Anti-disturbance event trigger control method for autonomous vehicle

Abstract

The invention discloses an automatic driving vehicle anti-disturbance event trigger control method which comprises the following steps: a, establishing a mathematical model of a nonlinear man-machine cooperation automatic driving vehicle; b, designing a disturbance observer and an adaptive estimator; c, defining error conversion of the man-machine collaborative automatic driving vehicle system; d, solving the derivative of the error; e, constructing a virtual controller; and f, constructing a dynamic event triggering controller. The controller designed by the invention not only can ensure the tracking capability of the vehicle and resist the influence of external disturbance, but also can reduce the communication burden and save communication resources.

Description

Technical Field

[0001] The present invention relates to the field of event triggering control of autonomous driving vehicles, and in particular to an anti-disturbance event triggering control method for autonomous driving vehicles. Background Art

[0002] Cars have become an important part of people's lives and economic development. As the number of cars in use grows rapidly, the intelligent control algorithms of cars are bound to become a research hotspot in the current control field.

[0003] In recent years, autonomous driving technology has also achieved significant development, and it has increasingly shown the potential to replace drivers. However, the real traffic conditions are complex, and intelligent algorithms can never surpass the driving level of humans. Therefore, the control algorithm of human-vehicle collaborative autonomous driving technology has important research significance. During the driving process, due to the influence of factors such as the driver's misoperation and the road environment, the car is inevitably subject to unknown disturbances, which poses a challenge to the anti-interference ability of human-machine collaborative autonomous driving technology. At the same time, since autonomous driving technology needs to transmit integrated information to each actuator, this undoubtedly increases the communication burden and transmission cost between various components of the car. Therefore, it is crucial to design an autonomous driving vehicle control method that has anti-interference ability and effectively reduces the communication burden. Summary of the invention

[0004] The purpose of the present invention is to provide an anti-disturbance event triggering control method for an autonomous driving vehicle. Based on disturbance observer technology and adaptive technology, the autonomous driving vehicle has anti-disturbance capability. Through the dynamic event triggering mechanism, the communication burden of the controller is alleviated, ensuring that the autonomous driving vehicle can achieve ideal control performance at a lower cost.

[0005] The invention includes the following steps: a. Establishing a mathematical model of a nonlinear human-machine cooperative autonomous driving vehicle; b. Designing a disturbance observer and an adaptive estimator; c. Defining the error conversion of the human-machine cooperative autonomous driving vehicle system; d. Solving the derivative of the error; e. Constructing a virtual controller; f. Constructing a dynamic event triggering controller. The invented controller can not only ensure the tracking capability of the vehicle and resist the influence of external disturbances, but also reduce the communication burden and save communication information.

[0006] Furthermore, in step (a), the two-degree-of-freedom dynamic model of the vehicle is described as follows:

[0007]

[0008] Where β and γ represent the vehicle's center of mass sideslip angle and yaw rate, respectively. and are the lateral forces on the front and rear axles of the vehicle, m represents the mass of the vehicle, V x is the longitudinal velocity of the vehicle, a and b represent the distance from the center of mass of the vehicle to the front axle and the rear axle respectively, I z is the vehicle's yaw moment of inertia. The vehicle will be subject to unknown disturbances from the outside world during driving, so the lateral force on the front and rear axles can be expressed as,

[0009]

[0010] Where μ is the tire road adhesion coefficient, C f , C r are the lateral stiffness of the front and rear axles, α f , α r are the tire side slip angles of the front and rear axles, d f d r are the unknown lateral disturbances of the front and rear axles respectively, V is the vehicle speed, and δ is the front wheel steering angle.

[0011] The vehicle dynamics model is combined with the path, and the lateral displacement deviation error e at the center of mass of the vehicle is combined l and heading error The path tracking error e is obtained,

[0012]

[0013] Among them, x p is the preview distance, V y is the lateral velocity of the vehicle, s is the distance along the reference path, is the vehicle heading angle, is the path reference heading angle.

[0014] The vehicle path model is further transformed into,

[0015]

[0016] in

[0017] The steering system model of the car is:

[0018]

[0019] Among them I eq and are the equivalent moment of inertia and equivalent damping coefficient of the steering system, and are the steering resistance torque on the front wheels and the driver input torque, is the equivalent friction of the steering system, and are the reduction ratios of the steering motor and the steering system, is the steering motor torque. Both the steering load torque and the driver input torque can be obtained through sensors. However, the data obtained by the sensor is not accurate in extreme cases. The model is further converted into

[0020]

[0021] in for The estimated value of for Estimation error, for The estimated value of for The estimation error.

[0022] Finally, the steering system model is integrated into the vehicle path model, and the mathematical model of the nonlinear human-machine collaborative autonomous driving vehicle system can be expressed as:

[0023]

[0024] where x1 = e, x3=δ, Usually, time-varying disturbances are mainly caused by exogenous effects, and the energy of exogenous effects is limited, so it is usually assumed that: there are unknown constants D1, D2, D 1d The composite perturbations d1 and d2 satisfy |d1|≤D1, |d2|≤D2,

[0025] Furthermore, in step (b), the disturbance observer is designed as:

[0026]

[0027] Among them, ∈=d1-λx2 is an auxiliary variable, is the estimated value of ∈, λ>1 is the observer gain, is the estimated value of D1.

[0028] Design the adaptive disturbance estimator as,

[0029]

[0030] in is the estimate of , γ, σ, ρ are the designed parameters, and z4 is the auxiliary variable that will be designed in the following steps.

[0031] Furthermore, in step (c), the error conversion is defined as,

[0032]

[0033] where α i-1 is the input of the filter, E i is the filtering error, is the output of the filter.

[0034] Furthermore, in step (d), the derivative of the error is,

[0035]

[0036] Furthermore, in step (e), during the controller design process, a candidate Lyapunov function is selected at each step to construct a virtual controller until the real controller is constructed in the last step. The virtual controllers α1, α2, and α3 are constructed as follows:

[0037] α1=-c1z1-q2z1,

[0038]

[0039]

[0040] Among them, c1, c2, c3, q2, q3, q4, b2, b3>0 are the parameters to be designed.

[0041] Furthermore, in step (f), the dynamic event triggering mechanism is:

[0042]

[0043] in For the controller after triggering, Auxiliary functions designed for s>0, [0,1], h∈(0,1), is the design parameter, is the event trigger error. The actual controller is designed as follows:

[0044]

[0045] Among them, c4>0 is the parameter to be designed.

[0046] Compared with the prior art, the beneficial effects of the present invention are as follows: the anti-disturbance event trigger control method of an autonomous driving vehicle provided by the present invention considers how to design a dynamic event trigger controller with anti-disturbance capability for the mathematical model of a human-machine collaborative autonomous driving vehicle system, so as to ensure that the vehicle can effectively resist the influence of disturbances, stably and safely track the target trajectory, and at the same time reduce the communication burden and improve resource utilization. First, the human-machine collaborative autonomous driving vehicle is modeled to obtain a more practical system. Secondly, under the stability theory in the sense of Lyapunov, with the help of disturbance observer, adaptive technology, dynamic surface technology, and event triggering technology, a dynamic event trigger controller with anti-disturbance capability is designed for the considered system. The controller involved can not only stably track the target trajectory under disturbances, but also save communication resources and reduce the communication burden. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is the trajectory of the actual control input given by the present invention;

[0048] Figure 2 The track of the car route tracking provided by the present invention;

[0049] Figure 3 The track of the vehicle tracking error provided by the present invention;

[0050] Figure 4 is the trajectory of the disturbance observer observation value given by the present invention;

[0051] Figure 5 This is the trajectory of the disturbance adaptive estimation value given by the present invention. DETAILED DESCRIPTION

[0052] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0053] The present invention provides an anti-disturbance event triggering control method for an autonomous driving vehicle, which is specifically implemented through the following steps.

[0054] Step a. Establish a mathematical model of nonlinear human-machine collaborative autonomous driving vehicle. The two-degree-of-freedom dynamic model of the vehicle is described as follows:

[0055]

[0056] Where β and γ represent the vehicle's center of mass sideslip angle and yaw rate, respectively. and are the lateral forces on the front and rear axles of the vehicle, m represents the mass of the vehicle, V xis the longitudinal velocity of the vehicle, a and b represent the distance from the center of mass of the vehicle to the front axle and the rear axle respectively, I z is the vehicle's yaw moment of inertia. The vehicle will be subject to unknown disturbances from the outside world during driving, so the lateral force on the front and rear axles can be expressed as,

[0057]

[0058] Where μ is the tire road adhesion coefficient, C f , C r are the lateral stiffness of the front and rear axles, α f , α r are the tire side slip angles of the front and rear axles, d f d r are the unknown lateral disturbances of the front and rear axles respectively, V is the vehicle speed, and δ is the front wheel steering angle.

[0059] The vehicle dynamics model is combined with the path, and the lateral displacement deviation error e at the center of mass of the vehicle is combined l and heading error The path tracking error e is obtained,

[0060]

[0061] Among them, x p is the preview distance, V y is the lateral velocity of the vehicle, s is the distance along the reference path, is the vehicle heading angle, is the path reference heading angle.

[0062] The vehicle path model is further transformed into,

[0063]

[0064] in

[0065] The steering system model of the car is:

[0066]

[0067] Among them I eq and are the equivalent moment of inertia and equivalent damping coefficient of the steering system, and are the steering resistance torque on the front wheels and the driver input torque, is the equivalent friction of the steering system, and are the reduction ratios of the steering motor and the steering system, is the steering motor torque. Both the steering load torque and the driver input torque can be obtained through sensors. However, the data obtained by the sensor is not accurate in extreme cases. The model is further converted into

[0068]

[0069] in for The estimated value of for Estimation error, for The estimated value of for The estimation error.

[0070] Finally, the steering system model is integrated into the vehicle path model, and the mathematical model of the nonlinear human-machine collaborative autonomous driving vehicle can be expressed as:

[0071]

[0072] where x1 = e, x3=δ, Usually, time-varying disturbances are mainly caused by exogenous effects, and the energy of exogenous effects is limited, so it is usually assumed that: there are unknown constants D1, D2, D 1d The composite perturbations d1 and d2 satisfy |d1|≤D1, |d2|≤D2,

[0073] Step b. Design disturbance observer and adaptive estimator. The designed disturbance observer is:

[0074]

[0075] Among them, ∈=d1-λx2 is an auxiliary variable, is the estimated value of ∈, l>1 is the observer gain, is the estimated value of D1.

[0076] The designed adaptive disturbance estimator is,

[0077]

[0078] in is the estimate of , γ, σ, ρ are the designed parameters, and z4 is the auxiliary variable that will be designed in the following steps.

[0079] Step c. Define the error transformation of the human-machine collaborative autonomous driving vehicle system. Define the error transformation as,

[0080]

[0081] where α i-1 is the input of the filter, E i is the filtering error, is the output of the filter.

[0082] Step d. Solve for the derivative of the error. The derivative of the error is,

[0083]

[0084] Step e. Construct a virtual controller. In the controller design process, a candidate Lyapunov function is selected at each step to construct a virtual controller until the last step to construct the real controller. The virtual controllers α1, α2, and α3 are constructed as follows:

[0085] α1=-c1z1-q2z1

[0086]

[0087]

[0088] Among them, c1, c2, c3, q2, q3, q4, b2, b3>0 are the parameters to be designed.

[0089] Step f. Construct a dynamic event trigger controller. The dynamic event trigger mechanism is:

[0090]

[0091] in For the controller after triggering, Auxiliary functions designed for s>0, [0,1], h∈(0,1), is the design parameter, is the event trigger error. The actual controller is designed as follows:

[0092]

[0093] Among them, c4>0 is the parameter to be designed.

[0094] The present invention is further described below through a specific embodiment of the present invention. In this embodiment, the system parameters shown in Table 1 and the design parameters given during simulation are set.

[0095] Table 1 System parameters and design parameters

[0096]

[0097] In CarSim, the E-class sedan is selected as the experimental object, and the Federal Highway Administration Alt3 Highway provided by CarSim is used as the experimental road. The actual control input, vehicle route tracking situation, and vehicle tracking error are analyzed under the action of the designed anti-disturbance event trigger controller. 、 Disturbance Observer Observations 、 The trajectories of the perturbation adaptive estimates are Figure 1 , Figure 2 , Figure 3 , Figure 4 and Figure 5 As shown. Figure 1 It can be seen that the designed controller realizes on-demand update transmission with less transmission times, which shows that the proposed control design scheme can save communication resources and reduce the communication burden. Figure 2 and Figure 3 It can be concluded that the proposed design can ensure that the vehicle's tracking error converges to the vicinity of the origin, thereby ensuring the stability of the vehicle during driving. Figure 4 and Figure 5 It can be seen that the disturbance observer effectively observes the size of the disturbance, and the disturbance adaptive estimator also compensates for the disturbance accordingly, so that the vehicle has better anti-disturbance ability.

[0098] This method is close to the real problem and designs an anti-disturbance event trigger controller for the human-machine collaborative automatic driving vehicle system model. Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0099] (1) The existing human-machine collaborative autonomous driving vehicle system model overuses small angle approximation and equivalent infinitesimal approximation during the modeling process, so the established model is relatively simple and inaccurate. This study only uses angle approximation for the vehicle's center of mass sideslip angle, so the obtained control model is more accurate and the designed control method also has better control effect, which has important research significance.

[0100] (2) Since the vehicle will inevitably be affected by the external environment during driving, the control scheme should be designed to compensate for external disturbances to improve the robustness of control. However, the existing control schemes only use simple methods such as adaptive rate to estimate the upper limit of the disturbance. Compared with the existing control schemes, the present invention introduces a disturbance observer to accurately observe the external disturbance. The designed disturbance observer has a simple structure and low coupling, and has better readability and strong applicability.

[0101] (3) The autonomous driving vehicle system consists of multiple sensors, actuators, and controllers, which leads to a large number of communication requirements between different electronic components. When there is too much communication data, the system may find it difficult to maintain smooth communication. In addition, when the actuator performs control tasks, some unnecessary control transmissions may be generated, resulting in a waste of resources. Therefore, the present invention introduces an event trigger mechanism and designs a dynamic event trigger controller. On the basis of ensuring system performance, the controller can dynamically update control information according to demand, reduce the number of control signal generation times, thereby avoiding the transmission of redundant data, reducing the communication burden, and effectively saving communication resources.

[0102] (4) In the backstepping control design process, the computational complexity explosion problem occurs due to the requirement to derive the virtual controller, making the controller design very difficult. Therefore, the present invention introduces the dynamic surface technology and uses a first-order low-pass filter to filter the virtual control law to replace the complex differential operation, thereby simplifying the difficulty of system control design.

[0103] The technical means disclosed in the scheme of the present invention are not limited to the technical means disclosed in the above-mentioned implementation mode, but also include technical schemes composed of any combination of the above-mentioned technical features. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications are also regarded as the protection scope of the present invention.

Claims

1. A method for triggering control of an anti-disturbance event of an autonomous driving vehicle, characterized in that: The following steps are included: a. Establish a mathematical model for nonlinear human-machine collaborative autonomous driving vehicles; b. Design disturbance observer and adaptive estimator; c. Define the error transformation of the human-machine collaborative autonomous driving vehicle system; d. Solve the derivative of the error; e. Construct a virtual controller; f. Construct a dynamic event trigger controller.

2. The anti-disturbance event triggering control method for an autonomous driving vehicle according to claim 1, further characterized in that: In step (a), a two-degree-of-freedom dynamic model of the vehicle is established. Where β and γ represent the vehicle's center of mass sideslip angle and yaw rate, respectively. and are the lateral forces on the front and rear axles of the vehicle, m represents the mass of the vehicle, V x is the longitudinal velocity of the vehicle, a and b represent the distance from the center of mass of the vehicle to the front axle and the rear axle respectively, I z is the vehicle yaw moment of inertia; The vehicle will be subject to unknown disturbances from the outside world during driving, so the lateral force on the front and rear axles can be expressed as: Where μ is the tire road adhesion coefficient, C f , C r are the lateral stiffness of the front and rear axles, α f , α r are the tire side slip angles of the front and rear axles, d f ,d r are the unknown lateral disturbances of the front and rear axles, V is the vehicle speed, and δ is the front wheel steering angle; The vehicle dynamics model is combined with the path, and the lateral displacement deviation error e at the center of mass of the vehicle is combined l and heading error The path tracking error e is obtained, Among them, x p is the preview distance, V y is the lateral velocity of the vehicle, s is the distance along the reference path, is the vehicle heading angle, is the path reference heading angle; The vehicle path model is further transformed into, in The steering system model of the car is: Among them I eq and are the equivalent moment of inertia and equivalent damping coefficient of the steering system, and are the steering resistance torque on the front wheels and the driver input torque, is the equivalent friction of the steering system, and are the reduction ratios of the steering motor and the steering system, is the steering motor torque; the steering load torque and the driver input torque are both obtained through sensors; but the data obtained by the sensor is not accurate in extreme cases, so the model is further converted into in for The estimated value of for Estimation error, for The estimated value of for The estimation error of Finally, the steering system model is integrated into the vehicle path model, and the mathematical model of the nonlinear human-machine collaborative autonomous driving vehicle is expressed as: where x1 = e, x3=δ, The time-varying disturbance is mainly caused by exogenous effects, and the exogenous effects have limited energy, so it is assumed that: there are unknown constants D1, D2, D 1d The composite perturbations d1 and d2 satisfy |d1|≤D1, |d2|≤D2, 3. The anti-disturbance event triggering control method for an autonomous driving vehicle according to claim 2, further characterized in that: In step (b), the disturbance observer is designed as, Among them, ∈=d1-λx2 is an auxiliary variable, is the estimated value of ∈, λ>1 is the observer gain, is the estimated value of D1; Design the adaptive disturbance estimator as, in is the estimate of , θ, σ, ρ are the designed parameters, and z4 is the auxiliary variable that will be designed in the following steps.

4. The anti-disturbance event triggering control method for an autonomous driving vehicle according to claim 3, further characterized in that: In step (c), the error transformation is defined as, where α i-1 is the input of the filter, E i is the filtering error, is the output of the filter.

5. The anti-disturbance event triggering control method for an autonomous driving vehicle according to claim 4, further characterized in that: In step (d), the derivative of the error is, 6. The anti-disturbance event triggering control method for an autonomous driving vehicle according to claim 5, further characterized in that: In step (e), during the controller design process, a candidate Lyapunov function is selected at each step to construct a virtual controller until the real controller is constructed in the last step. The virtual controllers α1, α2, and α3 are constructed as follows: α1=-c1z1-q2z1 Among them, c1, c2, c3, q2, q3, q4, b2, b3>0 are the parameters to be designed.

7. The anti-disturbance event triggering control method for an autonomous driving vehicle according to claim 6, further characterized in that: In step (f), the dynamic event triggering mechanism is, in For the controller after triggering, Auxiliary functions designed for s>0, [0,1], h∈(0,1), is the design parameter, The error is triggered by the event; The actual controller design is, Among them, c4>0 is the parameter to be designed.

Images