Continuously Variable Transmission Control System

U.S. patent application Ser. 07/592,933 claiming priority on Japanese Patent Appln. No. 1-259376.

The present invention relates to a ratio control system for a continuously variable transmission in a motor vehicle.

U.S. Pat. No. 4,735,113 discloses a continuously variable transmission for a vehicle. According to a known ratio control system for the continuously variable transmission, a feedback control is effected whereby a reduction ratio in the continuously variable transmission is controlled in such a direction as to decrease an input revolution speed error or deviation toward zero. In this known feedback control, proportional and integral gains are constant. If such feedback gains are set so as to effect an optimum upshift toward the smallest or minimum reduction ratio during operation with large throttle openings, an excessively quick upshift results during operation with small throttle openings. On the contrary, if the gains are set so as to effect an optimum upshift during operation with small throttle openings, an excessively slow upshift results during operation with large throttle openings. Thus, a difficulty arises in setting the feedback gains.

An object of the present invention is to provide a ratio control for a continuously variable transmission which has solved the above-mentioned problem.

According to the present invention, there is provided a ratio control system for a continuously variable transmission which is shiftable from a present reduction ratio to a target reduction ratio in response to an output signal given by a feedback control equation involving at least one gain and an input revolution speed deviation, wherein the gain is variable with a first time derivative of the target reduction ratio.

According to another aspect of the present invention, there is provided a method of controlling a ratio change in a continuously variable transmission in a motor vehicle, the method comprising the steps of:

repeatedly determining a target value in a predetermined variable representative of a reduction ratio in the continuously variable transmission;

repeatedly determining a first time derivative of said target value;

repeatedly setting proportional and integral gains of a feedback control equation in response to said first time derivative of said target value;

determining an output signal from said feedback control equation; and

controlling a ratio change in the continuously variable transmission in response to said output signal.

FIG. 1 is a characteristic curve of a gain versus a first time derivative of a target reduction ratio, the gain being representative of both a proportional gain and an integral gain;

FIG. 2 is a schematic illustration of a continuously variable transmission drivingly connected to an engine of a motor vehicle;

FIGS. 3A and 3B, when combined, are a hydraulic circuit diagram showing a control system for the continuously variable transmission;

FIG. 4 is a block diagram showing a control unit for executing a flowchart shown in FIG. 5; and

FIG. 5 is the flowchart illustrating the gist of the invention.

Referring to the accompanying drawings, and particularly to FIGS. 2, 3A, 3B, and 4, reference is made to a description in connection with FIGS. 2, 1A, 1B, and 3 of U.S. Pat. No. 4,735,113 which is hereby incorporated by reference.

Referring to the flowchart shown in FIG. 5, at a step 502, reading operation is performed to store output signals of a vehicle speed sensor 302, a throttle opening sensor 303, and a turbine revolution speed sensor 305 as a vehicle speed VSP, a throttle valve opening TVO and an input revolution speed Nin, respectively. At the subsequent step 504, a target input revolution speed T.multidot.Nin is determined after performing a table look-up operation of a predetermined shift point mapping based on vehicle speed VSP and throttle opening degree TVO. A target reduction ratio inew given in the preceding operation cycle of this routine is stored as an old target reduction ratio iold. Then, at a step 506, the target reduction ratio inew is updated by a result from calculating an equation that T.multidot.Nin/VSP.times.C, where C is constant. At the next step 507, a first time derivative of reduction ratio di is given by substracting iold from inew. Then, at a step 508, a proportional gain Kp is given as a function of the first time derivative of reduction ratio di, namely Kp=fl(di), and an integral gain Ki is given as a function of the first time derivative of reduction ratio di, namely Ki=f2(di). The functions fl(di) and f2(di) are illustrated by a characteristic curve shown in FIG. 1. As shown in FIG. 1, the proportional and integral gains Kp and Ki are proportional to the first time derivative of reduction ratio di. As readily seen from FIG. 1, the gains are larger when the first time derivative of reduction ratio di is relatively large than they are when the first time derivative of reduction ratio di is relatively small. At the subsequent step 509, an input revolution speed error or deviation e is given by subtracting the actual input revolution speed Nin from the target input revolution speed T.multidot.Nin, namely e=T.multidot.Nin-Nin. At the next step 510, a target motor position 74 m is given after calculating a feedback control equation that .theta. m=Kp.multidot.e+.intg.Ki.multidot.e. Then, at a step 512, a control signal is output whereby a shift actuator in the form of a stepper motor 110 is rotated to the target motor position .theta. m. Subsequently, the other control is processed at a step 514.

According to the above described control process, the proportional and integral gains Kp and Ki are increased as the first time derivative of target reduction ratio di is increased, and the target motor position .theta. m is also increased. Accordingly, when the throttle opening degree TVO falls in a wide open throttle opening area, a speed of ratio change becomes high to provide a good shift response, while when the throttle opening degree TVO falls in a small throttle opening area, a speed of shift becomes low, thus avoiding occurrence of overshoot phenomena.