System and method for controlling vehicle velocity and inter-vehicle distance

[0002] This application claims the benefit of and priority to PCT Application PCT/JP01/03972 filed May 14, 2001, Japanese Application 2000-143581 filed May 16, 2000; Japanese Application 2000-148742 filed May 19, 2000; Japanese Application 2000-148762 filed May 19, 2000; and Japanese Application 2000-148847 filed May 19, 2000.

[0003] The present invention relates to an automatic vehicular velocity controlling system and method having such an inter-vehicle distance controlling function as controlling a vehicular velocity to maintain an appropriate inter-vehicle distance.

[0004] An European Patent Application First Publication No. EP 0 982 172 A2 published on Mar. 1, 2000 exemplifies a previously proposed automatic vehicular velocity controlling system. (A U.S. Pat. No. 5,959,572 issued on Sep. 28, 1999 also exemplifies another previously proposed automatic vehicular velocity controlling system.)

[0005] In such a previously proposed automatic vehicular velocity controlling systems as described in the above-described European Patent Application First Publication, a response characteristic of a feedback inter-vehicle distance control system is determined in accordance with a relative velocity of an automotive vehicle in which the above-described vehicular velocity controlling system is mounted (hereinafter, also referred to as a host vehicle) to a preceding vehicle which is running ahead of the vehicle. Such an inter-vehicle distance control system as described above has aimed at providing a sharp response to a variation in the relative velocity by enlarging feedback constants when the magnitude of the relative velocity is large.

[0006] In addition, in such a previously proposed automatic vehicular velocity controlling system as described above, a deviation from a dynamic characteristic of an object to be controlled is estimated (so-called, external disturbance estimation) and an added value of an external disturbance to a target vehicular velocity is determined to be a new target vehicular velocity so that the inter-vehicle distance control is carried out in accordance with the external disturbance value (a running resistance or so forth).

[0007] Furthermore, in the above-described previously proposed automatic vehicular velocity controlling system, if a time duration from a time at which the preceding vehicle is stopped to a time at which the host vehicle has reached to the preceding vehicle at the present vehicular velocity of the host vehicle is defined as an inter-vehicle time duration, a target inter-vehicle distance is defined as a product between a velocity of the preceding vehicle and the inter-vehicle time duration.

[0008] In this case, the setting of inter-vehicle time duration is carried out by a vehicular occupant's manipulation on a three-stage changeover switch which switches the inter-vehicle time duration into a remote (long) distance, a middle distance, and a close (short) distance.

[0009] There are various methods of deriving the relative velocity used in the inter-vehicle distance control. One of the methods include detecting the inter-vehicle distance and deriving the relative velocity from a variation of the detected inter-vehicle distance.

[0010] However, an error in a detection of the inter-vehicle distance causes an occurrence of an error in the relative velocity. As the inter-vehicle distance becomes longer, the error tends to be increasingly included in the relative velocity. Hence, even in an actual case where the distance from the preceding vehicle becomes remote and the relative velocity is small, it is recognized that the error in the detection of the inter-vehicle distance causes the relative velocity to be changed to tend to become large. At this time, the feedback response characteristic is caused to become quick and a sharp control is resulted. Therefore, an unmatched feeling to a driving sense may be given to a vehicular occupant of the host vehicle.

[0011] It is noted that even in a case of a direct detection of the relative velocity, the error of calculation to detect the relative velocity occurs and, therefore, the unmatched feeling to the driving sense may be given to the occupant in the similar way as described above.

[0012] In addition, in a case where another vehicle which is running at a slightly slower velocity than the host vehicle is interrupted in front of the host vehicle, the relative velocity is small so that a feedback response is slow.

[0013] Hence, a control response is delayed and the host vehicle becomes remote from the preceding vehicle with a large delay to maintain a predetermined inter-vehicle distance form the interrupted other vehicle. Consequently, the unmatched feeling to the driving sense is given by the vehicular occupant.

[0014] Next, since the external estimation is the detection of the deviation of the external disturbance from the normal dynamic characteristic of the object to be controlled (i.e., the host vehicle), the deviation of the external disturbance from the normal dynamic characteristic of the object to be controlled (the normal dynamic characteristic is found during a flat road run of the vehicle) continuously occurs during a ascending slope run of the vehicle. Hence, the deviation is continuously added to the target vehicular velocity so that the vehicle (host vehicle) is continuously controlled to be accelerated. In this case, when the host vehicle has caught up with the preceding vehicle and the inter-vehicle distance has reached to a set value of inter-vehicle distance, the target vehicular velocity of the host vehicle is changed to the target vehicular velocity of the host vehicle is changed to a deceleration direction. However, since the above-described deviation is continued to be added, a start of deceleration is delayed so that the host vehicle becomes approached excessively to the preceding vehicle. This phenomenon also gives the vehicular occupant the unmatched feeling to the driving sense.

[0015] Next, suppose that the vehicular driver tries to widen the inter-vehicle distance and modifies the inter-vehicle time duration to a large value (for example, from the close distance to the remote distance) in a case where the host vehicle is running to follow the preceding vehicle at a small relative velocity to the preceding vehicle, for example, like such a case that the host vehicle is running at a substantially constant inter-vehicle distance to the preceding vehicle. In this modification of the inter-vehicle time duration, the relative velocity becomes gradually large but the response of the vehicular velocity variation to this modification is not favorable since the response characteristic is set slowly due to the small relative velocity. The above-described situation is found in such a case that the driver feels for the host vehicle to approach excessively to the preceding vehicle and manipulates the three-stage changeover switch to try to widen the inter-vehicle distance. Hence, if the vehicular velocity variation is dull upon the driver's manipulation on the three-stage changeover switch, the driver does not feel that the three-stage changeover switch has been operated to widen the inter-vehicle distance but gives the unmatched feeling to the driving sense.

[0016] With the above-described problems in mind, it is an object of the present invention to provide automatic vehicular velocity controlling system and method for an automotive vehicle which are capable of performing an appropriate inter-vehicle distance control with an appropriate response characteristic which accords with a vehicular running situation and without giving a driver's unmatched feeling to the driving sense.

[0017] According to one aspect of the present invention, there is provided an automatic vehicular velocity controlling system for an automotive vehicle, comprising: an inter-vehicle distance detector to detect an inter-vehicle distance from the vehicle to a preceding vehicle which is running ahead of the vehicle; a vehicular velocity detector to detect a vehicular velocity of the vehicle; a relative velocity detector to detect a relative velocity of the vehicle to the preceding vehicle; an inter-vehicle distance command value calculating section that calculates a command value of the inter-vehicle distance; a control response characteristic determining section that determines a response characteristic of an inter-vehicle related feedback control system in accordance with the relative velocity; a response characteristic correcting section that corrects the response characteristic of the inter-vehicle related feedback control system determined by the control response characteristic determining section in accordance with at least the inter-vehicle distance detected by the inter-vehicle distance detector; a vehicular velocity command value calculating section that calculates an inter-vehicle related controlling vehicular velocity command value on the basis of the response characteristic of the inter-vehicle related feedback control system corrected by the response characteristic correcting section; and a vehicular velocity controlling section that controls at least one of a driving force of the vehicle, a braking force of the vehicle, and a speed ratio of a transmission in such a manner that the vehicular velocity detected by the vehicular velocity detector is made substantially equal to the vehicular velocity command value.

[0018] According to another aspect of the present invention, there is provided an automatic vehicular velocity controlling method for an automotive vehicle, comprising: detecting an inter-vehicle distance from the vehicle to a preceding vehicle which is running ahead of the vehicle; detecting a vehicular velocity of the vehicle; detecting a relative velocity of the vehicle to the preceding vehicle; calculating a command value of the inter-vehicle distance; determining a response characteristic of an inter-vehicle related feedback control system in accordance with the relative velocity; correcting the determined response characteristic of the inter-vehicle related feedback control system in accordance with at least the detected inter-vehicle distance; calculating an inter-vehicle related controlling vehicular velocity command value on the basis of the corrected response characteristic of the inter-vehicle related feedback control system; and controlling at least one of a driving force of the vehicle, a braking force of the vehicle, and a speed ratio of a transmission in such a manner that the detected vehicular velocity is made substantially equal to the vehicular velocity command value.

[0019] This disclosure of the invention does not necessarily describe all necessary features so that the invention may also be a sub-combination of these described features.

[0020] FIG. 1A is a functional block diagram of a whole automatic vehicular velocity controlling system in a preferred embodiment according to the present invention.

[0021] FIG. 1B is a schematic top view of an example of an inter-vehicle time duration setting block shown in FIG. 1A.

[0022] FIG. 2 is a functional block diagram of an inter-vehicle distance command value calculating block shown in FIG. 1A.

[0023] FIG. 3 is a characteristic graph representing a step response of a transfer function of a set inter-vehicle time duration phase advance compensating block shown in FIG. 1A.

[0024] FIG. 4 is a functional block diagram of a vehicular velocity control system feedback characteristic determining block shown in FIG. 1A.

[0025] FIGS. 5A and 5B are characteristic graphs used to determine respective coefficients in the vehicular velocity control system feedback characteristic determining block shown in FIGS. 1 and 4, particularly FIG. 5A representing a map to determine an inter-vehicle distance control feedback system damping factor ζ_nDBand FIG. 5B representing a map to determine an inter-vehicle distance control feedback system specific angular frequency ω_nDB, respectively.

[0026] FIG. 6 is a characteristic graph representing a map to determine a correction coefficient C_D1.

[0027] FIG. 7 is a characteristic graph representing a map to determine a road surface gradient φ_A(t).

[0028] FIG. 8 is a characteristic graph representing a map to determine a correction coefficient C_D3.

[0029] FIG. 9 is a characteristic graph representing a map to determine a correction coefficient CD4.

[0030] FIG. 10 is a functional block diagram of a whole vehicular velocity controlling block shown in FIG. 1A.

[0031] FIG. 11 is a functional block diagram of a lateral G dependent vehicular velocity correction quantity calculating block shown in FIG. 10.

[0032] FIG. 12 is a characteristic graph representing a relationship between a host vehicular velocity V_A(t) and a cut-off frequency fc of a low-pass filter.

[0033] FIG. 13 is a characteristic graph representing a relationship between a correction coefficient to calculate a vehicular velocity correction quantity V_SUB(t) and a value of a lateral G, Y_G(t).

[0034] FIG. 14 is a characteristic graph representing a relationship between the specific angular frequency and the host vehicular velocity V_A.

[0035] FIG. 15 a characteristic graph representing a relationship between an absolute value of a deviation between the host vehicular velocity V_A(t) and a vehicular velocity command maximum value V_SMAXand a vehicular velocity command value variation rate Δ_VCOM(t).

[0036] FIG. 16 is a functional block diagram of one example of a drive torque command value calculating block shown in FIG. 10.

[0037] FIG. 17 is a characteristic graph representing one example of an engine non-linear steady state characteristic.

[0038] FIG. 18 is a characteristic graph representing one example of a throttle opening estimation map.

[0039] FIG. 19 is a characteristic graph representing one example of a CVT (Continuously Variable Transmission) speed ratio (also called, gear ratio or a transmission ratio).

[0040] FIG. 20 is a characteristic graph representing one example of a engine whole performance.

[0041] FIG. 21 is a functional block diagram of another example of the drive torque command value calculating block shown in FIG. 10 which is alternative of that shown in FIG. 16.

[0042] Reference will hereinafter be made to the drawings in order to facilitate a better understanding of the present invention.

[0043] FIG. 1A shows a whole functional block diagram of an automotive vehicular velocity controlling system in a preferred embodiment according to the present invention applicable to an automotive vehicle.

[0044] An inter-vehicle distance controlling block 105 (a portion enclosed by a broken line) is constituted by a microcomputer and its peripheral circuitry. Each block present inside of inter-vehicle distance controlling block 105 shown in FIG. 1A is a block representation of respective contents of calculations implemented by the microcomputer.

[0045] The inter-vehicle distance controlling block 105 receives an inter-vehicle time duration signal d_T(t), a relative velocity signal ΔV(t), and a vehicular velocity signal V_A(t) of the vehicle (host vehicle), calculates and supplies an inter-vehicle controlling vehicular velocity command value V*(t) to a vehicular velocity controlling block 500. A detailed functional explanation of the vehicular velocity controlling block 500 will be described later with a chief reference to FIG. 10.

[0046] It is noted that a symbol (t) denotes a value which varies with time t and is not always attached to another symbol in the drawings.

[0047] Vehicular velocity sensor 10 detects the vehicular velocity of the host vehicle from a revolution speed of one or any two of tire wheels of the vehicle.

[0048] Inter-vehicle distance sensor 15 which, for example, utilizes a laser radar detects an inter-vehicle distance L_A( t) to a preceding vehicle running ahead of the host vehicle according to reflected waves of light beams or electromagnetic waves and a relative velocity ΔV(t) from a change in time of the detected inter-vehicle spatial distance and receives the vehicular velocity signal V_A(t) from the vehicular velocity sensor 10, and outputs a (active) flag F (hereinafter referred to as a preceding vehicle flag) upon a decision that an object present in front of the host vehicle is the preceding vehicle in a case where a difference between the relative velocity ΔV(t) and the host vehiculart velocity VA(t) falls out of a range expressed as, for example, 5%×VA(t) Km/h.

[0049] Inter-vehicle time duration setting block 150 sets the inter-vehicle time duration dT(t) in response to a manipulation by the vehicular driver.

[0050] It is noted that the inter-vehicle time duration dT(t) is a time duration from a time at which the preceding vehicle is stopped to a time for the host vehicle to reach to the stopped preceding vehicle at the present vehicular velocity, if the preceding vehicle which follows the host vehicle is stopped.

[0051] Inter-vehicle time duration setting block 150 includes a switch such as shown in FIG. 1B to changeover three stages of, for example, remote (long) distance, middle distance, and close (short) distance according to the vehicular driver's operation thereof so that three-kinds of inter-vehicle time durations are selected. For example, the remote distance indicates 2.2 seconds, the middle distance indicates 1.8 seconds, and the close distance indicates 1.4 seconds. 1.8 seconds in the case of the middle distance corresponds to about 50 meters of the inter-vehicle distance when the host vehicle is running at 100 Km/h.

[0052] An inter-vehicle distance command value calculating block 110 constitutes a part of the inter-vehicle distance controlling block 105 shown in FIG. 1A and includes, as shown in FIG. 2, a set inter-vehicle time duration phase advance compensating block 111 and inter-vehicle distance command value determining section 112.

[0053] Set inter-vehicle time duration phase advance compensating block 111 receives inter-vehicle time duration d_T(t) from inter-vehicle time duration setting block 150 and outputs an inter-vehicle time duration phase advance compensation value d_THPF(t) in a case where the present inter-vehicle time duration d_T(t) is different from a previous inter-vehicle time duration d_T(t−1), viz., when the inter-vehicle distance controlling block 105 determines that the vehicular driver has tried to modify the setting of the inter-vehicle time duration through inter-vehicle time duration setting block 150.

[0054] The following equation represents a transfer function of inter-vehicle time duration phase advance compensating block:

[0055] D_THPF(t)=d_T(t)·(T₁·s+1)/(T₂·s+1)

[0056] In this equation, T₁and T₂denote time constants and T₁>T₂and s denotes a differential operator.

[0057] This time constant condition described above can cause the phase of the inter-vehicle time duration d_T(t) to be advanced with respect to the previous thereof.

[0058] FIG. 3 shows a step response of the transfer function expressed above in the inter-vehicle time duration phase advance compensating block 111.

[0059] As shown in FIG. 3, a phase advance compensation can be performed to the inter-vehicle time duration d_T(t) such as shown in the transfer function of set inter-vehicle time duration phase advance compensating block 111.

[0060] In details, in the case where the vehicular driver modifies the setting of the inter-vehicle time duration, for example, as shown by an example of FIG. 3, in a case where the setting is modified in such a manner as to modify the inter-vehicle time duration from d_TMwhich corresponds to the middle distance to d_TLwhich corresponds to the remote distance or d_TSwhich corresponds to the close distance, a change quantity of the inter-vehicle time duration is temporarily enlarged rather than a target new inter-vehicle time duration of d_TLor d_TS(if the inter-vehicle time duration is largely modified, the value is more larger and if this time duration is small modified, the value is more smaller) and, thereafter, the time duration is converged to new target inter-vehicle time duration d_TLor d_TS.

[0061] Consequently, in the case where the vehicular driver modifies the setting of the inter-vehicle time duration, the set inter-vehicle time duration can be controlled to speedily responded to an intention of the vehicular driver.

[0062] In addition, referring to FIG. 2, inter-vehicle distance command value determining block 112 calculates an inter-vehicle distance command value L*(t) from vehicular velocity V_A(t), relative velocity ΔV(t), and inter-vehicle time duration phase advance compensation value d_THPF(t) set arbitrarily by the vehicular driver in accordance with the following equation.

[0063] L*(t)={V_A(t)+ΔV(t)}·d_THPF(t)

[0064] As defined in this equation, the inter-vehicle distance command value L*(t) is given by an addition of the vehicular velocity V_A(t) to relative velocity ΔV(t) which is multiplied by the time duration phase advance compensation value d_THPF(t).

[0065] Hence, in a case where the inter-vehicle time duration is modified as described above, the change quantity of the time duration is temporarily enlarged and, thereafter, the inter-vehicle time duration is converged to the new target inter-vehicle time duration (by a minus exponential curve(e^−xcurve)). Therefore, the inter-vehicle distance is quickly changed upon the operation of the setting switch of inter-vehicle time duration setting block 150.

[0066] After the modification of the setting of the inter-vehicle time duration to enlarge the inter-vehicle time duration since, for example, the vehicular driver decides that the vehicle is to be approached to the preceding vehicle, the inter-vehicle distance is immediately changed so that there is no possibility that the vehicular driver gives the insufficient feeling due to an moderate change in the inter-vehicle distance.

[0067] Next, referring back to FIG. 1A, a target inter-vehicle distance calculating block 120 receives the preceding vehicle flag F, relative velocity ΔV(t), and inter-vehicle distance L_A(t) from the inter-vehicle distance sensor 15, sets relative velocity ΔV(F) and inter-vehicle distance L_A(F) at a time point at which the preceding vehicle is recognized as initial values of a target relative velocity ΔV_T(t) and a target inter-value distance L_T(t), and calculates target inter-value distance L_T(t) and target relative velocity ΔV_T(t) in a case where the input is the inter-vehicle distance command value L*(t) using a filter represented by a matrix equation shown in Table 1.

[0068] In the equation shown in Table 1, ω_nTdenotes a specific angular frequency of a target inter-vehicle distance response and a value thereof arbitrarily set by a designer, ζ_Tdenotes a damping factor of the target inter-vehicle distance response and a value thereof arbitrarily set by the designer, and Lv denotes a dead time due to a lag in a vehicular power train system as will be described later.

[0069] For the equation shown in Table 1, a transfer function in a case where the inter-vehicle distance command value L*(t) is the input and the target inter-vehicle distance L(t) is the output is expressed in the following equation:

[0070] L_T(t)=ω_nT²·e^−LV·s·L*(t)/(s²+2ζ_T·s+ω_nT²)

[0071] Pre-compensation vehicular velocity command value calculating block 130 is constituted by an inverse of a transfer function having a product between a transfer function Gv(s)′ (Gv(s)′=1/(Tv·s+1)) which neglects a dead time of vehicular velocity controlling block 500 and an integrator and pre-compensation vehicular velocity command value calculating block 130 calculates pre-compensation vehicular velocity command value V_C(t) shown in the following equation:

[0072] Vc(t)=ω_nT²·s(Tv·s+1)·L*(t)/(s²+2ζT·s+ω_nT²)

[0073] In the above equation, Tv denotes a time constant used in the transfer function of vehicular velocity controlling block 500.

[0074] In addition, when pre-compensation vehicular velocity command value Vc(t) is calculated from a state space representation using the equation shown in Table 1, an equation shown in Table 2 is given.

[0075] An inter-vehicle related vehicular velocity command value calculating block 140 calculates an inter-vehicle related controlling vehicular velocity command value V*(t) using the following equation according to actual inter-vehicle distance L_A(t), actual vehicular velocity V_A(t), actual relative velocity ΔV(t), target inter-vehicle distance L_T(t), and target relative velocity ΔV_T(t), and feedback constants f_Land fv as will be described later.

[0076] V*(t)=V_A(t)+ΔV(t)−Vc(t)−{L_T(t)−L_A(t)}·f_L−{ΔV_T(t)−ΔV(t)}·f_v

[0077] Inter-vehicle related feedback characteristic determining block 300 receives inter-vehicle distance L_A(t), relative velocity ΔV(t), and set inter-vehicle time duration d_T(t) and calculates feedback constants f_Land f_V.

[0078] Hereinafter, a determining method of feedback constants f_Land f_Vwith reference to FIG. 4 will be described below.

[0079] FIG. 4 shows a functional block diagram of inter-vehicle related control feedback characteristic determining block 300.

[0080] Inter-vehicle related control feedback characteristic determining block 300 includes: a feedback system damping factor determining block 310; a feedback system damping factor correcting block 311; a feedback system specific angular frequency determining block 320; a feedback system specific angular frequency first correcting block 330; a feedback system specific angular frequency second correcting block 331; and feedback constants determining block 340, each for inter-vehicle distance controlling block 105.

[0081] The block diagram of this inter-vehicle related feedback characteristic determining block 300 is expressed in terms of a transfer function G_DSfrom target inter-vehicle distance L_T(t) to actual inter-vehicle distance L_A(t) as follows:

[0082] G_DB(S)={ω_nDB²(T_VB·s+1)}/(s²+2ζ_nDB·ω_nDB+ω_nDB²)

[0083] wherein ζ_nDB=(f_V+1)/2{square root over ( )}(f_L/T_V) (denotes a damping factor in the inter-vehicle related feedback system), ω_nDB={square root over ( )}(f_L/T_V) and T_VB=f_L/T_V; a value corresponding to a zero (point) in the inter-vehicle related control feedback, and Tv denotes the time constant on a vehicular velocity feedback control in the vehicular velocity controlling block 500.

[0084] A feedback system damping factor determining block 310 inputs relative velocity ΔV(t) and determines an inter-vehicle related control feedback system damping factor ζ_nDBfrom a map shown in FIG. 5A in accordance with relative velocity ΔV(t).

[0085] As shown in FIG. 5A, ζ_nDBis constant even if relative velocity ΔV(t) is varied. This is because it is most appropriate that the value of ζ_nDBis set to 1 (when ζ_nDB=1, a critical damping occurs) in order to prevent a fluctuation tendency and to improve a response characteristic.

[0086] Feedback system damping factor determining block 320 inputs relative velocity ΔV(t) and determines the specific angular frequency ζ_nDBin the inter-vehicle related feedback control system according to a map shown in FIG. 5B with respect to relative velocity ΔV(t).

[0087] As shown in FIG. 5B, the specific angular frequency ζ_nDBis made smaller (lowered) if an absolute value of relative velocity ΔV(t)is small so that a slow control is carried out. If ζ_nDBis made larger (higher), a quick control is carried out in order not to develop a lag (delay) to a motion of the preceding vehicle.

[0088] Feedback system specific angular frequency first correcting block 330 determines a correction coefficient CD1 to correct the specific angular frequency ζ_nDBof the inter-vehicle related control feedback system, corrects specific angular frequency ζ_nDBby C_D1, and outputs a post-correction specific angular frequency ω_nBC1.

[0089] That is to say, ω_nDBC1can be expressed as follows: ω_nDBC1=C_D1·ω_nDB.

[0090] Correction coefficient C_D1, as appreciated from FIG. 6, is present in such a manner that the correction coefficient C_D1is a value equal to or larger than 1 if the inter-vehicle distance is shorter than a first predetermined value (for example, about 20 meters) so that the specific angular frequency ω_nDBis enlarged (made higher) to quicken the response characteristic of the inter-vehicle distance control and CD1 is a value less than one if the inter-vehicle is longer than a second predetermined value (for example, about 90 meters) so that the specific angular frequency ω_nDBis made smaller (lowered) to slow the response characteristic of the inter-vehicle distance control.

[0091] As described above, the value of specific angular frequency ω_nDBis corrected in accordance with the detected inter-vehicle distance. In details, when the inter-vehicle distance is large (long), the specific angular frequency ω_nDBis made small (lowered). when the inter-vehicle distance is small (short), the specific angular frequency ω_nDBis made large (high). Thus, if the inter-vehicle distance is long, the response characteristic is made dull (slowed) so that even when the relative velocity is largely varied due to a calculation error of detecting the relative velocity, the correction of the response characteristic can prevent a steep or sharp control from occurring. Consequently, the vehicular driver does not give the un-matched feeling to the driving sense.

[0092] Feedback system specific angular frequency second correcting block 331 receives the inter-vehicle time duration d_T(t) and the post-correction specific angular frequency ω_nDBC1of the inter-vehicle related control feedback system and temporarily enlarges ω_nDBC1by changing another correction coefficient CD2 from a preset value of 1 to 1.5 for one second in a case where d_T(t−1) to d_T(t), namely, where the driver has modified, the setting of the inter-vehicle time duration. In a normal state, the value of the other correction coefficient C_D2is 1. However, only if the setting of the inter-vehicle time duration is modified, C_D2is changed from 1 to 1.5 (or larger) so that ω_nDBC1is temporarily enlarged to achieve a quick modification of the actual inter-vehicle time duration.

[0093] It is noted that ω_nDBCafter this correction is as follows: ω_nDBC=C_D2·ω_nDBC1.

[0094] The above-described feedback system specific angular frequency second correcting block 331 improves the response characteristic by enlarging the specific angular frequency (gain) of the inter-vehicle distance feedback system. In the above-described feedback system specific angular frequency second correcting block 331, however, the vehicle would respond oversensitively if a quick motion of the preceding vehicle is made so that a vehicular comfortability would become worsened to some degree.

[0095] In this respect, as described in the inter-vehicle distance command value determining block 112, the value of the inter-vehicle time duration is temporarily enlarged or made smaller than the new inter-vehicle time duration d_TLor d_TSand, thereafter, converged the new one d_TLor d_TSwhen the inter-vehicle time duration is modified. Consequently, the above-described oversensitive response cannot be found.

[0096] Feedback system damping factor correcting block 311 receives an external disturbance value d_V(t) calculated by drive torque command value calculating block 530 of vehicular velocity controlling block 500, receives the specific angular frequency ζ_nDBfrom the feedback system damping factor determining block, and estimates a road surface gradient φ_A(t) from feedback system damping factor determining block 310.

[0097] Specifically, as shown in FIG. 7, if the external disturbance value d_V(t) is negative, the gradient indicates an ascending slope. If it is positive, the gradient indicates a descending slope. Thus, road surface gradient φ_A(t) is derived.

[0098] Thus, the correction coefficient C_D3is derived on the basis of the map shown in FIG. 8. The damping factor ζ_nDBCis determined by correcting the damping factor ζ_nDBof the inter-vehicle related control feedback system.

[0099] That is to say, the damping factor ζ_nDBCis expressed as follows:

[0100] That is to say, ζ_DBC=ζ_nDB·C_D3.

[0101] The value of correction coefficient C_D3, as appreciated from FIG. 8, is set to 1, when the road surface gradient φ_A(t) falls in a predetermined range. As the absolute value of the road gradient φ_A(t) becomes larger, the value of C_D3is set to becomes larger than 1.

[0102] Feedback constants determining block 340 receives damping factor ζ_nDBCof inter-vehicle distance control feedback system and calculates feedback constants f_Land f_Vfrom the equation expressed below.

[0103] f_L=ω_nDBC²·Tv;

[0104] and

[0105] f_V=2·ζ_nDBC·ω_nDBC·Tv−1.

[0106] Consequently, as the inter-vehicle distance L_A(t) becomes short, the correction coefficient CD2 becomes small but the specific angular frequency ω_nDBbecomes large.

[0107] Then, both of feedback constants f_Land f_Vare enlarged so that the deceleration becomes faster. At this time, if the inter-vehicle distance L_A(t) is short, specific angular frequency ω_nDBCis enlarged and, in place of the enlargement for the feedback constants f_Land f_V, the inter-vehicle related constant f_Lmay directly be corrected in accordance with the inter-vehicle distance.

[0108] In addition, as the gradient of road becomes enlarged, correction coefficient C_D3becomes large, the damping factor ζ_nDBCis accordingly enlarged, and feedback constant f_Vis enlarged so that the deceleration is made faster.

[0109] It is noted that as the gradient of road becomes large, the correction coefficient C_D3becomes large and damping factor ζ_nDBCbecomes large so that the deceleration becomes faster.

[0110] It is also noted that inter-vehicle distance command value L*(t) of the inter-vehicle distance command value calculating block 110 may be enlarged in place of the modification of specific angular frequency ζ_nDBof the inter-vehicle related feedback system. Specifically, inter-vehicle distance command value calculating block 110 may receive the external disturbance value d_V(t) calculated by drive torque command value calculating block 530 of vehicular velocity controlling block 500, may estimate road surface gradient φ_A(t) according to the map shown in FIG. 7 from external disturbance value, determines correction coefficient C_D4(>1) from a map shown in FIG. 9, and may calculate inter-vehicle distance command value L*(t) at the inter-vehicle distance command value determining section 112 shown in FIG. 2 by the correction coefficient C_D4.

[0111] That is to say, L*(t)=[V_A(t)+ΔV(t)]·d_T(t)·C_D4.

[0112] In this case, since as the external disturbance value d_V(t) representing the gradient angle of the road surface becomes larger, C_D4becomes larger and inter-vehicle distance command value L*(t) becomes large. Hence, a start of deceleration becomes earlier.

[0113] As described above, the values of each or either of the correction coefficients C_D3or C_D4are set in accordance with the external disturbance value of d_V(t) representing the road surface gradient angle. Hence, in a case where the control toward the acceleration direction is carried out according to the deviation of the external disturbance from the normal dynamic characteristic of the object to be controlled (host vehicle) during the vehicular run on the ascending slope, the feedback response characteristic is increased or the target inter-vehicle distance is set to be larger. Thus, if the host vehicle has reached to the set inter-vehicle distance, a speedily deceleration or the earlier deceleration start occurs. Consequently, even if the host vehicle is running on the ascending slope, the start of deceleration cannot be delayed. In addition, since the road surface gradient angle is calculated from the external disturbance value d_V(t), no additional sensor to detect the road surface gradient is needed.

[0114] Next, the vehicular velocity controlling block 500 will be described with a chief reference of FIG. 10.

[0115] First, suppose that a system switch (not shown) is turned on. At this time, a power supply of the whole system is turned on so that the system is a stand-by state. In this state, if a set switch 20 is turned to ON, the control is initiated.

[0116] The vehicular velocity controlling block 500 (a portion enclosed by a broken line) is constituted by a microcomputer in the preferred embodiment in the same manner as the inter-vehicle distance controlling block 105 shown in FIG. 1A. However, a single chip microcomputer with the inter-vehicle distance controlling block 105 may be installed together.

[0117] Within the vehicular velocity controlling block 500, vehicular velocity command value determining block 510 calculates vehicular velocity command value VCOM(t) for each control period of 10 milliseconds.

[0118] Vehicular velocity command maximum setting block 520 sets the vehicular velocity V_A(t) when set switch 20 is pressed (operated) as a vehicular velocity command maximum value V_SMAX(target vehicular velocity).

[0119] After the vehicular velocity command maximum value V_SMAXis set through set switch 20, whenever a coast switch 30 is pressed once, a vehicular velocity command maximum value setting section 520 sets the vehicular velocity command maximum value V_SMAXto a lower value in a stepwise manner by a unit of 5 Km/h.

[0120] That is to say, if coast switch 30 is pressed by an n number of times, the value of V_SMAXis set to a lower value by, e.g., n×5 (Km/h) (if continued to be pressed, suppose that a time duration during which the switch is continued to be pressed is T and, for example, this case is T/1(second)×5 Km/h.

[0121] After vehicular velocity command maximum value V_SMAXto a higher value in the stepwise manner by a unit of 5 Km/h.

[0122] Whenever n number of times accelerate switch 40 is pressed once after the set of V_SMAXby set switch 20, vehicular velocity command maximum value setting block 520 sets V_SMAXin the stepwise manner by the unit of 5 Km/h to a high value.

[0123] That is to say, whenever n number of times switch 40 is pressed, V_SMAXis set to the higher value by n×5 Km/h (or, for example, T/1 (second)×5 Km/h if pressed continuously).

[0124] Next, lateral G dependent vehicular velocity correction quantity calculating block 580 receives steering angle θ(t) of a steering wheel outputted from a steering angle sensor 100 and vehicular velocity V_A(t) and calculates a vehicular velocity correction quantity V_SUB(t) to correct a vehicular velocity command value as will be described later in accordance with a lateral acceleration (hereinafter, referred to as a lateral G).

[0125] The lateral G dependent vehicular velocity correction quantity calculating block 580, specifically as shown in FIG. 11, includes a steering angle signal low-pass filter LPF (hereinafter, called steering angle signal LPF block), a lateral G calculating block 582, and a vehicular velocity correction quantity calculating block 583.

[0126] First, steering angle signal LPF block 581 calculates a steering angle LPF value θLPF(t) according to inputted vehicular velocity V_A(t) and steering angle θ(t).

[0127] θLPF(t)=θ(t)/(TSTR·s+1)

[0128] In the equation described above, TSTR denotes a time constant of LPF (=1/2π·fc) and a cut-off frequency fc of LPF is determined from a map shown in FIG. 12 representing a relationship between the cut-off frequency and vehicular velocity V_A(t). This map of FIG. 12 indicates that as the vehicular velocity V_A(t) becomes higher (falls in a high velocity range), the cut-off frequency fc becomes lower. For example, fc at 50 Km/h is higher than that at 100 Km/h.

[0129] The lateral G calculating block 582 receives steering angle LPF value θLPF(t) and vehicular velocity V_A(t) and calculates lateral G value Y_G(t) in accordance with the following equation;

[0130] Y_G(t)={V_A(t)²·θLPF(t)}/{N·W·[1+A·V_A(t)]}

[0131] In this respect, W denotes a wheel base of the vehicle, N denotes a steering gear ratio, and A denotes a stability factor.

[0132] It is noted that the above-described equation indicates a case where the lateral G is detected from the vehicular steering angle but the lateral G may be detected through the low pass filtering to yaw rate ψ(t) using a yaw rate (yaw velocity) sensor. In the latter case, use is made of the following equation.

[0133] Y_G(t)=V_A(t)·ψLPF

[0134] ψLPF=ψ(t)/(T_YAW·s+1)

[0135] in this equation, T_YAWdenotes a time constant of the low pass filter for the yaw rate and as the vehicular velocity VA(t) becomes larger (higher), T_YAWis varied to be a larger value.

[0136] Vehicular velocity correction quantity V_SUB(t) is calculated by multiplying a correction coefficient determined by a lateral G with a predetermined vehicular velocity command value variation rate limitation value [for example, 0.021 (Km/h)/10(millisecond)=0.06 G].

[0137] It is noted that a value of the above-described vehicular velocity command value variation rate limitation value is equal to a maximum value of a vehicular velocity command value variation rate ΔV_COM(t) shown in FIG. 15 described later.

[0138] V_SUB(t)=correction coefficient×0.021 (Km/h)/10 (milliseconds).

[0139] As will be described later, in order to calculate a vehicular velocity command value V_COM(t) which is a value to control the vehicular velocity finally, vehicular velocity correction quantity V_SUB(t) is to be included as a subtraction term. Hence, as vehicular velocity correction quantity V_SUB(t) becomes larger, the vehicular velocity command value V_COM(t) is more limited.

[0140] The correction coefficient becomes large as the lateral G value Y_G(t) becomes larger as shown in FIG. 13.

[0141] This is because as the lateral G becomes large, there is a large limitation on a variation in the vehicular velocity command value V_COM(t).

[0142] However, as shown in FIG. 13, in a case where the lateral G is equal to or lower than 0.1 G as shown in FIG. 13, the correction coefficient is zeroed determining that no correction is needed for the vehicular velocity command value. In addition, if the lateral G is equal to or larger than 0.3 G, the correction coefficient is set to a constant value (for example, 2) in order to prevent an excessive correction in a case where the lateral G detected value is erroneously large and since in a normal driving no such a large lateral G as described above is developed.

[0143] As will be described later in vehicular velocity command value determining block 510, the above-described accelerate switch operation causes the target vehicular velocity to be raised, viz., the acceleration is requested, vehicular velocity command value variation rate ΔV_COM(t) is added to the present vehicular velocity value V_A(t) from which the vehicular velocity correction value V_SUB(t) is subtracted to calculate the final vehicular velocity command value V_COM(t). Hence, if the vehicular velocity command value variation rate ΔV_COM(t) is larger than the vehicular velocity correction value, the vehicle is accelerated but if ΔV_COM(t) is smaller than vehicular velocity correction value V_SUB(t), the vehicle is decelerated. As described above, the vehicular velocity correction value V_SUB(t) is derived by multiplying the correction coefficient as shown in FIG. 13 with the vehicular velocity command value variation rate limitation value (maximum value of the vehicular velocity command value variation rate). Hence, for example, if vehicular velocity command value variation rate limitation value=vehicular velocity command value variation rate, the present vehicular velocity is maintained since the acceleration value is balanced with the deceleration value when the correction coefficient is 1 (Y_G(t)=0.2 in the example of FIG. 13). In details, in this example, if the lateral G value Y_G(t) is smaller than 0.2, the host vehicle is accelerated but if, Y_G(T) is larger than 0.2, the host vehicle is decelerated. On the other hand, if the operation on the coast switch 30 causes the target vehicular velocity to be reduced, i.e., the deceleration is requested, the vehicular velocity command value ΔV_COM(t) and vehicular velocity correction value V_SUB(t) are subtracted from the present vehicular velocity V_A(t) to derive the vehicular velocity command value V_COM(t). Hence, in this case, the host vehicle is always decelerated. A magnitude of deceleration becomes larger as the vehicular velocity correction value V_SUB(t) becomes larger, that is to say, as the lateral G becomes larger. It is noted that the above-described value 0.021 (Km/h)/10 (milliseconds) for the vehicular velocity command value variation rate limitation value is a value presumed when the vehicle is used in a freeway.

[0144] As described above, the vehicular velocity correction value V_SUB(t) is determined according to a product between the correction coefficient according to the lateral G and the vehicular velocity variation rate limitation value and the vehicular velocity is controlled so that the lateral G is prevented from being large with the value of subtraction term (vehicular velocity correction value) enlarged when the lateral G becomes large. However, as explained in the steering angle signal LPF block 581 shown in FIG. 11, the cut-off frequency fc is lowered as the vehicular velocity falls in the higher velocity range. Hence, the time constant TSTR of the LPF becomes large and steering angle LPF value θ_LPF(t) is made smaller so that the lateral G estimated by the lateral G calculating block 582 is accordingly small. As a result of this, the vehicular velocity correction value V_SUB(t) derived via the vehicular velocity correction quantity calculating map 583 is small. Then, the correction from the steering angle to the vehicular velocity command value tends to become difficult to be incorporated (the correction in the direction of acceleration).

[0145] A detailed explanation in this respect will be made hereinbelow.

[0146] A characteristic of a specific angular frequency ωn_STRon a vehicle response to the steering angle will be described in the following equation:

[0147] ωn_STR=(2W/V_A){square root over ( )}[Kf·Kr·(1+A·V_A²)/m_V·I]

[0148] Wherein Kf and Kr denote front and rear tire wheel cornering power (for one wheel), W denotes a wheel base, mv denotes a vehicular weight, A denotes a stability factor, I denotes a vehicular yaw inertia moment.

[0149] The characteristic of the specific angular frequency ωn_STRis shown in FIG. 14.

[0150] As shown in FIG. 14, this characteristic is such that as the vehicular velocity is increased, specific angular frequency ωn_STRbecomes reduced so that the vehicular response characteristic to the steering wheel becomes worsened. As the vehicular velocity becomes decreased, the specific angular frequency ωn_STRbecomes higher and the vehicular response characteristic to the steering angle becomes favorable. In order words, as the vehicular velocity becomes higher, lateral G is not easy to be developed even when the vehicular steering wheel is operated. As the vehicular velocity becomes reduced, a slight steering operation causes the lateral G to be easy to be developed. Therefore, as shown in FIG. 12, the cut-off frequency fc is reduced as the vehicular velocity becomes increased so that the responsive characteristic is slowed not to be easy to apply the correction to the vehicular velocity command value.

[0151] Next, a vehicular velocity command value variation rate determining block 590 shown in FIG. 10 calculates a vehicular velocity command value variation rate ΔV_COM(t) from a map shown in FIG. 15 on the basis of an absolute value of a deviation between vehicular velocity V_A(t) and vehicular velocity maximum value V_SMAX.

[0152] The map shown in FIG. 15 represents that as the absolute value of the deviation (|V_A−V_AMAX| becomes larger as vehicular velocity command value variation rate ΔV_COM(t) becomes larger to such a degree that ΔV_COM(t) does not exceed an acceleration control limitation value α as will be described in a vehicular velocity control stop deciding block 610 so that the vehicle velocity is increased or decreased (accelerated or decelerated) as speedily as possible (this occurs in a range B shown in FIG. 15).

[0153] Then, as the absolute value of the deviation becomes small, the vehicular velocity command value variation rate ΔV_COM(t) is made small to a degree that the driver' feeling of acceleration is not affected thereby to prevent vehicular velocity command maximum value V_SMAXdoes not overshoot (in a range C in FIG. 15).

[0154] In a range A shown in FIG. 15, ΔV_COM(t) is set to be a constant value which does not exceed the acceleration limitation value α (for example, 0.06 G) and in the range C in FIG. 15, ΔV_COM(t) is set to another constant value (for example, 0.03 G).

[0155] Furthermore, the vehicular velocity command value variation rate determining block 590 monitors vehicular velocity correction value V_SUB(t) outputted from lateral G dependent vehicular velocity correction quantity calculating block 580. If a value of vehicular velocity correction value VsuB(t) is varied from zero to once that other than zero and is again returned to zero, vehicular velocity command value variation rate determining block 590 determines that the host vehicle has ended a run on a curved road and detects whether the vehicular velocity VA(t) becomes equal to vehicular velocity command maximum value V_SMAX.

[0156] If determining that the host vehicle has ended the run of the curved road, vehicular velocity command value variation rate determining block 590 determines vehicular velocity command value variation rate A V_COM(t) from vehicular velocity V_A(t) in place of determination of vehicular velocity command value variation rate ΔV_COM(t) using the map shown in FIG. 15 on the basis of the absolute value of deviation between vehicular velocity V_A(t) and vehicular velocity command maximum value V_SMAX. In this case, the characteristic of ΔV_COM(t) uses that representing generally the same tendency as that shown in FIG. 15.

[0157] That is to say, a new map in which a lateral axis of FIG. 15 is replaced from |V_A−V_SMAX| with vehicular velocity V_A(t) is used. The new map is set such that as vehicular velocity V_A(t) becomes small the vehicular velocity command value variation rate ΔV_COM(t) becomes small. This process using the new map is ended upon the determination that vehicular velocity V_A(t) becomes equal to vehicular velocity command maximum value V_SMAX.

[0158] It is noted that, in place of the above example in which the determination of vehicular velocity command value variation rate ΔV_COM(t) from actual vehicular velocity V_A(t) when the determination of the end of the curved road run, vehicular velocity command value variation rate determining block 590 may determine that the vehicular curved run is started when vehicular velocity correction value V_SUB(t) becomes the value other than zero, may previously store the instantaneous vehicular velocity V_A(start) in a memory location, and may determine vehicular velocity command value variation rate ΔV_COM(t) from a magnitude of a difference ΔV_A=V_A(start)−V_A(end) (namely, a vehicular velocity drop quantity due to the correction of the vehicular velocity command value). The used characteristic at this time is a characteristic representing an inverse tendency. That is to say, a new map is such that the lateral axis of FIG. 15 is replaced from |V_A−V_SMAX| to the vehicular velocity difference ΔV_Aand, as vehicular velocity difference ΔV_Abecomes large, vehicular velocity command value variation rate ΔV_COM(t) is set to a smaller value. This process is continued until vehicular velocity V_A(t) is made equal to vehicular velocity command maximum value V_SMAX.

[0159] Since the vehicular velocity command value is corrected so that the lateral G value does not indicate an excessive value during the vehicular run on the curved road, the vehicular velocity is generally reduced. Therefore, as described above, vehicular velocity command value variation rate determining block 590 is so structured that, after the curved road run has been ended and the vehicular velocity has dropped, vehicular velocity command value variation rate ΔV_COM(t) is modified in accordance with vehicular velocity V_A(t) at the end of the curved road run or vehicular velocity difference ΔV_Abetween those start of the curved road run and end thereof (namely, before and after the vehicular velocity drops due to the correction made for the vehicular velocity command value).

[0160] It is estimated that in a case where the vehicular velocity is low when the curved road run is ended or in a case where the vehicular velocity difference ΔV_Ais large, the vehicular velocity has dropped due to a small radius of curvature of the curved road on which the host vehicle has run (tight corner). Then, there is a great possibility that the above-described situation is resulted when the host vehicle is running on a continued curved road such as a letter-S curve (or hairpin curve). Therefore, vehicular velocity command value variation rate ΔV_COM(t) is made small, in the case where the vehicular velocity is low when the curved road run is ended or in the case where the vehicular velocity difference ΔV_Ais large, so that the acceleration during the active vehicular velocity control by means of the vehicular velocity command value is made small. Thus, the large acceleration for each turn of the curved road is not carried out when the host vehicle is running on the continued letter S shaped curved road.

[0161] Similarly, in a case where the vehicular velocity is high when the curved road run is ended or in a case where the vehicular velocity difference ΔV_Ais small, vehicular velocity command value variation rate determining block 590 determines that the curved road on which the host vehicle is running is a single curved road and enlarges vehicular velocity command value variation rate ΔV_COM(t). This causes an immediate acceleration upon the end of the single curved road run to be carried out so that there is no possibility that the acceleration becomes dull and the vehicular driver gives an insufficient feeling.

[0162] As described above, vehicular velocity command value variation rate determining block 590 is so structured that vehicular velocity command value variation rate ΔV_COM(t) is enlarged (made large) in the case where the deviation between actual vehicular velocity V_A(t) and post-modification set vehicular velocity (in the case described above, corresponds to vehicular velocity command maximum value V_SMAX) is large, as shown in the map of FIG. 15.

[0163] Therefore, in a case where the host vehicle is following the preceding vehicle on the basis of the output signal from inter-vehicle distance controlling block 105 shown in FIG. 1A and is running under a situation wherein the set vehicular velocity in the inter-vehicle distance control is different from the actual vehicular velocity, vehicular velocity command value variation rate ΔV_COM(t) is set in accordance with the deviation between the modified set vehicular velocity and the actual vehicular velocity.

[0164] In details, the vehicular velocity is varied according to a small vehicular velocity command value variation rate ΔV_COM(t) (acceleration or deceleration) in a case where the deviation between actual vehicular velocity VA(t) and post-modification (after modification) set vehicular velocity is small even if the set vehicular velocity before and after the modification is remarkably differed. Hence, this variation in the vehicular velocity can be matched with the driver's drive feeling in a case where the driver has demanded that the vehicular velocity is slightly reduced. In addition, the vehicular velocity is varied at the large vehicular velocity variation rate in a case where the deviation between the actual vehicular velocity and the set vehicular velocity is large. Hence, the vehicular velocity can speedily be converged to the new set vehicular velocity after the modification. For example, such a control that is in conformity to the driver's drive feeling such as to desire for the vehicular velocity to be largely decreased can be achieved.

[0165] Next, referring back to FIG. 10, a vehicular velocity command value determining block 510 calculates vehicular velocity command value V_COM(t) in the following way, receiving vehicular velocity V_A(t), vehicular velocity correction value V_SUB(t), vehicular velocity command value variation rate ΔV_COM(t), and vehicular velocity command maximum value V_SMAX.

[0166] (1) A case where vehicular velocity command maximum value V_SMAXis larger than vehicular velocity V_A(t), i.e., the acceleration request through the operation of accelerate switch 40 (or resume switch) occurs.

[0167] V_COM(t)=min[V_SMAX, V_A(t)+ΔV_COM(t)−V_SUB(t)]

[0168] In other words, vehicular velocity command value VCOM(t) is determined through a selection from one of V_SMAXand a value of V_A(t)+ΔV_COM(t)−V_SUB(t) which is smaller than the other.

[0169] (2) A case where V_SMAX=V_A(t), i.e., the vehicular velocity is maintained at a constant velocity.

[0170] V_COM(t)=V_SMAX−V_SUB(t)

[0171] In other words, vehicular velocity V_COM(t) is determined to be a subtraction result of vehicular velocity command maximum value V_SMAXfrom vehicular velocity correction value V_SUB(t).

[0172] (3) A case where vehicular velocity command maximum value V_SMAXis smaller than vehicular velocity V_A(t), i.e., a deceleration request through an operation of coast switch 30 occurs.

[0173] V_COM(t)=max [V_SMAX, V_A(t)−ΔV_COM(t)−V_SUB(t)]

[0174] In other words, vehicular velocity command value V_COM(t) is determined through a selection from one of V_SMAXand a value of V_A(t)−ΔV_COM(t)−V_SUB(t) which is larger than the other.

[0175] It is however, noted that vehicular velocity command value determining block 510 inputs inter-vehicle related controlling vehicular velocity command value V*(t) from inter-vehicle distance controlling block 105 and preceding flag F from inter-vehicle distance sensor 15 and performs a process as described below.

[0176] (4) A case where preceding flag F is received.

[0177] If preceding flag F (it is noted that this flag F indicates that the preceding vehicle is present) is received, vehicular velocity command value determining block 510 compares inter-vehicle related controlling vehicular velocity commandvalue L*(t) with vehicular velocity command maximum value V_SMAX, selects one of these two compared values which is smaller than the other as V_CO(t), and calculates V_COM(t) using the following equation.

[0178] That is to say, V_COM(t)=V_CO(t)−V_SUB(t).

[0179] As described above, vehicular velocity command value determining block 510 determines vehicular velocity command value V_COM(t) and controls the vehicular velocity in accordance with the determined vehicular velocity command value V_COM(t).

[0180] Next, a drive torque command value calculating block 530 receives vehicular velocity command value V_COM(t) and vehicular velocity V_A(t) and calculates a drive torque command value D_FC(t) in the following way.

[0181] FIG. 16 shows an example of an internal block of drive torque command value calculating block 530.

[0182] A transfer characteristic Gv(s) with vehicular velocity command value V_COM(t) being an input and vehicular velocity V_A(t) being an output and can be expressed in the following equation.

[0183] Gv(s)=1/(Tv·s+1)·e^(−Lv·s)

[0184] In the above equation, Tv denotes a first order lag time constant and Lv denotes the dead time due to the lag in the power train system of the vehicle.

[0185] In addition, a vehicle model of an object to be controlled is modeled with a drive torque command value dFC(t) as a manipulated variable and vehicular velocity VA(t) as a controlled variable so that a motion of the power train of the vehicle can be expressed in a simple linear model which is described below.

[0186] V_A(t)=1/(m_v·Rt·s)e^(−Lv·s)·d_FC(t)

[0187] In the above equation, Rt denotes a tire effective revolution radius and m_Vdenotes a vehicular mass.

[0188] Since the vehicular model with-drive torque command value d_FC(t) being input and vehicular velocity V_A(t) being output is in a form of 1/s, the vehicular model consequently has an integration characteristic.

[0189] It is noted that if the response characteristic of the object to be controlled in the case where the vehicular velocity V_A(t) is the output and the vehicular velocity VA(t) is the input is made coincident with the characteristic of transfer characteristic Gv(s) having elements of the predetermined first order lag Tv and dead time Lv, the following can be defined using C₁(s), C₂(s), and C₃(s). It is also noted that C₁(s)denotes an external disturbance estimator determined by means of an approximation zeroing technique, C₂(s) denotes another external disturbance estimator determined by means of the approximation zeroing technique, these estimators being compensators acted to suppress an influence caused by an external disturbance and a modeling error, and C₃(S) denotes a compensator determined by means of a model matching technique.

[0190] Compensator C₁(s)=e^{(−Lv·s)/(T}_H·s+1); and

[0191] Compensator C₂(s)=(m_V·Rt·s)/(T_H·s+1).

[0192] At this time, external disturbance estimation value d_V(t) is given as follows:

[0193] d_V(t)=C₂(s)·V_A(t)−C₁(s)·d_FC(t).

[0194] In addition, suppose that the dead time of the object to be controlled is neglected and a normal model Gv(s) is a first order low pass filter of the time constant Tv. Compensator C₃(S) is the constant described in the following:

[0195] Compensator C₃(s)=m_v·Rt/Tv

[0196] The drive torque command value d_FC(t) is calculated and determined in the following equation by means of the three compensators C₁(s), C₂(s), and C₃(s).

[0197] That is to say, d_FC(t)=C₃(s)·{V_COM(t)−V_A(t)}−{C₂(s)·V_A(t)−C₁(s)·d_FC(t)}.

[0198] The drive torque is controlled on the basis of drive torque command value d_FC(t).

[0199] In details, drive torque command value calculating block 530 calculates throttle opening command value which coincides an actual drive torque d_FA(t) with drive torque command value d_FC(t) using a map representing an engine non-linear steady characteristic previously measured as shown in FIG. 17 and distributes d_FC(t) to a vehicular transmission system or brake system to compensate for an insufficient negative engine drive torque if the engine negative torque is insufficient. In this way, the control for the throttle valve opening angle, a continuously variable transmission, and brake system permits a linearization of the engine non-linear steady characteristic.

[0200] It is noted that in a case where the continuously variable transmission 70 is provided with a fluid converter with a lock up mechanism equipped, a lock up state signal LUs is inputted from a controller of the continuously variable transmission 70 and the drive torque command value calculating block 530 determines if the converter is in the lock up state and, in the case of the lock up state determination, enlarges the time constant T_H(this time constant is described in each denominator of C₁(s), C₂(s), and C₃(s) in FIG. 16). Thus, the vehicular velocity control feedback correction quantity (a correction coefficient of a feedback loop to maintain a desired response characteristic) becomes small so that the response characteristic is made conformity to that characteristic of the object to be controlled which delays at the time of an un-lock up state as compared with the lock up state. Hence, a stability of the vehicular velocity control system can be assured at each of both times of the lock up state and of the un-lock up state.

[0201] Although FIG. 16 shows drive torque command value calculating block 530 including compensators C₁(s) and C₂(s) to compensate for the transfer characteristic of the object to be controlled and compensator C₃(s) to achieve the response characteristic that the designer defines, FIG. 21 shows an alternative of drive torque command value calculating block 530 including a pre-compensator C_F(s) to compensate for giving an arbitrary response characteristic defined by the designer, a normal model calculating block C_R(s) to calculate the arbitrary response characteristic defined by the designer, and a feedback compensator C₃(s)′ to compensate for a deviation from the response characteristic of normal model calculating block C_R(s).

[0202] In details, in FIG. 21, pre-compensator C_F(s) calculates a reference drive torque command value d_FC1(t)using a filter defined in the following equation in order to achieve a transfer function Gv(s) of actual vehicular velocity V_A(t) with respect to vehicular velocity command value V_COM(t).

[0203] That is to say, d_FC1(t)=m_V·R_T·V_COM(t)/(Tv·s+1).

[0204] The normal model calculating block C_R(s) calculates a target response V_T(t) from transfer function Gv(s) and vehicular velocity command value V_COM(t).

[0205] That is to say, V_T(t)=Gv(s)·V_COM(t).

[0206] Feedback compensator C₃(s)′ calculates a drive torque command value correction quantity d_V(t)′ so as to eliminate such a deviation as that between target response V_T(t) and actual vehicular velocity V_A(t) if this deviation occurs.

[0207] In details, drive torque command value correction quantity d_V(t)′ is expressed in the following equation.

[0208] That is to say, d_V(t)′=[(Kp·s+K_I)/s][V_T(t)−V_A(t)].

[0209] In the above equation, Kp denotes a proportional gain of feedback compensator C₃(s)′ and K_Idenotes an integration control gain of feedback compensator C₃(s)′.

[0210] It is noted that drive torque commandvalue correction quantity d_V(t)′ corresponds to external disturbance estimated value d_V(t) described above with reference to FIG. 16.

[0211] If drive torque command value correction quantity calculating block shown in FIG. 21 determines that the un-lockup state occurs according to the lock-up state signal LUs, this block calculates correction quantity d_V(t)′.

[0212] That is to say, d_V(t)′=[(K_P′·s+K_I′)/s][V_T(t)−V_A(t)].

[0213] In the above equation, K_P′<K_Pand K_I′<K_I.

[0214] Hence, drive torque command value d_FC(t) is calculated from reference drive torque command value d_FC1(t) and drive torque command value correction quantity d_V(t)′ as follows:

[0215] That is to say, d_FC(t)=d_FC1(t)+d_V(t)′.

[0216] Since the feedback gain is made small during the un-lockup time as compared with the lock-up time, a variation rate of the drive torque command value correction quantity is made small. Since the response characteristic can be conformed to that of the response characteristic of the object to be controlled delayed during the un-lock up state as compared with the lock-up state, the stability of the vehicular velocity control system during each of both times of the lock up state and un-lock up state can be assured.

[0217] Next, an actuator drive system shown in FIG. 10 will be described below.

[0218] A transmission ratio (gear ratio) command value calculating block 540 inputs drive torque command value d_FC(t), vehicular velocity V_A(t), and output of coast switch 30, and output of an accelerator pedal sensor 90, calculates a transmission gear ratio command value DRATIO(t), and outputs the calculated gear ratio command value DRATIO to the continuously variable transmission (CVT) 70.

[0219] (1) A case where coast switch 30 is in an off state.

[0220] Transmission (gear) ratio command value calculating block 540 calculates throttle opening angle estimated value TVO_EST1from a throttle valve opening angle estimation map as shown in FIG. 18 on the basis of the vehicular velocity V_A(t) and drive torque command value d_FC(t).

[0221] Then, transmission (gear or speed) ratio command value calculating block 540 calculates an engine revolution speed command value N_INCOMfrom a CVT transmission (gear) speed ratio map shown in FIG. 19 on the basis of the throttle valve estimated value TVO_EST1and vehicular velocity V_A(t). Then, speed ratio command value DRATIO(t) is derived from vehicular velocity V_A(t) and engine speed command value V_A(t) using the following equation.

[0222] DRATIO(t)=N_INCOM·2π·Rt/[60·V_A(t)·Gf], wherein Gf denotes a final gear ratio.

[0223] (2) A case where coast switch 30 is in an on state.

[0224] Transmission (gear or speed) ratio command value calculating section 540 holds a previous transmission (gear) ratio command value DRATIO(t−1) as transmission (gear) ratio command value DRATIO(t) in a case where vehicular velocity command maximum value V_SMAXis reduced with coast switch 30 operated to be turned on. Even if coast switch 30 is continuously turned to ON, the transmission (gear or speed) ratio holds a previous value immediately before coast switch 30 is turned to ON (namely, holds the previous value until coast switch 30 is turned from on state to the off state. Therefore, no downshift occurs. Hence, in a case where set vehicular velocity is returned to the original set vehicular velocity by accelerate switch 40 after set vehicular velocity is largely dropped, the engine throttle valve is controlled in an open direction to accelerate host vehicle. However, even if this occurs, since the downshift operation is not carried out, the engine speed does not quickly increase and a generation of noise given to the vehicular driver can be prevented.

[0225] An actual transmission (gear) ratio (speed ratio in the case of continuously variable transmission 70) calculating block 550 calculates actual speed ratio RATIO(t) according to an engine speed N_E(t) and vehicular velocity V_A(t) using the following equation:

[0226] That is to say, RATIO(t)=N_E(t)/[V_A(t)·Gf·2π·Rt].

[0227] It is noted that engine speed N_E(t) is detected by an engine speed sensor 80 detecting an engine crankshaft revolution angle from an ignition signal of engine.

[0228] Engine torque command value calculating block 560 shown in FIG. 10 calculates an engine torque command value TE_COM(t) in accordance with the following equation from drive torque command value d_FC(t) and RATIO(t).

[0229] That is to say, TE_COM(t)=d_FC(t)/[Gf·RATIO(t)].

[0230] A target throttle opening (angle) calculating block 570 calculates a target throttle opening angle TVO_COMfrom an engine whole performance map as shown in FIG. 20 on the basis of engine torque command value TE_COM(t) and engine speed N_E(t) and outputs calculated TVO_COM(t) to a throttle actuator 60.

[0231] A brake pressure command value calculating block 630 shown in FIG. 10 calculates an engine brake torque TE_COM′ during a full closure throttle from the engine whole performance map shown in FIG. 20, calculates a brake pressure command value REF_PBRK(t) from engine brake torque TE_COM′ and engine torque command value TE_COM(t) in accordance with the following equation, and outputs REF_PBRK(t) to abrake actuator 50.

[0232] That is to say, REF_PBRK(t)=(TE_COM−TE_COM′)·Gm·Gf/{4·(2·AB·RB·μB)}.

[0233] In the above equation, Gm denotes a speed ratio of the continuously variable transmission 70, AB denotes a wheel cylinder force (cylinder pressure·area), RB denotes a disc rotor effective radius, and μB denotes a brake pad frictional coefficient.

[0234] Next, a suspension processing of the vehicular velocity control will be described below.

[0235] A vehicular velocity suspension deciding block 620 shown in FIG. 10 inputs an accelerator manipulated variable APO detected by accelerator pedal sensor 90 and compares the accelerator manipulated variable APO with a predetermined value.

[0236] This predetermined value is equivalent to accelerator manipulated variable APO1 corresponding to target throttle opening angle TVOCOM inputted from a target throttle opening angle calculating block 570, namely, a value of the throttle valve opening angle corresponding to the automatically controlled vehicular velocity at that time point at which the automatic vehicular velocity control is carried out.

[0237] If accelerator manipulated variable is larger than the predetermined value, i.e., the vehicular driver has depressed the accelerator pedal, (in other words, when the throttle valve is opened more widely than the throttle valve opening angle opened by means of throttle actuator 60 at that time point), a vehicular velocity control suspending signal is outputted.

[0238] Upon receipt of the vehicular velocity control suspending signal, drive torque command value calculating block 530 and target throttle opening (angle) calculating block 570 initializes the present arithmetic operations (zeroes the variables) and the transmission (CVT) 70 performs a switching from a cruise run purpose speed ratio (gear ratio) map to a normal run purpose speed ratio map through the CYT controller. In other words, the cruise run through the automatic control is suspended and the normal run control is carried out which responds to the vehicular driver's accelerator manipulation.

[0239] The continuously variable transmission 70 is provided with the normal run purpose speed ratio map and the cruise run speed ratio map. During the suspension of the cruise run control, the vehicular velocity controlling block 500 outputs a switch command to the continuously variable transmission 70 to switch the map from the cruise run purpose speed ratio map to the normal run purpose speed ratio map. The normal run purpose speed ratio map is a control map such that the downshift during the acceleration is not moderate but is sharp (the response characteristic is favorable) and the cruise run purpose speed ratio map is a control map such that a comfortable feeling is given and response characteristic is moderate. Consequently, the switching causes the vehicular driver not to feel insufficiently.

[0240] In addition, vehicular velocity control suspension deciding block 620 halts the output of the vehicular velocity suspending signal when the accelerator manipulated variable APO is returned to a value smaller than the predetermined value and outputs a deceleration request to drive torque command value calculating block 530 in a case where the APO is returned to the value smaller than the predetermined value and vehicular velocity V_A(t) is larger than vehicular velocity command maximum value V_CMAX. Then, drive torque command value calculating block 530 outputs the calculated driving force command value d_FC(t) to target throttle valve opening angle calculating block 570 so as to achieve the deceleration control by the calculated throttle valve opening angle from the driving force command value d_FC(t) in the case where the output of the vehicular velocity suspending signal is halted and the deceleration request is inputted. In a case where a braking force only with the throttle valve fully closed is insufficient, transmission (gear) ratio command value calculating block 540 outputs the transmission (speed) ratio command value DRATIO (downshift request) to the continuously variable transmission so that the downshift operation of CVT 70 is carried out to compensate for the insufficient braking force, irrespective of the vehicular run on the ascending slope, descending slope or flat road to achieve the throttle valve opening angle control and the speed ratio control of the continuously variable transmission.

[0241] In addition, in a case where the driving (in this case, braking) force command value d_FC(t) is large and the braking force caused by the downshift operation of the continuously variable transmission 70 gives an upper limit, the driving force is normally (usually) compensated by the activation of brake system when the vehicle is running on the flat road. However, when the vehicle is running on the descending slope, a brake control inhibit signal Bp is outputted from drive torque command value calculating block 530 to the brake pressure command value calculating block 630, thereby the brake control during the run on the descending slope being inhibited. The reason that such a control as described above is carried out is as follows. That is to say, if the deceleration is carried out using the brake system when the vehicle is running on the descending slope, it becomes necessary to continue to be braked through the brake system. Such a problem as a brake fade phenomenon may occur. To avoid an occurrence of the above-described brake fade phenomenon, the vehicular velocity control is carried out to obtain the necessary braking force during the run on the descending slope only by the deceleration caused by the throttle valve opening angle and the downshift control of the continuously variable transmission 70 without use of the brake system of the vehicle.

[0242] According to the method described above, even if the cruise run control is recovered after the suspension of the cruise run control thorough a temporal depression of accelerator pedal by the vehicular driver, the downshift operation of the continuously variable transmission can achieve a larger deceleration than that only through the full closure of the throttle valve. Hence, a convergence time duration to the target vehicular velocity can be shortened.

[0243] In addition, since the continuously variable transmission 70 is used in the vehicular velocity control, a, so-called, gear shift shock does not occur even on the vehicular run on a long descending slope. Furthermore, since the deceleration is larger than that only through the full closure control for the throttle valve and both the throttle valve opening angle and transmission (gear or speed) ratio are controlled to achieve drive torque based on vehicular velocity command value variation rate ΔV_COM(t), a smooth deceleration can be achieved maintaining a predetermined deceleration.

[0244] It is noted that since the gear shift shock is found in a normally available multiple-ratio, geared transmission during the downshift operation, the downshift control for such an automatic transmission as described above has not been carried out but only the throttle valve full closure control has been carried out even in a case where the deceleration control request occurs if the automatic transmission has been used in a previously proposed vehicular velocity control system. However, since the continuously variable transmission is used, a smooth downshift operation can be achieved without gear shift shock. Consequently, a smooth deceleration can be achieved by an implementation of the above-described control at a larger deceleration than that only thorough the full closure control of the throttle valve opening angle.

[0245] Next, a stop process of the vehicular velocity control will be described below.

[0246] A driven wheel acceleration calculating block 600 shown in FIG. 10 inputs vehicular velocity VA(t) and calculates a driven wheel acceleration α_OBS(t) using the following equation. That is to say, α_OBS(t)=[K_OBS·s/(T_OBS·s²+s+K_OBS)]·V_A(t).

[0247] In the above equation, K_OBSdenotes a constant and T_OBSdenotes a time constant.

[0248] It is noted that since vehicular velocity V_A(t) in the above equation is a value calculated from a revolution velocity of the tire wheel (the driven wheel) as described above, the value thereof corresponds to the vehicular velocity of the driven wheel and the driven wheel acceleration α_OBS(t) is a value of a variation rate (driven wheel acceleration) of the vehicular velocity derived from driven wheel acceleration V_A(t).

[0249] A vehicular velocity control stop deciding block 610 compares driven wheel acceleration a α_OBS(t) with a predetermined acceleration limitation value α (α denotes the acceleration value corresponding to the variation rate of the vehicular velocity and, for example, 0.2 G). If driven wheel acceleration α_OBS(t) is in excess of the acceleration limitation value α, a vehicular velocity control stop signal is outputted. According to this vehicular velocity control stop signal, drive torque command value calculating block 530 and target throttle opening angle calculating block 570 initialize the arithmetic operations (calculations) thereof. It is noted that once the vehicular velocity is once stopped, a control is not returned to the vehicular velocity control until set switch 20 is again turned to ON.

[0250] The vehicular velocity controlling block 500 recited in FIG. 10 is a system controlling the vehicular velocity at the vehicular velocity command value based on the vehicular velocity command value variation rate ΔV_COMDetermined according to vehicular velocity command value variation rate determining block 590. Therefore, in the normal state, no variation in the vehicular velocity exceeding vehicular velocity command value variation rate limitation value occurs [for example, 0.06 G=0.021 (km/h)/10 (milliseconds)]. Hence, there is a high possibility that a slip occurs on the driven wheel(s) if driven wheel acceleration α_OBS(t) is in excess of a predetermined acceleration limitation value α (for example, 0.2 G) larger than a value corresponding to the above-described vehicular velocity command value variation rate limitation value. Thus, the slip on driven wheel(s) can be detected by comparing driven wheel acceleration α_OBS(t)with a predetermined acceleration limitation value α (for example, 0.2 G). Therefore, it is not necessary to install additionally an acceleration sensor in a slip suppression system such as a traction control system (TCS) or to detect a difference in revolution speed between one of the driven wheels and one of non-driven wheels. Driven wheel acceleration α_OBScan be derived according to the output of vehicular velocity sensor (the sensor to detect the revolution speed of driven wheel) so that a slip occurrence determination and a stop determination of the vehicular velocity control can be made.

[0251] In addition, the response characteristic to the target vehicular velocity can be improved by enlarging vehicular velocity command value ΔV_COM. It is noted that in place of a determination of the cruise run control stop according to a comparison result between driven wheel acceleration with a predetermined value, the vehicular velocity control stop may be carried out in a case where a difference between vehicular velocity command value variation rate ΔV_COMcalculated by vehicular velocity command value variation rate determining block 590 and driven wheel acceleration Δ_OSB(t) becomes equal to or larger than a predetermined value.

[0252] In addition, vehicular velocity command value determining block 510 determines whether the preceding vehicle is detected or not according to an indication of the preceding vehicle flag F. If determining that no preceding vehicle is detected, vehicular velocity command value determining block 510 determines whether vehicular velocity command value ΔV_COM(t) calculated by itself is higher than the inputted vehicular velocity V_A(t) and determines whether the vehicular velocity is changed in the deceleration mode (i.e., if V_SMAX<V_A(t)). Then, vehicular velocity command value V_COMis set to vehicular velocity V_A(t) or a smaller predetermined vehicular velocity VCOM(t) (for example, a value of subtraction of 5 km/h from vehicular velocity of the host vehicle). Then, an initial value of the integrator of each of C₁(s) and C₂(s) is set to vehicular velocity V_A(t) so as to zero the output of C₂(s)·V_A(t)−C₁(s)·d_FC(t)=dv(t) in drive torque command value calculating block 530 shown in FIG. 16. Consequently, each output of C₁(s) and C₂(s) indicates vehicular velocity V_A(t). Consequently, external disturbance estimation value d_v(t) indicates zero.

[0253] Furthermore, a case where ΔV_COM(t) which is a variation rate of V_COM(t) is larger at a deceleration side than the predetermined value (0.06G) is a timing at which the above-described stop control is carried out.

[0254] Since this reduces an unnecessary initialization (V_A(t)→V_COM(t) initialization and integrator initialization), a deceleration shock can be relieved.

[0255] In a case where the vehicular velocity command value (instantaneous control command value until the vehicular velocity reaches to a target vehicular velocity) is larger than actual vehicular velocity V_A(t) and a variation in time of the vehicular velocity command value occurs in the deceleration side, the vehicular velocity command value is modified to actual vehicular velocity or predetermined vehicular velocity equal to or lower than actual vehicular velocity so that the vehicular velocity can quickly be converged to target vehicular velocity. A control continuity can be maintained by initializing drive torque command value calculating block 530 using the set actual vehicular velocity or lower than the actual vehicular velocity.

[0256] It is noted that in a vehicular velocity control system in which the vehicular velocity is controlled to make the actual inter-vehicle distance coincident with a target inter-vehicle distance so that the vehicle is running maintaining the target inter-vehicle distance to the preceding vehicle set by the vehicular driver, the vehicular velocity command value is set to maintain the target inter-vehicle distance. However, if the vehicular velocity control system (corresponding to the vehicular velocity controlling block 500) determines that the preceding vehicle is detected according to the input preceding vehicle flag F, the modification of vehicular velocity command value V_COM(t) [V_A(t)→V_COM(t)] and the initialization of drive torque command value calculating block 530 (specifically, the integrator incorporated therein) are carried out in a case where the actual inter-vehicle distance is equal to or shorter than the predetermined value and the vehicular velocity command value variation rate ΔV_COM(t) is larger than the predetermined value (0.06 G) in the deceleration side. Thus, the actual inter-vehicle distance can speedily be converged to the target inter-vehicle distance. Hence, there is no possibility of excessive approach to the preceding vehicle and the continuity of control can be maintained. In addition, this causes reductions in unnecessary initializations [initialization of V_A(t) ΔV_COM(t)] and initialization of the integrators] so that the number of the deceleration shocks becomes reduced.

[0257] It is noted that the vehicular driver is included in a vehicular occupant or an operator who operates the automatic vehicular velocity controlling system mounted in the host vehicle through the above-described switches.

[0258] The entire contents of Japanese Patent Applications No. 2000-148742 (filed in Japan on May 19, 2000), No. 2000-148762 (filed in Japan on May 19, 2000), No.2000-148847 (filed in Japan on May 19, 2000), and No. 2000-143581 (filed in Japan on May 16, 2000) are herein incorporated by reference.

[0259] Modifications and variations of the embodiment described above will occur to those skilled in the art in the light of the above teachings. The scope of the invention is defined with reference to the following claims.

[0260] Industrial Applicability

[0261] Automatic vehicular velocity controlling system and method according to the present invention is applicable to an automotive vehicle in which a continuously variable transmission and an inter-vehicle distance sensor to detect a distance from the vehicle to a preceding vehicle which is running ahead of the vehicle are equipped.