FR2737760A1 - Method of control of automatic transmission in motor vehicle - involves using fuzzy logic rules to determine ratio changing with different rules for acceleration and deceleration, and using position of accelerator pedal to change driving regimes - Google Patents

Abstract

The method of control of ratio changing involves using data on the operation of the vehicle to guide decisions on changing gear ratio. If the vehicle is accelerating the decision to engage a different ratio is taken by application of fuzzy logic rules concerned with the difference between the requested acceleration and the potential acceleration of the vehicle in each of the available ratios. If the vehicle is braking or decelerating the decision to engage a lower ratio is taken from fuzzy logic rules concerned with the speed of the engine. Different regimes are selected by detection of the position of the accelerator pedal.

Description

METHOD FOR CONTROLLING AUTOMATIC TRANSMISSION
The present invention relates to the control of automatic transmissions. More particularly, it relates to a method of controlling gear changes on an automatic transmission with staggered ratios.

The automatic gearbox control strategies developed by the manufacturers are mainly designed to take into account the road conditions, for example to automatically adapt the behavior of the box during a trip in the mountains, reduce the number of shifts in curves or on winding roads, etc.

Publication EP 0503 948 discloses a method for controlling gear changes in an automatic transmission based on the use of five laws of passage, dedicated respectively to urban and flat terrain routes, to particularly winding mountain paths, to descents requiring engine braking, descents requiring a lot of engine braking, and mountain paths with little winding. According to this publication, a control system recognizes the type of path taken by the driver, thanks to the fuzzy logic processing, of various parameters measured on the vehicle, so that this system is able to choose, in all circumstances, the law of passage best suited to the journey.

The driving strategy described in the publication EP 0503 948 has the main advantage of taking into account the behavior of the driver, and the driving conditions of the vehicle, and to automatically adjust the control passages to those ci. On the other hand, it does not take into account its true intentions as to whether it wishes to accelerate, brake or maintain its constant speed.

As a result, the rules implemented in this strategy do not include fuel consumption criteria. Finally, the principle of switching between mappings of predetermined pass laws results in significant memory congestion in the calculator, and the difficulty of adopting in real time the solution best suited to driving conditions, the development different maps of passage assuming in all cases a large number of operations.

Other known steering strategies aim to adapt the behavior of an automatic gearbox to the driving style of each user. The publication FR 2697 884 describes a gearbox switching device capable of modifying the behavior of the gearbox after having identified a "nervous" behavior, based on an analysis of the actuation of an element. power control, such as the throttle valve. This publication therefore provides for modifying the behavior of an automatic gearbox according to the driving style, but does not propose a control strategy of the passages making it possible to apply such modifications.

The object of the present invention is to establish a global strategy for controlling gear changes on an automatic gearbox, taking into account each situation, the driving style, the real intentions of the driver, the consumption, and the conditions of the gearbox. use of the engine.

The present invention aims to eliminate the aforementioned drawbacks, by directly taking into consideration the intentions of the driver, and by inducing the volume of data stored in memory.

It relates to a method of controlling the gear changes of an automatic transmission from signals relating to the gear engaged, the engine speed, the opening of the throttle valve, the depressing of the brake pedal, when the accelerator pedal is depressed and at the speed of movement of the vehicle.This method is characterized in that if the vehicle accelerates, the decision to engage a report nl different from n is taken in application of logic rules fuzzy relative to the belonging of the difference between a setpoint acceleration γ; c and the potential acceleration of the vehicle ri on each of the ratios of the transmission to predetermined ranges of difference, whereas if the vehicle brakes or decelerates, the decision to commit a report n2 is taken under fuzzy logic rules relating to the membership of regime #n on ratio n and regime # n-1 on ratio nl, to the predetermined rhyme ranges.

The invention however provides that if the vehicle is traveling below a speed limit, the first gear ratio is imposed.

According to the invention, the decision to engage a ratio n2 to adapt the behavior of the vehicle deceleration is taken if the driver presses his brake pedal, if the opening is lower than a first threshold ai or the derivative a 'of a with respect to time is less than a second threshold g, whereas the decision to engage a ratio nl making it possible to adapt the acceleration behavior of the vehicle is considered in all the other cases.

Other features and advantages of the invention will become clear from reading the following description of a particular embodiment of those, in conjunction with the appended drawings in which:
- Figure 1 shows the different parameters taken in
account,
- Figure 2 shows the overall strategy implemented
artwork,
- Figure 3 relates to the determination of the ratio of
transmission in a situation of acceleration,
- Figure 4 relates to the determination of the ratio of
transmission in a deceleration situation,
FIGS. 5A and 5C illustrate the application of
fuzzy rules in a deceleration situation,
- Figure 6 illustrates the application of fuzzy rules to
measures of the deceleration of the vehicle,
FIGS. 7A and 7B illustrate the application of rules
fuzzy in acceleration,
- Figures 8A, 8B and 8C relate to rules of
consumption reduction, and
FIGS. 9A and 9B relate to rules relating to
criterion of respect for the operation of the engine.

FIG. 1 shows various sensors associated respectively with the engine 1 of a vehicle, with its automatic transmission 2, and with the manual selector 3 thereof, including the usual functions D (drive), N (neutral) and R ( reverse). These sensors connected to a transmission control unit 4 of the transmission are as follows. A first sensor, or gear ratio sensor 6 is mounted on the transmission 2. A second sensor 7 mounted on the engine 1 detects the speed X thereof. A third sensor 8 delivers a signal a representative of the opening of the throttle valve expressed in degrees. A fourth sensor 9 delivers an all-or-nothing accelerator pedal position signal k (k = 0 or 1) detecting depression of the accelerator pedal by the driver, or "kick-down", corresponding at a demand for maximum torque. A fifth sensor, or brake sensor 11, also of the all-or-nothing type, signals the driver's support on his brake pedal. Finally, a sixth sensor 12 delivers a signal Yt representative of the transverse acceleration of the vehicle, expressed in meters per square second (m / s), and a seventh sensor 13 indicates the linear speed of movement of the vehicle v in meters per second (m / s).

According to the invention, the control unit 4 processes the signals emitted by all of these sensors in order to determine the ratio n to be retained or to engage on the transmission 2. This determination, carried out using calculation rules pertaining to the so-called "fuzzy hierarchical" technique is carried out at a regular rate, for example of the order of 30 Hz, when the driver has selected the D (drive) mode using his selector 3.

FIG. 2 indicates that the determination of the gear to be engaged is based on the evaluation of the parameters f, a, and a '(derived with respect to the time of the throttle opening computed in step 201). The parameters are used in the calculation step 202 of checking whether one of the following conditions is fulfilled:
- depressed brake pedal (f = 1), or
butterfly opening less than a first threshold,
a <al, or
- derived from the butterfly opening less than a second
threshold, a <a'2.

The calculation step 202 makes it possible to establish whether the driver brakes or releases the foot of the accelerator. If this is the case (signal 0), step 203 is imposed in order to determine the ratio n2 making it possible to adapt the engine brake to the deceleration of the vehicle, otherwise (signal 1) it is the step 204 that is set in order to determine the ratio nl which will make it possible to adapt the acceleration of the vehicle to the acceleration request of the driver. According to step 205, the report to be engaged can be the ratio nl, the ratio n2, or necessarily the first gear ratio, if the vehicle is traveling slowly (v <vmin) for example below 10 km / h.

FIG. 3, relating to the calculation step 204, shows that it exploits the information n, a, and v as well as a signal s relating to the driving style of the driver, between 0 and 1 and determined, for example in the acceleration phase, based on an analysis of the ratio between the desired setpoint acceleration by the driver and the maximum setpoint acceleration that he would obtain at full speed, and / or the value of the measured deceleration &gamma; m in the deceleration phase.

ki étant le rapport de démultiplication global de la transmission sur le rapport i, et r étant le rayon des roues du véhicule.

The signals, v, n and s are used in step 304, to calculate a setpoint acceleration γ, representative of the acceleration desired by the driver, for example by weighting by the coefficient s of the driving style an acceleration setpoint, called economic, SYcons-eco, and an acceleration of sports instruction, sport-cons, so-called sports, calculated from the measures of a and v so as to obtain respectively a setting of economic type favoring the minimization of the consumption , and a sports type setting favoring the brilliance of the car, for example in application of the formula ::
The calculation step 301 determines the potential regimes # 1 of the engine on each of the ratios of the transmission, by the formula:

where k is the overall ratio of the transmission on the ratio i, and r is the radius of the wheels of the vehicle.

- C étant le couple moteur selon la cartographie de celuici, pour les valeurs de a et oi existantes, -r étant le rayon de la roue du véhicule -M étant la masse standard du véhicule transportant deux personnes, - Ta étant la traînée aérodynamique du véhicule, calculée par exemple selon la formule

oùp est la densité de l'air, et S le coefficient de pénétration dans l'air du véhicule, et V la vitesse du véhicule, et - Tr étant la traînée de roulement du véhicule, calculée par exemple selon la formule

oùgest
l'accélération de la pesanteur et kr un coefficient de
résistance au roulement.The calculation step 302 determines the potential accelerations of the vehicle ai on each of the reports of the transmission, for example as follows:

- C being the engine torque according to the cartography thereof, for the values of a and oi existing, -r being the radius of the wheel of the vehicle -M being the standard mass of the vehicle carrying two people, - Ta being the aerodynamic drag of vehicle, calculated for example according to the formula

wherep is the density of the air, and S is the coefficient of penetration into the vehicle air, and V is the speed of the vehicle, and - Tr is the rolling drag of the vehicle, calculated for example according to the formula

oùgest
the acceleration of gravity and kr a coefficient of
rolling resistance.

The calculation step 303 determines the potential instantaneous consumption Ci on each report (in liters per 100 km), in particular from the engine-specific consumption map statements.

Finally, the computation step 305 ensures the determination of n1 from the set of parameters n, ri, wi, ci and a by applying fuzzy logic type calculation algorithms making it possible to weight each of them according to the importance that we wish to bring to the operating conditions they allow to quantify. These calculation methods, known per se, allow the simultaneous consideration in the decision process of cumulative or contradictory objectives, such as the brilliance of the engine, the respect of its operation, the minimization of consumption, the responsiveness of the system. , etc ...

Figure 4, relating to step 203 of Figure 2, specifies the nature of the parameters and operations leading to the determination of n2.

These parameters are n, s n-l the engaged ratio, the engine speed calculated on the lower ratio rm, deceleration of the vehicle and co. The calculation of rm is performed from a v low pass type filtering (step 402) and the derivation of the filtered rate signal vf (step 403). The determination of n2 is carried out in step 404 by applying fuzzy logic algorithms.

FIG. SA reproduces a table indicating fuzzy logic rules applicable to determining the ratio in the deceleration situation, in step 404. According to this table, the engine speeds o, and xn-1 on the ratios n and nl are classified " low "," medium "or" high ". The downshift decision may be made based on whether xn and xn-1 belong to these three overlapping ranges in accordance with FIGS. SB and 5C, and vary according to the deceleration, so that the more the deceleration is negative (the more the driver brakes), the higher the terminals B1 to B4 of these ranges, in order to increase the engine brake accordingly. The shift on the ratio n + 1 will be inhibited when the driver performs a fast foot lift of its accelerator pedal because the strategy resulting from Figure 5A does not provide for passage over the upper gear.

According to the invention, the values required for the parameters
B1, B2, B3 and B4 of the terminals of these three fuzzy subsets can themselves be calculated by a method called fuzzy logic, depending on the membership of the deceleration to the three subsets defined for example according to the Figure 5C.

Moreover, the deceleration value Srm calculated in step 403 is compared with the value ranges as defined by FIG. 6 and with a coefficient defining a degree of membership in each range noted # i, where i is the index of the corresponding range.

The invention proposes to calculate the values B1, B2, B3 and B4 by conventional resolution or defuzzification methods of the rules of FIG. 6. The numerical values mentioned in the table of FIG. understood as a mere indicative value. In addition they can be modules during the development of the strategy. Once determined, these values make it possible to calculate degrees of membership of a and edn-1 at the three sets defined above, eg notes A and j, j equal to 1, 2 or 3, being the index of the subset. corresponding set.

As before, the application of conventional defiizzification methods makes it possible to determine a ratio, calculated for example from the conclusions of the fuzzy rules of FIG. 10, and the degrees of membership of the regimes #n and # n-1. This report, which is not necessarily an integer, makes it possible to determine n2 by a simple boundary of the type:
if y # 3,5, n2 = 4
if y # [2.5; 3.5 [, n2 = 3
if y # [1.5; 2.5 [, n2 = 2
siy <1.5, n2 = 1
FIG. 7A reproduces, by way of non-limiting example, a summary table of decision rules contributing to determining the ratio, or in such a way that it is the closest to FYc. These rules are based on the calculation of the differences #i = &gamma; i - &gamma; c on each report of the box (four in the example described) and on the computation of the degree of membership between 0 and 1 of ei to five subsets notes TN (very negative), N (negative) ), Null, P (positive) and TP (very positive), for example from the indications of Figure 7B defining partially overlapping ranges of values. If the degree of belonging of ei to each subset TN, N, Null,
#
P, TP is denoted i, j, i being the index of the ratio and j the index of the subset (j = 1 to 5 from TN to TP), each conclusion in terms of
# ratio decision will be assigned the coefficient i, j corresponding to it in such a table, which will be taken into account in the algorithms of step 305.

FIG. 8A reproduces another summary table, making it possible to implement the function of minimizing the consumption in the calculation step 305. On this table appears the box C + between the potential consumption on the report n uses and on the ratio higher n + 1 and the difference in consumption between the third and fourth C3 ratios. Depending on the value of the ratio n used, the degree of membership between 0 and 1, the three subsets N (N4ative), Nul or P (positive) as defined for example in Figure 8B is accompanied by a coefficient noted for example yk where k is the index of the corresponding fuzzy set which will be taken into consideration together with H, j in the calculation step 305. In this respect, the invention also provides for the implementation of a function of deactivation of consumption rules. This function, illustrated by FIG. 8C, aims at reducing or even canceling the overall effect of the rules in question at step 305 by modulating them by a coefficient of between 0 and 1, reflecting the fact that more the driver accelerates, that is to say more is important, the less he cares about his consumption.

FIG. 9A reproduces a third summary table allowing, in conjunction with FIG. 9B, to take into account the criterion of respect for the operation of the engine "in the algorithm developed in 305, as follows: the potential regimes on each ratio of the transmission, the membership degrees of the variables wi are assigned a degree of membership of the fuzzy subsets in accordance with a graph such as that of FIG. 9B. The degree of membership can for example be denoted flil, i being the index of the ratio and 1 the index of the corresponding set (1 = 1 for the first set (low schemes) and 1 = 2 for the second set (high schemes).) As before, each proposed shift decision according to Figure 9A is assigned the corresponding coefficient Pil.

The determination of the ratio n1 in the step 305, corresponding to a defuzzification of the parameters and coefficients mentioned above, can be carried out in various ways, including for example notions of hysteresis for adjusting the sensitivity of the driving method, the notion downshifting, an anti-pumping strategy that makes it possible to adapt the gear engaged to the mass of the vehicle, or a strategy of behavior in turn ensuring the adaptation of the control to the sinuosity of the course. In all the cases, the decision to engage a report nl will be taken by weighting the instructions resulting from the calculation of the potential accelerations and the setpoint acceleration, the calculation of the potential regimes and that of the potential accelerations.

In conclusion, it should be emphasized that the application of so-called "fuzzy hierarchical" calculation techniques proposed by the invention makes it possible to take into account the real desires of the driver in terms of acceleration or braking while reducing consumption and respecting the operation of the engine through the rules of consumption and respecting the operation of the engine through the regime rules of that. In particular, the invention makes it possible to adapt the downshifts to the measured deceleration level, representative of the level of braking desired by the driver, without resorting to information of the "brake pressure" type, or degree of depression of the pedal. Furthermore, the strategy implemented easily adapts to all types of engine as well as to the driving style of the driver, its development requires only a small memory and a limited number of operations. Calculation.

Finally, thanks to the invention, the total number of adjustments to be made is less than with pre-established passage maps integrating various potential functions (brio, reduction of consumption, management of organs), the development of which is particularly delicate.

Claims (10)

1) method for controlling the gear changes of an automatic transmission from signals relating to the gear engaged n, to the engine speed o, to the opening a of the throttle valve, to the depression f of the pedal of the brake, the depression k of the accelerator pedal and the speed of movement v of the vehicle, characterized in that if the vehicle accelerates, the decision to engage a report nl different from n is taken in application of rules fuzzy logic relating to the membership of the difference between an acceleration of setpoint &gamma;; c and the potential acceleration of the vehicle ri on each of the reports of the transmission to predetermined ranges of differences, while if the vehicle brakes or decelerates , the decision to commit a report n2 different from n is taken under fuzzy logic rules relating to the membership of the regime #n on the ratio n and the regime xn-l on the ratio nl, to pla predetermined regime conditions. 2) A control method according to claim 1, characterized in that if the vehicle remains below a limit speed, the first gear ratio is imposed. 3) A control method according to claim 1 or 2, characterized in that the decision to engage a report n2 to adapt the behavior of the vehicle deceleration is taken if the driver presses his brake pedal, if the opening a is less than a first threshold ai or if the derivative a 'of a with respect to time is less than a second threshold a2, while the decision to initiate a report nl making it possible to adapt the behavior of the vehicle in acceleration is envisaged in all other cases. 4) Control method according to claim 1, 2 or 3, characterized in that the decision to engage the ratio n2 is taken according to the degree of membership of xn and xn-1 to fuzzy subsets partially overlapping . 5) Control method according to claim 4, characterized in that the values bounding the value ranges of the engine speed are themselves determined according to fuzzy logic rules according to the membership of the acceleration or deceleration SYm from the vehicle to fuzzy subsets. 6) A control method according to one of the preceding claims, characterized in that the ratio n is determined by taking into account the difference between the consumption on the engaged ratio n and on the upper ratio n + 1. 7) A control method according to claim 6, characterized in that the decision taken to initiate the report nl takes into account the degree of belonging of the difference between the potential consumptions on the ratio n and the ratio n + 1 , as well as the difference between the potential consumptions on the two higher ratios of the transmission, with fuzzy sub-levels. 8) A control method according to one of the preceding claims characterized in that the decision to engage the report n2 takes into account the degree of membership of the potential schemes on each of the reports of the transmission to fuzzy sets. 9) Control method according to one of the preceding claims characterized in that the decision to engage the report nl taken by weighting the instructions results from the calculation of potential accelerations and the acceleration of instructions, potential regimes and calculation of potential consumption. 10) A control method according to one of the preceding claims, characterized in that the setpoint acceleration is calculated from the speed v of the opening angle a and a coefficient s, relative to the driving style of the driver.  11) Control method according to one of claims 7 to 10, characterized in that it comprises a function of deactivation of the consumption criterion, a function of a.