DEVICE FOR PREDICTIVELY CONTROLLING THE MOVEMENT OF A MOTOR VEHICLE

Abstract

A device for controlling the movement of a motor vehicle, including a longitudinal controller and a lateral controller which are capable of generating, from first information relating to the road layout and second information relating to the dynamic behaviour of the vehicle, control commands intended for actuators for controlling the longitudinal and lateral movement of the vehicle. The device includes a prediction model which is supplied with the first and second information and is capable of determining future states of the vehicle for future positions of the vehicle over a plurality of iterations defining a future road portion. The model is connected to a module for determining whether driving limit values are violated, which module is capable of determining, for each future state, whether one of the state variables defining the future state reaches or exceeds a driving limit value, and of deducing a future risk situation.

Claims

1-8. (canceled)

9. A device for controlling the movement of a motor vehicle on a road, comprising: a module configured to manage least one driving assistance system; an onboard device configured to detect the ground markings for traffic lanes, configured to provide first information relating to the road layout according to the detected markings; a set of onboard measurement sensors configured to provide second information relating to the dynamic behavior of the vehicle; a device configured to control the longitudinal movement of the vehicle and a device configured to control the lateral movement of the vehicle, which are configured to generate, from the first and second information, control commands for the longitudinal movement and for the lateral movement sent to actuators for longitudinal and lateral control of the steering system of the vehicle; and a model configured to predict the dynamic behavior of the vehicle supplied with the first and second information, which is configured to determine a set of future states of the vehicle corresponding to future positions of the vehicle over a plurality of iterations defining an upcoming road portion, said prediction model being connected to a module configured to determine the violation of vehicle driving limit values, which is configured to determine, for each determined future state, when at least one of the state variables defining said future state of the vehicle reaches or exceeds a vehicle driving limit value, and configured to infer therefrom an upcoming risk situation for the vehicle for at least one future state of the vehicle.

10. The device as claimed in claim 9, wherein the driving assistance system is an electronic stability program.

11. The device as claimed in claim 9, wherein the model configured to predict the dynamic behavior of the vehicle receives as input predicted control commands for the actuators of the vehicle delivered by a regulator equivalent to the devices to control the longitudinal movement and the lateral movement of the vehicle, state variables of the future state of the vehicle determined during a current iteration, and its current states, which are the lateral position of the vehicle, the lateral speed of the vehicle, the yaw angle and the yaw rate of the vehicle.

12. The device as claimed in claim 9, wherein the module configured to determine the violation of the vehicle driving limit values includes, among the parameters defining a vehicle driving limit value, physical parameters relating to said actuators and/or to the dynamics of the vehicle.

13. The device as claimed in claim 12, wherein said parameters comprise at least a maximum steering angle value, a maximum steering wheel angle speed value, a value representative of a maximum braking capability, and a value representative of a maximum acceleration capability of the vehicle.

14. The device as claimed in claim 12, wherein the module configured to determine the violation of the vehicle driving limits is configured to compare the determined future states of the vehicle with respect to the vehicle driving limit values and to generate an alert signal as soon as at least one of the state variables defining the future states reaches the corresponding driving limit value.

15. The device as claimed in claim 14, wherein the module configured to manage the driving assistance system is configured to receive said generated alert signal and, upon receiving said alert signal, to control an anticipatory corrective action for said state variable.

16. The device as claimed in claim 15, wherein said anticipatory corrective action comprises a braking and/or steering action.

17. A motor vehicle, comprising: the device as claimed in claim 9.

Description

[0020] Other features and advantages of the present invention will become more clearly apparent from reading the following description, given by way of illustrative and non-limiting example and with reference to the appended drawings:

[0021] FIG. 1 is a diagram illustrating an architecture of the control device of the invention;

[0022] FIG. 2 is a graph illustrating an example of a trajectory of the vehicle on which the control device of the invention is evaluated;

[0023] FIG. 3A is a graph of the trajectory portion formed by the pronounced curve illustrated in FIG. 2, illustrating the change in the actual and predicted states of the vehicle on this trajectory portion;

[0024] FIG. 3B is a detail view of the entry into the curve of FIG. 3;

[0025] FIG. 4 is a graph illustrating the variations over time in the lateral error of the actual and predicted states of the vehicle with respect to the trajectory to be followed, throughout the course of FIG. 2;

[0026] FIG. 5 is a graph illustrating the variations over time in the actual steering angle of the steering wheel and in the predicted angle, with respect to a maximum value of the steering angle of the steering wheel;

[0027] FIG. 6 is a graph illustrating the corresponding output of the module for determining violation of vehicle driving limit values.

[0028] With reference to FIG. 1, the device 1 for controlling the movement of the vehicle comprises an onboard device 10 for detecting traffic lane ground markings. This is typically an onboard camera oriented forward with respect to the motor vehicle, for example installed at the top of the windshield, at the junction with the vehicle's roof, or installed behind the vehicle's interior rearview mirror so as to be able to acquire images of the scene ahead of the vehicle and to be able to detect the road markings ahead of the motor vehicle. To do this, the camera is associated with a processing unit, which takes the image stream provided by the camera and analyzes it using image processing techniques, to detect in particular the road marking lines on the ground along the traffic lane followed by the vehicle. The processing unit is also designed to provide information relating to the road layout based on the detected markings, in particular the number of lanes, the width of the shoulders, the lateral deviation in the direction of the middle lane, and a curvature profile of the trajectory.

[0029] The vehicle is also equipped with a set of onboard sensors 11, arranged on various items of vehicle equipment (steering wheel, steering, brakes, etc.) which are able to provide information relating to the dynamic behavior of the vehicle, in particular information such as speed, heading angle, acceleration, yaw rate of the vehicle, etc.

[0030] The information relating to road layout and the information relating to the dynamic behavior of the vehicle are delivered to a device 12 for controlling the longitudinal movement of the vehicle, or longitudinal controller, and to a device 13 for controlling the lateral movement of a vehicle, or lateral controller, which are capable of generating, from this information, at each iteration point, commands for controlling longitudinal movement and lateral movement to actuators 14, in particular at least one vehicle longitudinal control actuator and at least one steering actuator of the vehicle steering system, so as to make it possible to steer the vehicle in autonomous mode, that is to say without actions from the driver, for example by applying a control strategy for following the middle lane. Thus, for the longitudinal positioning of the vehicle, the device 12 for controlling the longitudinal movement of the vehicle is used to control vehicle braking and acceleration actuators. For the lateral positioning of the vehicle, the device 13 for controlling the lateral movement of the vehicle is used to control actuators for controlling the steering angle of the wheels.

[0031] These two control devices 12 and 13 implement regulators that use the following input/output variables: current vehicle speed, current vehicle acceleration, desired vehicle speed at the given iteration point, desired vehicle acceleration at the given iteration point, current yaw rate, desired yaw rate at the given iteration point, current vehicle heading error, current vehicle lateral error.

[0032] The outputs of the control devices 12 and 13 are then used in a module 15 for managing the driving assistance systems of the vehicle, in particular an ESP system, with which the vehicle is equipped. These systems are therefore feedback-controlled just using information from the sensors, which ensures only reactive-type assistance. There is currently no system capable of indicating the vehicle driving limits for the upcoming road portion. In other words, these systems are blind and are unable to anticipate future states of the vehicle that are liable to cause situations in which vehicle driving limits are violated, such as emergency braking or excess speed in turns.

[0033] Thus, provision is made to supply a model 16 for predicting the dynamic behavior of the vehicle with, on the one hand, information relating to the dynamic behavior of the vehicle and, on the other hand, information relating to the road layout, so that the model 16 can use this information to predict future states of the vehicle.

[0034] This prediction model 16 is therefore sensitive to vehicle state signals comprising the vehicle lateral speed, the vehicle longitudinal speed, the vehicle yaw rate and the steering angle, and to the information relating to the road layout and more particularly the curvature of the trajectory, in order to make it possible to predict, as model output, a set of vehicle positions corresponding to future states of the vehicle, up to the electronic horizon. The electronic horizon is formed from a set of information relating to the road environment into which the vehicle is liable to move in the near future, i.e. in practice as far as the maximum visibility permitted by the onboard detection device.

[0035] Each of these future positions is associated with a vehicle steering, accelerator, and braking command, which makes it possible to assess whether one of these future positions is outside of vehicle driving limits.

[0036] The vehicle prediction model is determined as follows:

|X.sub.v=A.sub.v+B.sub.vu.sub.v

Y.sub.v=C.sub.vX.sub.v

[0037] where u.sub.v is the steering wheel angle command and X.sub.v is the state vector, defined as follows:

X.sub.v=[y.sub.vv.sub.yψ.sub.vω.sub.v]

[0038] where y.sub.v is the vehicle lateral position, v.sub.y is the vehicle lateral speed, ψ.sub.v is the vehicle yaw angle, and ω.sub.v is the vehicle yaw rate.

[0039] The matrices A, B and C are described below:

[00001]$A_v=\left[\begin{array}{cccc}0& 1& 0& 0\\ 0& \frac{-(C_f+C_r}{{\mathrm{mv}}_x}& 0& \frac{-{\mathrm{aC}}_f+{\mathrm{bC}}_r}{{\mathrm{mv}}_x}-v_x\\ 0& 0& 0& 1\\ 0& \frac{-{\mathrm{aC}}_f+{\mathrm{bC}}_r}{I_z{v}_x}& 0& \frac{-a^2{C}_f-b^2{C}_r}{I_z{v}_x}\end{array}\right]$$B_v={\left[0\frac{C_f}{m}0\frac{{\mathrm{aC}}_f}{I_z}\right]}^T$$C_v=\left[0001\right]$

[0040] where C.sub.f and C.sub.r correspond to the cornering stiffness at the front and rear wheels, respectively, v.sub.x is the speed of the vehicle, m is the mass of the vehicle, I.sub.z is the moment of inertia about the vertical axis Z, and a and b are the distances from the center of gravity of the vehicle to the front and rear wheels, respectively.

[0041] This vehicle-specific dynamic model can be connected to a descriptive model of tire behavior developed on the basis of the model proposed by PACEJKA or the like, to include the stability limits.

[0042] The output of the model 16 is connected to a module 17 for determining the future states of the vehicle in which the future positions of the vehicle are calculated. More specifically, this module 17 uses the current speed of the vehicle, the position and orientation of the vehicle, the wheelbase and the angle of the steering wheel to calculate the first iteration. Next, for the future states, the prediction model 16 is run forward, taking into account the closed-loop response of a prediction regulator 18 of the vehicle, which provides the same response as the regulators implemented in the devices 12 and 13 for controlling the longitudinal and lateral movement of the vehicle. Thus, the future positions of the vehicle are estimated using the same control strategy as that which will be applied later on by the devices for controlling the longitudinal and lateral movement of the vehicle to said positions.

[0043] Thus, the input values are the predicted control commands provided by the regulator 18, collected at the beginning of the current iteration, and the state values are the components of the updated state vector X.sub.v characterizing the state of the system at the previous iteration.

[0044] The output of the module 17 for determining the future states of the vehicle is thus provided first to the prediction regulator 18 of the vehicle which, as indicated above, is equivalent to the devices 12 and 13 for controlling the longitudinal and lateral movement of the vehicle. The main difference is that the output of the prediction regulator is also used to supply the prediction model 16, so as to allow the future states of the vehicle across the entire electronic horizon to be covered by the prediction.

[0045] The output of the module 17 for determining the future states of the vehicle is also delivered to a module 19 for determining the violation of vehicle driving limit values. This module 19 includes all of the physical limits from either the actuators of the vehicle or the dynamics of the vehicle itself. It takes into account the maximum steering wheel angle, the maximum steering wheel angle speed, the maximum braking capability and the maximum acceleration capability. It monitors all of the future states of the vehicle with respect to these maximum values and is designed to raise an indicator flag when one of these future states exceeds the limits of the vehicle.

[0046] The module 19 for determining the violation of vehicle driving limits is connected to the module 15 for managing the driving assistance systems of the vehicle. The latter thus receives the current outputs of the control devices 12 and 13 and the outputs of the prediction regulator with the violation indicators. It makes it possible to modify the dynamic behavior of the vehicle by reacting in advance according to the future states for which an indicator of violation of vehicle driving limits has been raised.

[0047] The following example illustrates the principles that have just been described. FIG. 2 illustrates, on X and Y coordinates, an example of a specific trajectory in which a pronounced curve C is taken. The whole trajectory is followed by a vehicle equipped with a device 1 for controlling the movement of the vehicle as described above with reference to FIG. 1. FIGS. 3A and 3B more precisely show the exact moment when the prediction model of the vehicle detects actuator control signals or future states of the vehicle that are outside of a controllable range, i.e. that reach or exceed a driving limit value of the vehicle. FIG. 3A shows the trajectory portion at the pronounced curve C of FIG. 2, while FIG. 3B is a detail view of the entry into the pronounced curve C. FIG. 3A shows the trajectory T to be followed by the vehicle, the actual states of the vehicle st_real as determined by the onboard sensors, and the future or predicted states of the vehicle st_predicted provided by the prediction model 16. The vehicle will therefore follow the trajectory defined by the set of actual states of the vehicle st_real, but the prediction model is capable of detecting in advance, for example 800 ms before, future states of the vehicle that are outside of the controllable range, which are represented by the end of the line st_predicted.

[0048] FIG. 4 shows the lateral error in meters of the vehicle with respect to the trajectory to be followed, throughout the course of FIG. 2. The curve ε_real represents the lateral error of the actual states of the vehicle with respect to the trajectory to be followed and the curve ε_predicted represents the lateral error of the predicted states of the vehicle 800 ms in advance with respect to the trajectory to be followed. Around the twelfth second, lateral errors of more than 1 meter appear, meaning that the vehicle departs from the controllable range.

[0049] FIG. 5 shows the behavior of the steering wheel steering angle α_real resulting from the actual state of the vehicle and from the predicted angle α_predicted, as a function of time, as well as the maximum value α_max of the steering wheel steering angle provided by the electric power steering system, set to about 65°. In this example, by virtue of the prediction, the vehicle can detect the steering wheel steering angle states that reach the maximum controllable value before they occur and make decisions accordingly. FIG. 5 illustrates the advance in time of the predicted states of the vehicle which, later on, will correspond to the actual states of the vehicle. Thus, the prediction model makes it possible to determine in advance the future states of the vehicle that correspond to potentially dangerous situations in which the steering angle reaches the maximum value.

[0050] FIG. 6 illustrates the corresponding output of the module for determining violation of vehicle driving limits. Thus, the raising of the violation indicator ld_predicted shows when the module for determining violation of vehicle driving limits detects that the future state of the vehicle with respect to the steering wheel steering angle violates the maximum allowed limit value. Such an alert may be issued for one or more predicted state variables of the vehicle outside of its controllable range. In the example presented here, the raising of the indicator ld_predicted makes it possible to inform the module 15 for managing the driving assistance systems acting on the future change in the steering angle about 0.8 second before the actual state occurs. The module 15 for managing the driving assistance systems of the vehicle is then able to act and prevent this situation in advance, by controlling an anticipatory corrective action with the aim of preventing the steering angle from violating the maximum allowable limit value. Such an anticipatory corrective action may, for example, be a slight braking and/or steering action. Thus, the invention makes it possible to increase the safety of the vehicle by anticipating future states of the vehicle in which its stability limits might be exceeded, and adopting appropriate safety strategies in advance.