EP1924478A1

EP1924478A1 - Method for controlling two actuators of a vehicle capable of being responsive to a common request

Info

Publication number: EP1924478A1
Application number: EP06808264A
Authority: EP
Inventors: Lionel Cordesses; Mehdi Gati; Ophélie Thomassin
Original assignee: Renault SAS
Current assignee: Renault SAS
Priority date: 2005-09-13
Filing date: 2006-08-31
Publication date: 2008-05-28
Also published as: JP2009508463A; WO2007031665A1; KR20080049093A; US20080288126A1; FR2890629B1; FR2890629A1

Abstract

The invention concerns a method for controlling multiple actuators (1, 2) of a vehicle capable of being responsive to a common request, at least one of the actuators having a bandwidth and/or a saturation, which consists in determining for at least one of the actuators (1, 2) a setpoint value (u1,u2) integrating an output physical quantity (T1T2) of at least another of the actuators or of the other actuator, such that the actuators or at least some of them operate jointly.

Description

METHOD FOR CONTROLLING TWO ACTUATORS OF A VEHICLE THAT CAN MEET THE SAME DEMAND

The invention relates to the control of actuators on board a vehicle.

In a vehicle, the decoupling between the driver and the various actuators is more and more frequent. Ten years ago, the first motorized butterflies of fuel flow control appeared in the automobile, and since then there is more and more a tendency to break the direct links between the driver and the mechanical parts. Hybrid engines in particular require this decoupling because an ordinary driver would be unable to drive a car by managing both engines heat and electric. Similarly, in an electric braking system, the fact that the driver presses on the brake pedal is interpreted as a braking instruction.

In this perspective of decoupling appear new servo control problems of the actuators. As we move from the control of a single actuator to the control of several actuators each having their own dynamics and their own operating range (saturation). A good example of this scenario is the brake of a car that can deliver only negative torque and has a different dynamics of the dynamics of the engine which, moreover, provides mainly positive torque.

The problem is to enslave a multivariable system saturated input (bandwidth) and / or output (see Figure 1). The very fact of enslaving a system with input saturations is a problem in itself. Answers to this kind of problem exist. But the fact that the system has multiple inputs makes the problem more difficult to deal with "classical" approaches. The specifications dictate that the output torque be as faithful as possible to the reference given by the driver, while making the best use of the dynamic characteristics of the actuators available.

Solutions that deal with similar problems are grouped below into two categories. The first category includes scientific articles.

An example of a linear control is described in "Sei-Bum Choi and Peter Devlin. Throttle and brake combined control for intelligent vehicle highway Systems. SAE Technical Paper Series, pages 53-60, August 1995. " The servo control of the engine is achieved using a sliding mode control, the objective being to minimize the distance between vehicles and the speed difference between two cars, the ultimate goal being to cruise control. The brake is controlled with a feed forward part in order to compensate for nonlinearities of the model (mainly hysteresis) and to offer a proportional return for monitoring the driver's instructions. The switching strategy is based on the principle of using the motor when positive torque is requested, and the use of the brake when the engine brake is insufficient to satisfy the braking demand. Two thresholds on the opening of the throttle angle are fixed (ai> α ₀ ), so that when the throttle opening is less than αo, the brake is applied. When the butterfly opening becomes larger than ai, the engine is tilted.

An optimal control solution is proposed in "Kyongsu Yi, Youngjoo Cho, Lee Sejin, Joonwoong Lee, and Namkyoo Ryoo. A throttle / brake law for vehicle. FISITA World Automotive Congress, pages 1-6, June 2000 ". The authors present a control strategy on three layers. In the first layer, the reference acceleration is generated by calculating the optimum acceleration to reach a certain speed of the vehicle and maintain a certain distance between two vehicles that follow each other. This acceleration passes through a saturation in order to avoid saturation of the two actuators. In the second layer, the acceleration demand distribution is effected between the actuators according to whether the acceleration of the vehicle is less than or above a certain threshold. Servo control of the powertrain is achieved with a P1, that of the brake is achieved using a PID plus a part of feed forward, the latter part being performed in the third layer.

In these solutions, the control of each of the two members is achieved independently of the other. The control law is therefore quite simple and inexpensive in computing time. However, these same solutions have certain disadvantages. The switching strategy between the two actuators is empirical. The switching thresholds are chosen arbitrarily. Failure to take into account the saturations of the actuators in most jobs can lead to a deterioration in the performance of the closed loop when the actuator reaches the limit of its operating range.

In the documents EP-0 798 150, EP-0 896 896 and US-5,054,570, which relate to a subject similar to the problem dealt with in our case, the proposed solutions make it possible to regulate the speed of the vehicle as a function of the distance which separates it from another vehicle and the difference in speed between these two. The switching between the acceleration and deceleration actuators is abrupt when certain thresholds have been crossed. The thresholds are fixed in an arbitrary manner and no criterion on the choice of these is given. Switching from one actuator to another simplifies the problem of studying stability. Each actuator is slaved independently of the other. The control laws remain fairly simple and therefore do not take too much computing time. The choice of thresholds is totally arbitrary and no indication is given on the criterion that allows the choice of these. We are limited by the bandwidth of the actuators since they are each used on their side. Sudden switching between the actuators may cause discontinuities in the delivered torque.

As indicated, the invention aims to improve the control of two actuators responding to the same request. For this purpose, provision is made according to the invention for a method of controlling several actuators of a vehicle capable of responding to the same request, at least one of the actuators having a bandwidth and / or a saturation, in which one determines for at least one of the actuators a setpoint taking into account an output of at least one other actuators or other actuator, so that the actuators or at least some of them act together.

The present invention aims to address the problem of control of two actuators that we will qualify, in the remainder of the document, asymmetrical (bandwidths and / or different saturations). As will be seen, the approach disclosed allows the synthesis of a control law for distributing the torque demand expressed by the driver between the various actuators.

The method according to the invention may also have at least one of the following characteristics:

a reference is determined for each actuator taking account of an output quantity of at least one other of the actuators or of the other actuator;

at least one of the actuators determines a setpoint taking account of an output quantity of each of the other actuators or of the other actuator;

a reference is determined for each actuator taking into account the output quantity of each of the other actuators or of the other actuator;

the determination is implemented by consulting a cartography;

- at least one of the following data is used as input data of the map:

the output quantity of at least one of the actuators; and

the sum of the output quantities of the actuators.

an integer i is determined such that:

Mi [T ₁ T ₂ ... T _in ] ≤ mi Where: Mj and rrij are predetermined matrices associated with i; T _n is the output quantity of the actuator n; and T _in is a quantity corresponding to the demand;

the mapping is generated by means of a constrained optimization algorithm;

- Mapping is generated by quadratic multiparameter programming;

the determination is implemented by means of a calculation;

- we calculate:

where: U _n (k) is the setpoint associated with the actuator n with k sampling parameter;

Lj and Ii are matrices given by mapping; and

T _n is the output quantity of the actuator n; and

T _in is the magnitude corresponding to the demand.

- The or each output quantity (Ti ₁ T ₂ ) is determined and the determination of the or each instruction is repeated, taking into account the or each determined quantity.

The invention also provides a vehicle comprising:

actuators capable of responding to the same request, at least one of the actuators having a bandwidth and / or saturation; and

a control member, the control member being arranged to determine for at least one of the actuators a setpoint taking into account an output quantity of at least one of the actuators or of the other actuator so that the actuators or at least some of them act together. Other features and advantages of the invention will become apparent in the following description of a preferred embodiment given by way of non-limiting example with reference to the accompanying drawings, in which:

FIG. 1 is a flowchart illustrating a configuration of the actuators to which the invention applies;

- Figure 2 is a view similar to Figure 1 showing the feedback loops involved in the context of the invention;

FIG. 3 is a diagram illustrating a request for stepped torque and the torque obtained at the time of a simulation of the operation of the invention;

FIG. 4 illustrates the reference signals sent to the motor and to the brake as well as the output torques produced by them in correspondence with the diagram of FIG. 3;

FIGS. 5 and 6 are two diagrams similar to FIGS. 3 and 5 corresponding to a request for ramp torque; and

FIG. 7 is a flowchart illustrating the progress of the method according to the invention.

In the present embodiment, a vehicle equipped with two actuators 1 and 2 respectively formed by a motor 1 and a braking device 2 will be considered. The engine may be an internal combustion engine, a gasoline engine, a diesel engine or an engine. electric, or even a hybrid engine.

These two actuators 1, 2 are each capable of providing a torque to satisfy a torque demand T _ref , formulated by the driver by means of the accelerator pedal or the brake pedal for example. The two actuators are able to act together so that the torques provided by them two add up to provide a torque output T _SO .

Both actuators each have their own bandwidth and operating range as shown in blocks 3.5. Thus, as shown in Figure 2, the motor can provide positive torque when a positive torque demand is formulated. When a couple request negative is formulated, it provides a zero torque. Moreover, the positive torque that can be supplied can not exceed a maximum value. Conversely, the braking device can only provide negative torque when negative torque is requested, this supply being also limited in absolute value by a maximum value. It provides zero torque in a positive torque request.

As can be seen, the operating ranges of the two actuators are therefore not overlapping here. Nevertheless, the invention is applicable in the case where the actuators have overlapping operating ranges. It is even particularly advantageous in this case.

Similarly, the number of actuators is here limited to 2. However, the invention may be applied to vehicles in which the number of actuators that can cooperate to respond to a request of the same nature is greater than or equal to 3.

The aim of the invention is to achieve simultaneous control of these two asymmetric actuators. For this, it implements a control algorithm based on the calculation of the explicit solution of a quadratic optimization problem under constraints.

For the implementation of the invention, the vehicle comprises a control member such as a computer or microcontroller 4 adapted to generate instructions Ui and U ₂ to control the respective actuators 1 and 2. In addition, the vehicle comprises feedback sensors informing the control member 4 output quantities Ti ₁ T ₂ actually generated by these actuators.

We will first expose the theoretical foundations of the invention and then present its practical implementation.

In Figure 1, we give the schema of the problem we want to treat. The invention is presented in the context of controlling an engine and a brake. We want to provide some torque here using two actuators. Each actuator delivers torque within a certain range expressed in using input saturations. This torque is delivered with a dynamics specific to each actuator (bandwidth and saturation).

In Figure 2, we present the block diagram of the control strategy. The inputs required to carry out this servocontrol are also defined.

The primary objective of the control law is to carry out a reference tracking as perfect as possible between the input T _in and the output T _or t of the system. We try to minimize this error: e = (T _1n - T _out ) ²

Since each of the two actuators considered has a dynamics that can be approximated by a first order, the model of the system can be expressed in the following form:

With

- ζi the time constant of the i ^th actuator;

- Tj the torque delivered by the ^ith actuator;

U ' _m and U' _M respectively the minimum and maximum limits of the operating range of the i ^th actuator; and

- The input (control) of the i ^th actuator.

If T _{3 is} the sampling period, then the discrete model deduced from the model (1) is given by:

T ₁ (k + I) - (I - ^) T ₁ (k) + - - U ₁ (k)

IT ₂ (k + V) = (1 - ¹ ^) T ₂ (k) + ^ u ₂ (k) [^ ₂ τ ₂

The exit that we want to enslave is

All (k) = T ₁ (k) + T ₂ (k) (3) We define the quadratic criterion to be minimized as follows

J _N = Σ JM ₁ ² (k) + + q [T _m (U) -T ₁ (U) -T ₂ (h)] ² } (4)

where q is a weighting parameter in order to penalize one term of the criterion with respect to the other.

The problem can be rewritten in a more compact form: min U ^T RU + X ^T QX (5)

where R> 0 and Q> 0 are square matrices of adequate order. Likewise for the matrices A and B which can be deduced from the constraints on the inputs and on the torques supplied at the output of the actuators 1 and 2. This formulation also comprises the vectors:

U = [U 1 (O), U ₂ (O), ..., U 1 (N-1), U ₂ (N-1)] and

X = [T ₁ (O), T ₂ (O), ..., T ₁ (N - 1), T ₂ (N - 1)]

According to equation (2), formula (5) can be rewritten

where H, F, C and D are appropriate size matrices derived from matrices A, B, R and Q and equation (2) with the change of variable:

Z = U + H ^"1 F x (0) 5 (b)

This last problem (5a) can be rewritten as a quadratic multiparametric programming problem in the following form:

GZ ≤ SX (0) + W ^ZTHZ (6)

(See in particular "Alberto Bemporad, Manfred Morari, Vivek Dua, and Efstratios N. Pistikopoulos, The Explicit Linear Quadratic Regulator for Constrained Systems, Automatica, 38: 3-20, 2002"). The resolution of the latter problem makes it possible to generate a map of the permissible operating range of the two actuators considered.

The torque requests Ui and U2 are then calculated as being an affine function of the outputs of the two actuators and of the global torque request T _in (see FIG. 2):

<m i = l ... Nr (7)

The generated map is stored in the calculator. Depending on the torque measurements returned by sensors and the torque requested by the driver, the computer gives the instructions for each of the two actuators.

FIG. 10 shows the details of the sequential sequence of operations making it possible to obtain the requested torque on board the vehicle by the driver.

In step 10, the control member receives a request for torque expressed by the driver and transmitted to the member via one or more sensors for example. This is the size Tin. This value must be taken into account in the next step 12. Also taken into account are values Ti and T ₂ corresponding to the output torque of the two actuators. These are the last values in memory or reference values used to start the iteration.

In step 12, the controller searches the map stored in memory two matrices Mj and rrii verifying the second part of equation 7 recalled in box 12 and corresponding to the same integer i. This identification is made using the three aforementioned torque values as input values.

In step 14, the controller then determines the two matrices Lj and Ij corresponding to the integer i. Then he calculates the values of reference Ui and U ₂ using the first part of equation 7 recalled in box 14 using again the values Ti, T ₂ and T _in .

Then, at the next step 16, the torque thus determined instructions are applied to the two actuators Ui and U ₂

In the next step 18, the output torques T ₁ , T ₂ of these two actuators are effectively measured and, thanks to the feedback loop 20, are reused with the new value T _n which corresponds to their sum, to achieve the same values. operations and constitute a bondage.

The output torques of the actuators can be obtained alternately by means of torque estimators.

Simulations of the operation of the invention are illustrated in Figures 3 to 6.

FIGS. (3) and (5) illustrate simulations carried out with the model described by equation (2), namely the torque that it is desired to provide and the torque actually delivered by the two actuators (FIG. sum of the two pairs Ti and T ₂ ).

Figures (4) and (6) show how the distribution of the torque is made between the different actuators.

The figure (4) where the demand is in steps shows that, for a positive torque demand of 50 Nm, the mapping expresses a torque request to the motor of 150 Nm (maximum torque) so that the torque supplied by the motor goes up. as quickly as possible. Once the latter reaches the value of 50Nm, the engine torque demand returns to 50Nm. Meanwhile, no torque demand is expressed for the brake. It is the same when the requested pair is negative.

In Figure (6), the torque demand is a ramp. When the positive torque is requested, the computer systematically solicits the motor, but this time the torque request is not too important compared to the global request. By cons and this is particularly interesting, when we ask for a decrease in the total torque and that the decline is achievable by the engine, the calculator continues to solicit. Yes this drop becomes too important, so the computer also seeks negative torque from the brake.

The application of this method to control two actuators considers the problem as a whole. The control law is calculated based on a model encompassing the dynamics of one and the other of the two actuators with their respective saturations. The generated map is the exact solution of a constrained optimization problem. The problem of choosing the thresholds to switch from one actuator to another no longer arises. The problem of stability of the loop is also solved by the optimization method.

This method has the following advantages:

the problem of calculating the thresholds for switching between the actuators is solved mathematically,

the servo law of each actuator is integrated in the cartography,

each of the two actuators works while having knowledge of the state of the other actuator,

- the approach can be applied in the case of several actuators with overlapping operating ranges.

The invention comprises the following elements:

a measurement or torque estimation system provided by the actuators;

a mapping computed with an optimization algorithm under constraints (Multi-Parametric Programming); and

a calculator in which the mapping is memorized and which calculates the requests to be sent to each actuator.

We have detailed how, when we have several actuators, divide a request for torque between them. The solution to the problem is obtained by solving an optimization problem under constraints. The approach has very good results but, and it is quite natural, it has some disadvantages. In particular, if the actuators have a dynamics of order greater than 1, the mapping will depend on the entire state of the system. Either we have the measure of the whole state of the system, or we must synthesize an observer that allows the reconstruction of the state of the system.

The size of the map can become very large if one tries to increase the N prediction horizon when solving the optimization problem given in equation (4). This has the consequence of increasing the calculation time.

Finally, this approach can only be applied to actuators that have linear dynamics and whose saturation remains linear in pieces. It should be noted that we often approach the dynamics of an actuator with a linear dynamics.

Of course, we can bring to the invention many changes without departing from the scope thereof.

The invention also applies to other actuators than the motor and the brake.

Claims

1. A method for controlling a plurality of actuators (1, 2) of a vehicle capable of responding to the same request, at least one of the actuators having a bandwidth and / or a saturation, characterized in that it determines for at least one of the actuators (1, 2) a setpoint (ui, U2) taking into account an output quantity (Ti ₁ T ₂ ) of at least one other of the actuators or of the other actuator, so that the actuators or at least some of them act jointly.

2. Method according to the preceding claim, characterized in that for each actuator is determined an instruction taking into account an output of at least one other actuators or the other actuator.

3. Method according to any one of the preceding claims, characterized in that for determining at least one of the actuators a setpoint taking into account an output of each of the other actuators or the other actuator.

4. Method according to any one of the preceding claims, characterized in that for each actuator is determined a setpoint taking into account the output of each of the other actuators or the other actuator.

5. Method according to any one of the preceding claims, characterized in that it implements the determination by consulting a map.

6. Method according to the preceding claim, characterized in that at least one of the following data is used as input data of the map:

the output quantity (Ti ₁ T ₂ ) of at least one of the actuators; and

the sum (T _out ) of the output quantities of the actuators.

7. Method according to any one of claims 5 or 6, characterized in that determines an integer i such that:

Mi [Ti T ₂ ... T _in ] ≤ ITIi

Where: Mi and mi are predetermined matrices associated with i; T _n is the output quantity of the actuator n; and T _in is a quantity corresponding to the demand.

8. Method according to any one of claims 5 to 7, characterized in that the mapping is generated by means of a constrained optimization algorithm.

9. Method according to any one of claims 5 to 8, characterized in that generates the mapping by quadratic multiparametric programming.

10. Method according to any one of the preceding claims, characterized in that implements the determination by means of a calculation.

11. Method according to any one of the preceding claims, characterized in that calculates:

U ₁ (Ic)

TJk) u ₂ (k) = L, + 1

^T J ^k \ where: U _n (k) is the setpoint associated with the actuator n with k sampling parameter;

Lj and Ij are matrices given by mapping.

T _n is the output quantity of the actuator n; and

T _in is the magnitude corresponding to the demand.

12. Process according to any one of the preceding claims, characterized in that the or each output quantity (Ti ₁ T ₂ ) is determined and the determination of the or each instruction is repeated, taking into account the or each determined quantity. .

13. Vehicle comprising:

actuators (1, 2) capable of responding to the same request, at least one of the actuators having a bandwidth and / or saturation; and

- A control member (4), characterized in that the control member is arranged to determine for at least one of the actuators a setpoint (ui, u ₂ ) taking into account an output quantity of at least another of the actuators or the other actuator so that the actuators or at least some of them act together.