CN-121995968-A - Lateral movement safety control method and system for automatic driving vehicle

CN121995968ACN 121995968 ACN121995968 ACN 121995968ACN-121995968-A

Abstract

The invention discloses a lateral movement safety control method and a lateral movement safety control system for an automatic driving vehicle. The method comprises the steps of obtaining course errors and transverse errors between a controlled vehicle system and a reference path, calculating projection errors, inputting the projection errors into an input channel of a dual-channel event trigger mechanism for sparsification updating, inputting a preset time fuzzy state observer to estimate unmeasurable dynamic data, designing a self-adaptive safety controller based on the updated projection errors and estimated values to obtain control input through a backstepping method, reconstructing a safety control signal by combining a spoofing attack model, sparsifying again through an output channel of the dual-channel event trigger mechanism, and finally realizing accurate and safe control on transverse movement of the vehicle by utilizing the updated control signal. The invention effectively saves communication and calculation resources, ensures that the system state is stable within the preset time, and enhances the security of the control system under network spoofing attack.

Inventors

LI KEWEN
ZHU YU
WU WEI
Shao xinfeng
DONG GUOWEI
LI YONGMING
CHENG SHUAI
HU JUN

Assignees

辽宁工业大学

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (9)

1. The method for controlling the lateral movement safety of the automatic driving vehicle is characterized by being applied to a lateral movement system of the vehicle, and comprises the following steps of: Step1, confirming course errors and transverse errors between a controlled vehicle system and a reference path in a vehicle transverse motion system; step 2, acquiring a projection error based on the heading error and the transverse error obtained in the step 1; step 3, inputting the projection error obtained in the step 2 into a dual-channel event trigger mechanism to obtain the projection error updated by the trigger threshold; step 4, inputting the updated projection error obtained in the step 3 into a preset time fuzzy state observer to obtain an estimated value of dynamic data; step 5, designing an adaptive safety controller for the estimated value of the dynamic data obtained in the step 4, and obtaining an adaptive control input; step 6, reconstructing the self-adaptive control input obtained in the step 5 according to the spoofing attack model to obtain a safety control signal; Step 7, inputting the safety control signal obtained in the step 6 into a dual-channel event trigger mechanism to obtain a control signal with a trigger threshold updated; and 8, controlling the target system based on the updated control signal in the step 7, and realizing the transverse movement control of the vehicle.
2. The method according to claim 1, wherein in the step 1, the vehicle lateral movement system model is constructed by: based on dynamics theory, constructing a two-degree-of-freedom model of the vehicle: wherein the variables are Indicating the slip angle of the vehicle, And Representing the front and rear wheel tire lateral forces respectively, Which is indicative of the longitudinal velocity, Indicating the yaw rate of the vehicle, Representing yaw moment of inertia; which is indicative of the mass of the vehicle, Representing the distance of the centroid to the front axis, Representing the distance of the centroid to the rear axis; based on the two-degree-of-freedom model of the vehicle, the following vehicle kinematic model is constructed: wherein the variables are Indicating a heading error and, Indicating the heading of the vehicle, The lateral velocity is indicated as such, Indicating the lateral error of the beam, The projection error is indicated as such, Representing a constant projection distance and, Representing the distance along the reference path, Representing the heading of the reference path; Based on the vehicle kinematic model, the nonlinear vehicle kinematic model is obtained as follows: wherein the variables are defined 、 And , Indicating the rotation angle of the front wheel, And Representing unknown disturbances in the front and rear axes, respectively; represents the curvature of the road surface, Indicating the slip angle of the front wheel, Representing a constant projection distance and, Representing the cornering stiffness of the front axle tyre, Represents the cornering stiffness of the rear axle tyre, Representing the adhesion coefficient; to stabilize the system, the unknown disturbance is transformed into the following form based on a nonlinear vehicle dynamics model: Wherein, the , Representing an external unknown disturbance, based on the above derivation, transforming the nonlinear vehicle dynamics model into the following form: In order to solve the problem that the vehicle dynamics model is insufficient to fully describe the response of the vehicle to the steering input, the following steering system model is built: wherein the variables are The steering resistance moment is indicated as a result, The coefficient of friction is indicated as such, Representing steering motor torque; Is the corner of the front wheel Is that The first derivative with respect to time is, Is that The second derivative with respect to time is, Indicating the transmission ratio of the steering system, Representing the transmission ratio of the steering motor; the steering system model is rewritten as: Wherein, the Indicating the damping of the steering system, Representing an equivalent inertia of the steering system; in the course of the actual running process, And Difficult to measure, thus, will transform the system model The rewriting is as follows: Wherein, the , Representing an external unknown disturbance; from the above analysis, the rewritten steering system model is changed to the following form: Definition of variables , , ; The lateral movement of the vehicle is determined by the input of the steering system and therefore a strict feedback system is used, which in combination with the steering system and the vehicle dynamics model is: Definition of intermediate variables 、、 And , Is a constant which is known as such, Is that The first derivative with respect to time is, Representing the system output with a spoofing attack, Is the steering angle speed of the front wheels.
3. The method of claim 1, wherein in the step 3, a dual-channel event trigger mechanism is designed based on a dual-channel event trigger control theory, and the dual-channel event trigger mechanism is as follows: Wherein, the Expressed in time The updated input signal is provided to the user, The time is represented by the time period of the day, The time of the update is indicated and, Representing a design parameter greater than 0, Is shown in The system output of the time of day update, Representing the system output updated via the event trigger mechanism.
4. The method according to claim 1, wherein in the step 4, the preset time blur state observer comprises: based on a vehicle transverse motion system, performing model conversion to obtain a dynamics model, wherein the dynamics model comprises unknown dynamics parameters, and a characterization formula of the dynamics model comprises: defining continuous functions And , , , , Is an unknown parameter, and thus a nonlinear function And Is totally unknown; Based on the universal approximation principle of the fuzzy logic system, solving unknown kinetic parameters to obtain a predetermined time fuzzy state observer, wherein a characterization formula of the predetermined time fuzzy state observer comprises the following steps: wherein the variables are Representation of Is a function of the estimated value of (2); is a function of a pre-set time scale, , Is an integer number of times which is the whole number, Is a positive constant with respect to the preset time, whereby it can be seen that, Is monotonically increasing and ; 、、、 Indicating that the observed gain is to be achieved, Representing system output Is used for the estimation of the (c), Representing parameters Is used for the estimation of the (c), Representing a fuzzy basis function; the characterization formula of the kinetic parameters comprises: Wherein, the Is an ideal weight vector for the vehicle, Is an approximation error and satisfies , Is a positive constant.
5. The method according to claim 1, wherein the step 5 comprises: Step 51, confirming and acquiring data of the preset time fuzzy state observer; Step 52, obtaining a virtual control input based on the projection error, the observer data and a preset time stability theory; step 53 of obtaining an adaptation law of the target system based on the adaptive virtual control input and the observer data; and step 54, acquiring an adaptive control input based on the virtual control input and the adaptive law.
6. The method according to claim 1, wherein the step 6 comprises: step 61, obtaining an error vector based on the projection error of the system, observer data and a preset time stability theory; step 62, constructing a lyapunov function based on the error vector; Step 63, obtaining an adaptive virtual control input based on the error vector and the lyapunov function.
7. The method according to claim 1, wherein in step 7, the safety controller is designed by using a back-stepping method based on the adaptive control input to obtain the system control signal, and the following coordinate transformation is defined: Wherein, the In the form of a virtual controller, the controller is configured to, Is a desired input to the system and, ; The characterization formula of the system control signal comprises: Wherein, the Representing a design parameter greater than 0, Representing virtual controllers Derivative with respect to time; The following fraud attack model is constructed: Wherein, the , In order to take advantage of the attack signal, Is an additive attack signal; based on the spoofing attack model, a system security control signal is obtained, and a characterization formula of the system security control signal comprises: 。
8. the method according to claim 1, wherein said step 8 comprises: Step 81, inputting a safety control signal into a dual-channel event trigger mechanism to obtain a control signal after the trigger threshold is updated; Step 82, designing a dual-channel event trigger mechanism based on a dual-channel event trigger control theory, wherein the dual-channel event trigger mechanism comprises: Wherein, the Representing the control signal updated by the event trigger mechanism, Is shown in A control signal for the time instant update, Representing a design parameter greater than 0; Based on the designed dual-channel event trigger mechanism and the system safety control signal, a control signal which meets the trigger threshold value and is updated is obtained, and the updated control signal comprises: 。
9. The lateral motion safety control system of the automatic driving vehicle is characterized by being applied to a vehicle lateral motion system, and comprises the following components: a heading error and lateral error acquisition module (10) for confirming a heading error and a lateral error between a controlled vehicle system and a reference path in a vehicle lateral motion system; a projection error acquisition module (20) for acquiring a projection error based on a heading error and a lateral error between the controlled vehicle system and the reference path; the system comprises a preset time fuzzy state observer module (30) and a dynamic data analysis module, wherein the preset time fuzzy state observer module is used for inputting projection errors after system updating into the preset time fuzzy observer to obtain estimated values of dynamic data; a backstepping technique module (40) for designing an adaptive safety controller for the projection error and the estimated value of the dynamics data after the system update to obtain an adaptive control input; The spoofing attack module (50) is used for obtaining a security control signal according to the spoofing attack model and the self-adaptive control input; The dual-channel event triggering module (60) is used for inputting the projection error and the safety control signal of the system into the dual-channel event triggering mechanism so as to obtain the projection error and the control signal after the triggering threshold value is updated; And the output control module (70) is used for controlling the target system based on the updated control signal to realize the transverse movement control of the vehicle.

Description

Lateral movement safety control method and system for automatic driving vehicle Technical Field The invention belongs to the field of automatic driving vehicle lateral movement control, and particularly relates to a lateral movement safety control method and system of an automatic driving vehicle. Background With the increasing popularization of automobiles, the automobile industry is accelerated to change and upgrade to electric, intelligent and networking directions, and the position of an automatic driving automobile in a modern traffic system is more and more critical, so that the daily life of people is greatly facilitated. At the same time, the falling of the smart city concept gradually pushes the autopilot car to be the focus of social attention. It is worth noting that the improvement of the intelligent degree of the vehicle puts forward a strict requirement on the communication efficiency among the sensor, the actuator and the vehicle controller, and the superposition influence of the factors such as uncertainty of the vehicle model parameters, unknown external interference, malicious network attack and the like put forward a strict requirement on the performance of the automatic driving automobile control system. Therefore, a plurality of scholars at home and abroad have developed extensive and intensive research around the technology of controlling the lateral movement of the automatic driving automobile, and a series of important research results are achieved. In the field of transverse motion control, the transverse motion control is used as a core technology for improving the intelligence and automation level of a vehicle, and active steering is realized by regulating and controlling a steering power-assisted motor of a steering system, so that a high-precision path tracking task is completed. Currently, although many control techniques are available for autonomous vehicle lateral motion control systems. However, the following problems still remain in the prior art: (1) In the control method of the automatic driving vehicle lateral movement control system, although the existing technology can complete the high-precision path tracking task, the problems of consumption and utilization of communication resources in the tracking process are not considered in most cases. The dual-channel event trigger control can effectively reduce the consumption of communication resources and greatly improve the utilization rate of the communication resources. (2) Among existing methods of controlling lateral motion of an autonomous vehicle, most are based on ideal assumptions that the state is fully measurable. However, in an actual vehicle control scenario, the key state inside the system is often difficult to directly obtain, and a large number of sensors are required to be installed. Meanwhile, the existing control method has few consideration on achieving an ideal control effect in a preset time. (3) In the existing automatic driving vehicle lateral movement control method, influences caused by fraud attacks in the vehicle communication process are not considered in most cases. From a practical point of view, the vehicles communicate via a network, which means that the vehicles are susceptible to fraud attacks during the communication. Disclosure of Invention The invention aims to provide a transverse movement safety control method and system for an automatic driving vehicle, which not only can ensure that a vehicle system is stable within a preset time, but also can effectively reduce the consumption of communication resources and greatly improve the utilization rate of the communication resources. In order to achieve the above purpose, the invention adopts the following technical scheme: a lateral motion safety control method of an autonomous vehicle, applied to a lateral motion system of the vehicle, comprising the steps of: Step1, confirming course errors and transverse errors between a controlled vehicle system and a reference path in a vehicle transverse motion system; step 2, acquiring a projection error based on the heading error and the transverse error obtained in the step 1; step 3, inputting the projection error obtained in the step 2 into a dual-channel event trigger mechanism to obtain the projection error updated by the trigger threshold; step 4, inputting the updated projection error obtained in the step 3 into a preset time fuzzy state observer to obtain an estimated value of dynamic data; step 5, designing an adaptive safety controller for the estimated value of the dynamic data obtained in the step 4, and obtaining an adaptive control input; step 6, reconstructing the self-adaptive control input obtained in the step 5 according to the spoofing attack model to obtain a safety control signal; Step 7, inputting the safety control signal obtained in the step 6 into a dual-channel event trigger mechanism to obtain a control signal with a trigger threshold updated; and 8, controlling the target