A vehicle lane-changing obstacle-avoiding method, system, device and storage medium

Abstract

The application discloses a lane-changing obstacle avoidance method, system and device and a storage medium, and is applied to the technical field of vehicle control, and can realize automatic control of lane-changing obstacle avoidance, improve lane-changing obstacle avoidance stability and reliability, and reduce the probability of vehicle collision accidents. The method comprises the following steps: constructing a hub motor vehicle dynamics model; constructing a safety distance model according to the motion relationship between the host vehicle and the front vehicle to construct a lane-changing risk prediction model; when it is determined that lane-changing obstacle avoidance intervention is needed, an optimal obstacle avoidance trajectory is obtained by fitting a lane-changing path through a preset polynomial; a lateral displacement tracking controller is constructed to predict preset motion data of the host vehicle; an optimized integral sliding film controller is constructed to calculate an expected additional yaw moment; a longitudinal vehicle speed self-adaptive fuzzy controller is constructed to calculate an expected longitudinal acceleration; and the hub motor torque is adjusted according to the optimal obstacle avoidance trajectory, the expected additional yaw moment and the expected longitudinal acceleration in combination with the preset motion data.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and in particular to a vehicle lane-changing obstacle avoidance method, system, device, and storage medium. Background Technology

[0002] In intelligent driving, active obstacle avoidance control technology is crucial. Its successful implementation can reduce the probability of traffic accidents to a certain extent, avoid personal injury and economic losses, and improve the utilization efficiency of road traffic resources. When a vehicle encounters obstacles at medium to high speeds or when the vehicle in front brakes suddenly, the vehicle can assist the driver in automatically avoiding obstacles, reducing the probability of accidents. In addition, lane-change obstacle avoidance control can solve the problem of limited longitudinal safe braking distance under high-speed driving conditions. However, in related technologies, lane-change obstacle avoidance risk prediction models often suffer from phenomena such as skidding and slippage when calculating safe distances, which directly affect the actual acceleration of the vehicle. Furthermore, in the path planning part, when finding the optimal obstacle avoidance trajectory, the paths planned in related technologies have poor environmental adaptability, and most path tracking focuses on tracking accuracy control, which is extremely prone to safety accidents under high-speed driving conditions. Summary of the Invention

[0003] To address at least one of the aforementioned technical problems, this invention proposes a vehicle lane-changing obstacle avoidance method, system, device, and storage medium, which enables automatic control of vehicle lane-changing obstacle avoidance, effectively improves the stability and reliability of vehicle lane-changing obstacle avoidance, and effectively reduces the probability of vehicle collision accidents.

[0004] On one hand, embodiments of the present invention provide a vehicle lane-changing obstacle avoidance method, including the following steps:

[0005] Construct a vehicle dynamics model for hub motors;

[0006] Based on the motion relationship between this vehicle and the vehicle in front, construct a safe distance model;

[0007] Construct a lane-change risk prediction model based on the aforementioned safe distance model;

[0008] When lane change risk prediction model determines that lane change obstacle avoidance intervention is required, the optimal obstacle avoidance trajectory is obtained by fitting the lane change path with a preset polynomial.

[0009] A lateral displacement tracking controller is constructed based on the in-wheel motor vehicle dynamics model, and the lateral displacement tracking controller predicts the preset motion data of the vehicle; wherein, the preset motion data includes lateral displacement and heading angle;

[0010] An optimized integral slip film controller is constructed based on the in-wheel motor vehicle dynamics model, and the desired additional yaw moment is calculated through the optimized integral slip film controller; wherein, the desired additional yaw moment is the additional yaw moment required for the vehicle to perform steady-state steering;

[0011] A longitudinal speed adaptive fuzzy controller is constructed, and the desired longitudinal acceleration is calculated through the longitudinal speed adaptive fuzzy controller.

[0012] The torque of the hub motor is adjusted based on the optimal obstacle avoidance trajectory, the desired additional yaw moment, and the desired longitudinal acceleration, combined with the preset motion data.

[0013] According to some embodiments of the present invention, constructing the in-wheel motor vehicle dynamics model includes:

[0014] A seven-degree-of-freedom (DOF) dynamic model of a vehicle is constructed; wherein, the seven-DOF dynamic model of a vehicle includes the longitudinal motion dynamic equilibrium equations of lane changing and obstacle avoidance, the lateral motion dynamic equilibrium equations of lane changing and obstacle avoidance, and the yaw motion dynamic equilibrium equations of lane changing.

[0015] According to some embodiments of the present invention, constructing a safe distance model based on the motion relationship between the vehicle and the vehicle in front includes:

[0016] Based on the first travel distance of the vehicle and the second travel distance of the vehicle in front, a safe braking and obstacle avoidance distance function is constructed.

[0017] The safe braking obstacle avoidance distance function is subjected to vehicle acceleration boundary processing to obtain the safe distance model.

[0018] According to some embodiments of the present invention, constructing a lane-changing risk prediction model based on the safe distance model includes:

[0019] Based on the safety distance model, a dual-fuzzy dynamic identification risk prediction model is constructed using a dual-fuzzy inference algorithm. The dual-fuzzy dynamic identification risk prediction model includes an upper fuzzy identification layer and a lower fuzzy identification layer. The upper fuzzy identification layer is used to identify the motion relationship between the vehicle in the first lane and the vehicle in front to obtain the upper fuzzy output. The lower fuzzy identification layer is used to identify the preset state parameters of the vehicle in front and the vehicle in front. Based on the upper fuzzy output and the preset state parameters, the lane change obstacle avoidance intervention factor is predicted.

[0020] According to some embodiments of the present invention, when lane-change obstacle avoidance intervention is determined by the lane-change risk prediction model, and the optimal obstacle avoidance trajectory is obtained by fitting the lane-change path using a preset polynomial, the step includes:

[0021] Based on the lane change and obstacle avoidance intervention factor, lane change and obstacle avoidance intervention is determined, and a fifth-degree polynomial function is constructed by a preset time coefficient matrix and a boundary condition output matrix.

[0022] Construct a trajectory objective function based on the vehicle's longitudinal acceleration, lateral acceleration, and obstacle avoidance longitudinal distance;

[0023] The optimal obstacle avoidance trajectory is obtained by solving the objective function of the trajectory using the Lagrange multiplier method based on the fifth-order polynomial function.

[0024] According to some embodiments of the present invention, the step of constructing an optimized integral slip controller based on the in-wheel motor vehicle dynamics model, and calculating the desired additional yaw moment through the optimized integral slip controller, includes:

[0025] Construct the desired yaw rate function and the centroid sideslip angle function;

[0026] Construct the system's sliding surface function based on the desired yaw rate function and the centroid sideslip angle function;

[0027] The system's synovial surface function is optimized using an integral term and a synovial factor to obtain an optimized synovial surface function. The first derivative of the synovial surface function is then obtained by differentiating the system's synovial surface function and the optimized synovial surface function.

[0028] A preset control system objective function is constructed based on the aforementioned seven-degree-of-freedom vehicle dynamics model;

[0029] The optimized integral sliding controller is constructed based on the first derivative of the sliding surface and the preset objective function of the control system.

[0030] The additional yaw moment is predicted by the optimized integral sliding controller.

[0031] According to some embodiments of the present invention, the construction of a longitudinal speed adaptive fuzzy controller, and the calculation of the desired longitudinal acceleration by the longitudinal speed adaptive fuzzy controller, includes:

[0032] Construct a universe-of-discourse scaling control function;

[0033] The scaling factor is constructed based on the universe scaling control function.

[0034] Construct a fuzzy proportional-integral control function;

[0035] The longitudinal vehicle speed adaptive fuzzy controller is constructed based on the fuzzy proportional-integral control function and the scaling factor.

[0036] The desired longitudinal acceleration is predicted by the longitudinal speed adaptive fuzzy controller.

[0037] On the other hand, embodiments of the present invention also provide a vehicle lane-changing obstacle avoidance system, including:

[0038] The first building module is used to build the vehicle dynamics model of the hub motor;

[0039] The second construction module is used to build a safe distance model based on the motion relationship between the vehicle and the vehicle in front;

[0040] The third construction module is used to construct a lane-changing risk prediction model based on the safe distance model.

[0041] The trajectory prediction module is used to obtain the optimal obstacle avoidance trajectory by fitting the lane change path through a preset polynomial when lane change risk prediction model determines that lane change and obstacle avoidance intervention is required.

[0042] The fourth construction module is used to construct a lateral displacement tracking controller based on the in-wheel motor vehicle dynamics model, and to predict the preset motion data of the vehicle through the lateral displacement tracking controller; wherein, the preset motion data includes lateral displacement and heading angle;

[0043] The first calculation module is used to construct an optimized integral slip film controller based on the vehicle dynamics model of the hub motor, and to calculate the desired additional yaw moment through the optimized integral slip film controller; wherein, the desired additional yaw moment is the additional yaw moment required for the vehicle to perform steady-state steering;

[0044] The second calculation module is used to construct a longitudinal speed adaptive fuzzy controller, and to calculate the desired longitudinal acceleration through the longitudinal speed adaptive fuzzy controller.

[0045] The control module is used to adjust the torque of the hub motor based on the optimal obstacle avoidance trajectory, the desired additional yaw moment, and the desired longitudinal acceleration, combined with the preset motion data.

[0046] On the other hand, embodiments of the present invention also provide a vehicle lane-changing obstacle avoidance device, comprising:

[0047] At least one processor;

[0048] At least one memory for storing at least one program;

[0049] When the at least one program is executed by the at least one processor, the at least one processor implements a vehicle lane-changing obstacle avoidance method as described in the above embodiments.

[0050] On the other hand, embodiments of the present invention also provide a computer storage medium storing a processor-executable program, which, when executed by the processor, is used to implement a vehicle lane-changing obstacle avoidance method as described in the above embodiments.

[0051] According to an embodiment of the present invention, a vehicle lane-changing obstacle avoidance method has at least the following beneficial effects: First, the embodiment of the present invention constructs a hub motor vehicle dynamics model. Simultaneously, based on the motion relationship between the vehicle and the preceding vehicle, a safety distance model is constructed. A lane-changing risk prediction model is then built based on the safety distance model. This model determines whether lane-changing obstacle avoidance intervention is necessary, effectively improving lane-changing obstacle avoidance safety. Next, when lane-changing obstacle avoidance intervention is determined through the lane-changing risk prediction model, the optimal obstacle avoidance trajectory is obtained by fitting a preset polynomial. This, combined with the vehicle's driving environment, selects the optimal lane-changing obstacle avoidance path while meeting obstacle avoidance requirements and safety margins. Then, the embodiment of the present invention constructs a lateral displacement tracking controller based on the hub motor vehicle dynamics model. This controller predicts preset motion data of the vehicle, including lateral displacement and heading angle. Furthermore, this embodiment of the invention constructs an optimized integral slip film controller based on the vehicle dynamics model of the in-wheel motor. This controller calculates the desired additional yaw moment, i.e., the additional yaw moment required for steady-state steering. By compensating for the steering yaw moment, the steady-state steering problem during lane changes is effectively alleviated. Simultaneously, this embodiment uses a longitudinal speed adaptive fuzzy controller to calculate the desired longitudinal acceleration of the vehicle. Then, based on the optimal obstacle avoidance trajectory, the desired additional yaw moment, and the desired longitudinal acceleration, combined with preset motion data, the in-wheel motor torque is adjusted. This ensures vehicle stability during lane changes while fully leveraging the advantages of the in-wheel motor. The output torque of the in-wheel motor is dynamically adjusted to improve the safety of emergency lane change and obstacle avoidance at high speeds, thereby achieving automatic lane change and obstacle avoidance control. This effectively improves the stability and reliability of lane change and obstacle avoidance, reducing the probability of vehicle collisions. Attached Figure Description

[0052] Figure 1 This is a flowchart of a vehicle lane-changing obstacle avoidance method provided in an embodiment of the present invention;

[0053] Figure 2 This is a schematic diagram of a seven-degree-of-freedom vehicle dynamics model provided in an embodiment of the present invention;

[0054] Figure 3 This is a schematic diagram of the relative motion of a vehicle braking and obstacle avoidance provided in an embodiment of the present invention;

[0055] Figure 4This is a schematic diagram illustrating the relative positions of the vehicle in the left lane and the vehicle itself, provided in an embodiment of the present invention.

[0056] Figure 5 This is a schematic diagram of a vehicle lane-changing obstacle avoidance device provided in an embodiment of the present invention. Detailed Implementation

[0057] The embodiments described in this application should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0058] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0060] In intelligent driving, active obstacle avoidance control technology is crucial. Its successful implementation can reduce the probability of traffic accidents to a certain extent, avoid casualties and economic losses, and improve the utilization efficiency of road traffic resources. According to relevant road traffic safety survey data, more than 80% of motor vehicle collisions are caused by drivers' inability to react in time, of which 24% could have been avoided by changing lanes. When a vehicle encounters obstacles at medium to high speeds or when the vehicle in front brakes suddenly, the vehicle can assist the driver in automatically avoiding obstacles, which can greatly reduce the probability of accidents, ensuring driver safety and improving road efficiency. In addition, lane-change obstacle avoidance control can effectively solve the problem of limited longitudinal safe braking distance under high-speed driving conditions. The vehicle can determine the intervention conditions of the system based on its own driving environment, plan a safe lane-change path, and achieve longitudinal and lateral path tracking. However, in related technologies, the lane-change obstacle avoidance risk prediction model is mainly based on the safe distance model. When calculating the safe distance, it often ignores phenomena such as slippage and skidding that may occur due to changes in the road surface adhesion coefficient, which directly affect the actual acceleration of the vehicle. In addition, in the path planning part, when seeking the optimal obstacle avoidance trajectory, the paths planned by related technologies have poor environmental adaptability, and most path tracking focuses on tracking accuracy control while ignoring the steady-state steering problem of the vehicle during lane changing, which is extremely easy to cause safety accidents under high-speed driving conditions.

[0061] One embodiment of the present invention provides a vehicle lane-changing obstacle avoidance method, system, device, and storage medium, which can realize automatic control of vehicle lane-changing obstacle avoidance, effectively improve the stability and reliability of vehicle lane-changing obstacle avoidance, and effectively reduce the probability of vehicle collision accidents. (Refer to...) Figure 1 The method in this embodiment of the invention includes, but is not limited to, steps S110, S120, S130, S140, S150, S160, S170 and S180.

[0062] Specifically, the application process of the method in this embodiment of the invention includes, but is not limited to, the following steps:

[0063] S110: Construct a vehicle dynamics model with a hub motor.

[0064] S120: Construct a safe distance model based on the motion relationship between this vehicle and the vehicle in front.

[0065] S130: Construct a lane-changing risk prediction model based on the safe distance model.

[0066] S140: When lane change risk prediction model determines that lane change obstacle avoidance intervention is required, the optimal obstacle avoidance trajectory is obtained by fitting the lane change path with a preset polynomial.

[0067] S150: A lateral displacement tracking controller is constructed based on the vehicle dynamics model of the in-wheel motor. The lateral displacement tracking controller predicts the preset motion data of the vehicle. The preset motion data includes lateral displacement and heading angle.

[0068] S160: An optimized integral slip controller is constructed based on the vehicle dynamics model of the in-wheel motor, and the desired additional yaw moment is calculated through the optimized integral slip controller. The desired additional yaw moment is the additional yaw moment required for the vehicle to perform steady-state steering.

[0069] S170: Construct a longitudinal speed adaptive fuzzy controller, and calculate the desired longitudinal acceleration through the longitudinal speed adaptive fuzzy controller.

[0070] S180: Adjust the torque of the hub motor based on the optimal obstacle avoidance trajectory, the desired additional yaw moment, and the desired longitudinal acceleration, combined with preset motion data.

[0071] In this specific embodiment, the present invention first constructs a vehicle dynamics model using a hub motor. Specifically, this embodiment provides theoretical support for trajectory tracking control by constructing a hub motor vehicle dynamics model. The hub motor vehicle dynamics model constructed in this embodiment can be a multi-degree-of-freedom vehicle dynamics model, such as a three-degree-of-freedom, five-degree-of-freedom, or seven-degree-of-freedom dynamics model. Next, this embodiment constructs a safe distance model based on the motion relationship between the vehicle and the preceding vehicle. Specifically, in this embodiment, the vehicle is the primary vehicle, i.e., the self-vehicle. Correspondingly, the preceding vehicle is the vehicle located ahead of the vehicle's travel path. This embodiment constructs a safe distance model by combining the motion relationship between the preceding and self-vehicles, thereby providing theoretical support for the lane-change obstacle avoidance prediction model. Then, this embodiment constructs a lane-change risk prediction model based on the safe distance model. By building a lane-change risk prediction model based on the safe distance model, it can combine changes in the surrounding environment of the vehicle, such as the motion state of the preceding vehicle, relative speed, and the state of vehicles in adjacent lanes, to predict lane-change risks, effectively improving the reliability of lane-change risk prediction. Furthermore, this embodiment of the invention uses a lane-change risk prediction model to predict lane-change risks and determine whether lane-change obstacle avoidance intervention is necessary. When lane-change obstacle avoidance intervention is determined by the lane-change risk prediction model, this embodiment of the invention uses a preset polynomial to fit the lane-change path and obtain the optimal obstacle avoidance trajectory. Specifically, this embodiment of the invention uses the fitted preset polynomial as the obstacle avoidance curve, i.e., the lane-change path, to predict the optimal obstacle avoidance trajectory. Further, this embodiment of the invention constructs a lateral displacement tracking controller based on the in-wheel motor vehicle dynamics model and predicts preset motion data of the vehicle using the lateral displacement tracking controller. Specifically, the preset motion data in this embodiment of the invention includes lateral displacement and heading angle. This embodiment of the invention tracks the lateral displacement and heading angle of the vehicle using the constructed lateral displacement tracking controller, tracking the vehicle's lateral displacement and heading angle using model prediction values. Simultaneously, this embodiment of the invention constructs an optimized integral slip film controller based on the in-wheel motor vehicle dynamics model and calculates the desired additional yaw moment using the optimized integral slip film controller. Specifically, in this embodiment of the invention, the desired additional yaw moment is the additional yaw moment required for the vehicle to perform steady-state steering. It is easy to understand that the steady-state steering problem of vehicles during lane changes can easily lead to safety accidents at high speeds. Therefore, this embodiment of the invention constructs an optimized integral slip film controller to supplement the control of the vehicle's steady-state steering yaw moment by predicting and calculating the additional yaw moment required for steady-state steering. This effectively alleviates the vehicle's steady-state steering problem and improves the safety and reliability of lane-changing obstacle avoidance. Furthermore, this embodiment of the invention constructs a longitudinal speed adaptive fuzzy controller and calculates the desired longitudinal acceleration using this controller.Specifically, this embodiment of the invention predicts the desired steering acceleration of the vehicle using a longitudinal speed adaptive fuzzy controller to track and control the vehicle's longitudinal speed. Next, this embodiment adjusts the torque of the wheel hub motors based on the optimal obstacle avoidance trajectory, the desired additional yaw moment, and the desired longitudinal acceleration, combined with preset motion data. Specifically, this embodiment redistributes the torque of each wheel based on longitudinal speed stability and steady-state steering yaw stability. By adjusting the torque distribution of each wheel according to the optimal obstacle avoidance trajectory, the desired additional yaw moment, the desired longitudinal acceleration, and preset motion data, automatic lane-changing obstacle avoidance control can be achieved, effectively improving the stability and reliability of lane-changing obstacle avoidance and significantly reducing the probability of vehicle collisions.

[0072] In some embodiments of the present invention, a vehicle dynamics model for a hub motor is constructed, including but not limited to:

[0073] A seven-degree-of-freedom (DOF) dynamic model of the vehicle is constructed. This model includes the longitudinal motion equilibrium equations, the lateral motion equilibrium equations, and the yaw motion equilibrium equations for lane-changing and obstacle avoidance.

[0074] In this specific embodiment, the in-wheel motor vehicle dynamics model constructed by this invention includes a seven-degree-of-freedom vehicle dynamics model. Specifically, the seven-degree-of-freedom vehicle dynamics model in this embodiment includes the longitudinal motion dynamics equilibrium equations for lane-changing and obstacle avoidance, the lateral motion dynamics equilibrium equations for lane-changing and obstacle avoidance, and the yaw motion dynamics equilibrium equations for lane-changing. (Refer to...) Figure 2 The vehicle seven-degree-of-freedom dynamic model constructed in this embodiment of the invention includes yaw motion, motion along the y-axis, motion along the x-axis, and longitudinal motion of the four hub motors. This embodiment of the invention provides a dynamic basis for the lateral displacement and heading angle tracking of the obstacle avoidance tracking layer, the calculation of yaw moment compensation required for steady-state steering, and the longitudinal speed tracking and holding control of the vehicle by establishing a seven-degree-of-freedom vehicle dynamic model. For example, the vehicle lane-changing obstacle avoidance longitudinal motion dynamic equilibrium equation constructed in this embodiment of the invention is shown in equation (1) below:

[0075] (1)

[0076] Accordingly, the lateral motion dynamics equation for vehicle lane-changing obstacle avoidance constructed in this embodiment of the invention is shown in equation (2) below:

[0077] (2)

[0078] Accordingly, the vehicle lane-changing yaw motion dynamics equilibrium equation constructed in this embodiment of the invention is shown in equation (3) below:

[0079] (3)

[0080] In equations (1) to (3), For the overall vehicle weight; , These are the vehicle's longitudinal acceleration and lateral acceleration, respectively. , For the vehicle's longitudinal speed and lateral speed; The vehicle's yaw rate; This refers to the front wheel steering angle; It is the moment of inertia; , These are the distances from the vehicle's center of gravity to the front and rear axles, respectively. , These are the front and rear track widths, respectively. , These are the longitudinal and lateral forces of the four tires.

[0081] In some embodiments of the present invention, a safe distance model is constructed based on the motion relationship between the vehicle and the vehicle in front, including but not limited to:

[0082] Based on the first travel distance of this vehicle and the second travel distance of the vehicle in front, construct a safe braking obstacle avoidance distance function.

[0083] By applying vehicle acceleration boundary conditions to the safe braking obstacle avoidance distance function, a safe distance model is obtained.

[0084] In this specific embodiment, the present invention first constructs a safe braking obstacle avoidance distance function based on the first travel distance of the current vehicle and the second travel distance of the preceding vehicle. Then, it performs vehicle acceleration boundary processing on the safe braking obstacle avoidance distance function to obtain a safe distance model. Specifically, before performing lane-changing obstacle avoidance, it is necessary to clarify the application scenarios and intervention conditions of lane-changing obstacle avoidance, analyze the safe distance model, and provide theoretical support for the establishment of a lane-changing obstacle avoidance prediction model. (Refer to...) Figure 3 Assuming that only the vehicle in front and the vehicle in front are in motion in the same lane, and ignoring the sliding distance when the wheels lock up, the relative positional relationship between the vehicle and the vehicle in front is as shown in equation (4):

[0085] (4)

[0086] Where, in the formula To ensure safe braking and obstacle avoidance distance, The distance traveled by the vehicle in front. This represents the distance traveled by the vehicle. To minimize the collision distance, ,Pick .

[0087] Accordingly, the vehicle's travel distance in this embodiment of the invention It can be shown in the following formula (5):

[0088] (5)

[0089] Where, in the formula Let be the initial velocity of the vehicle. To maintain the same following speed as the vehicle in front after slowing down, For the vehicle's braking deceleration, The braking system's delay time is generally taken as... In the embodiments of the present invention, the following are taken .

[0090] It should be noted that, in this embodiment of the invention, the distance traveled by the preceding vehicle can be divided into three cases based on the initial state of the preceding vehicle: stationary, constant speed, and braking, as shown in the following formula (6):

[0091] (6)

[0092] Where, in the formula The initial speed of the vehicle in front; The braking deceleration of the vehicle in front.

[0093] Accordingly, equation (4) is processed according to equations (5) and (6) to obtain the safe braking distance function for vehicles in front and behind on the road surface after processing, i.e., the safe braking distance function, as shown in equation (7) below:

[0094] (7)

[0095] It is easy to understand that, under the same initial conditions for the vehicles in the embodiments of the present invention, when , The braking safety distance is the longest when... At that time, based on actual speculation, The braking safety distance should be the shortest. Among them, , The classification of reference braking comfort levels is shown in the following formula (8):

[0096] (8)

[0097] In addition, one of the conditions for lane-changing obstacle avoidance intervention in this embodiment of the invention is: ,in, The real-time distance between the vehicle and the vehicle in front is given. Furthermore, in this embodiment of the invention, the vehicle acceleration is boundary-based based on the road surface adhesion of each wheel, as shown in equation (9):

[0098] (9)

[0099] Where, in the formula This is the minimum value of the four tire adhesion coefficients. In this embodiment of the invention, by combining the road surface adhesion of each wheel during the construction of the safety distance model, the vehicle acceleration is boundary-processed to optimize the safety distance model, effectively alleviating the problem of slippage and drift caused by changes in the road surface adhesion coefficient affecting the actual vehicle acceleration. Thus, the final safety distance model is constructed as shown in equation (10):

[0100] (10)

[0101] It should be noted that, in the embodiments of the present invention, for such irregular concepts that are biased towards the processing of empirical information, fuzzy theory in intelligent algorithms is an effective method that can mimic the imprecise and nonlinear information processing of the human brain.

[0102] In some embodiments of the present invention, a lane-changing risk prediction model is constructed based on a safe distance model, including but not limited to:

[0103] Based on the safety distance model, a dual-fuzzy dynamic identification risk prediction model is constructed using a dual-fuzzy inference algorithm. This model comprises an upper fuzzy identification layer and a lower fuzzy identification layer. The upper fuzzy identification layer identifies the motion relationship between the vehicle in the first lane and the vehicle in front, obtaining the upper-layer fuzzy output. The lower fuzzy identification layer predicts the lane-changing obstacle avoidance intervention factor based on the upper-layer fuzzy output and preset state parameters of the vehicle and the vehicle in front.

[0104] In this specific embodiment, the present invention constructs a dual-fuzzy dynamic identification risk prediction model based on a safety distance model and a dual-fuzzy inference algorithm. Specifically, the dual-fuzzy dynamic identification risk prediction model constructed in this embodiment includes an upper fuzzy identification layer and a lower fuzzy identification layer. The present invention identifies the motion relationship between the vehicle in the first lane and the vehicle itself through the upper fuzzy identification layer to obtain the upper-layer fuzzy output. For example, the present invention identifies the position and motion state of the vehicle in the left lane relative to the vehicle itself through the upper fuzzy identification layer. Next, the present invention identifies preset state parameters of the vehicle and the vehicle in front through the lower fuzzy identification layer, and predicts a lane-change obstacle avoidance intervention factor based on the upper-layer fuzzy output. For example, the present invention identifies key state parameters such as the relative distance, speed, and acceleration between the vehicle and the vehicle in front, i.e., preset state parameters, through the lower fuzzy identification layer, and predicts lane-change obstacle avoidance based on the upper-layer fuzzy output and preset state parameters to obtain the lane-change obstacle avoidance intervention factor, thereby enabling the determination of whether lane-change obstacle avoidance is necessary based on the lane-change obstacle avoidance intervention factor.

[0105] For example, in this embodiment of the invention, the relative positional relationship between the vehicle and the car in the left lane is as follows: Figure 4 As shown in the figure. , These are the centroid coordinates of the vehicle in the left lane and the vehicle itself, respectively. Define the length of the vehicle and the vehicle in the left lane. This refers to the longitudinal spacing between vehicles in the left lane and the vehicle itself. , The longitudinal speeds of the vehicle in the left lane and the vehicle itself are defined, and the relative speed between the two vehicles is defined. , This represents the longitudinal acceleration of the vehicle in the left lane. In this embodiment of the invention, the upper fuzzy recognition layer determines the longitudinal distance between the vehicle in the left lane and the vehicle itself. and the relative speed of the two vehicles Relative acceleration (Right now As the fuzzy recognition input for the driving status of vehicles in the left lane, the state of the vehicles in the left lane relative to the vehicle itself. As the system output, this embodiment of the invention first defines the corresponding input-output fuzzy subsets, as shown in equations (11) to (14) below:

[0106] (11)

[0107] (12)

[0108] (13)

[0109] (14)

[0110] Where, in the formula They represent negative, zero, and positive, respectively. These represent negative large, negative medium, negative small, zero, positive small, positive medium, and positive large, respectively. , , , , , , , , These represent deceleration from the left rear, constant speed from the left rear, acceleration from the left rear, adjacent deceleration, adjacent constant speed, adjacent acceleration, deceleration from the left front, constant speed from the left front, and acceleration from the left front. The basic domain of discourse is , The basic domain of discourse is , The basic domain of discourse is , The basic domain of discourse is .

[0111] Accordingly, in this embodiment of the invention, the membership function of the fuzzy algorithm is triangular. During the formulation of fuzzy rules, when the relative speed and relative acceleration of two vehicles are constant, the larger the longitudinal distance, the stronger the lateral attribute of the vehicle in the left lane; the smaller the longitudinal distance, the stronger the adjacency attribute. Alternatively, when the longitudinal distance and relative acceleration of two vehicles are constant, the smaller the relative speed, the more fixed the relative position attribute of the vehicles; or when the longitudinal distance and relative speed of two vehicles are constant, the smaller the relative acceleration, the more fixed the relative position attribute of the vehicles. The fuzzy rules formulated in this embodiment of the invention are shown in Table 1 below:

[0112] Table 1

[0113]

[0114] Accordingly, in this embodiment of the invention, the lower fuzzy recognition layer determines the state of the vehicle in the left lane relative to the vehicle itself. Actual vehicle distance With minimum braking distance deviation and the braking strength of the vehicle in front. As input to the fuzzy controller, the output of the entire system is an obstacle avoidance factor that evaluates the collision risk. And define the corresponding input and output fuzzy subsets as shown in equations (15) to (18) respectively:

[0115] (15)

[0116] (16)

[0117] (17)

[0118] (18)

[0119] Where, in the formula middle Let these represent negative, zero, and positive respectively, with the universe of discourse being... . middle , Let these represent small, medium, large, and maximal values, respectively, with a universe of discourse of . . middle They represent safe, relatively dangerous, dangerous, and extremely dangerous, respectively, with a domain of _____. Meanwhile, in this embodiment of the invention, the membership function of the fuzzy algorithm is triangular, and the overall obstacle avoidance rules followed by the formulation of the obstacle avoidance fuzzy rules for the three vehicles (the vehicle itself, the vehicle in front, and the vehicle in the left lane) are shown in Table 2 below:

[0120] Table 2

[0121]

[0122] Furthermore, embodiments of the present invention are based on obstacle avoidance risk factors. This refers to the lane-change obstacle avoidance intervention factor, which determines whether lane-change obstacle avoidance intervention should be initiated. In this embodiment of the invention, it is based on the obstacle avoidance risk factor. Three hazard levels are defined, from level one to level three. At level one, no lane change or obstacle avoidance intervention is initiated. At level two, lane change and obstacle avoidance intervention is initiated, and steering is performed. At level three, lane change intervention is not initiated, and braking is performed, as shown in Table 3 below.

[0123] Table 3

[0124]

[0125] In some embodiments of the present invention, when lane-change obstacle avoidance intervention is determined by a lane-change risk prediction model, the optimal obstacle avoidance trajectory is obtained by fitting a lane-change path using a preset polynomial, including but not limited to:

[0126] The lane-change obstacle avoidance intervention is determined based on the lane-change obstacle avoidance intervention factor, and a fifth-degree polynomial function is constructed by using a preset time coefficient matrix and boundary condition output matrix.

[0127] The trajectory objective function is constructed based on the vehicle's longitudinal acceleration, lateral acceleration, and obstacle avoidance longitudinal distance.

[0128] The optimal obstacle avoidance trajectory is obtained by solving the objective function of the trajectory using the Lagrange multiplier method based on the fifth-order polynomial function.

[0129] In this specific embodiment, the present invention first determines whether to intervene in lane-change obstacle avoidance based on a lane-change obstacle avoidance intervention factor. If lane-change obstacle avoidance intervention is determined, the present invention constructs a fifth-order polynomial function using a preset time coefficient matrix and a boundary condition output matrix. Simultaneously, the present invention constructs a trajectory objective function based on the vehicle's longitudinal acceleration, lateral acceleration, and obstacle avoidance longitudinal distance. Further, the present invention solves for the objective function of the vehicle using the Lagrange multiplier method based on the fifth-order polynomial function, thereby obtaining the optimal obstacle avoidance trajectory. Specifically, in this specific embodiment, the initial position of the vehicle lane-change obstacle avoidance... and completion position When the constraint condition shown in equation (19) is satisfied:

[0130] (19)

[0131] Where, in the formula Represents displacement; Represents speed; Represents acceleration; The time required to complete obstacle avoidance; The longitudinal distance required to complete obstacle avoidance; The lane width is equal to the lateral distance traveled by the vehicle when obstacle avoidance is completed. In this embodiment of the invention, we take... .

[0132] Furthermore, the coefficient vectors set in the embodiments of the present invention are respectively , Thus, we can obtain information about and Regarding time The coefficient matrix equation system is shown in equation (20) below:

[0133] (20)

[0134] Where, in the formula The time coefficient matrix, and This is the output matrix for the boundary conditions. Correspondingly, the time coefficient matrix in this embodiment of the invention... As shown in equation (21), the boundary condition output matrix and As shown in equations (22) and (23) respectively:

[0135] (twenty one)

[0136] (twenty two)

[0137] (twenty three)

[0138] Furthermore, in this embodiment of the invention, by calculating the above equation (20), a fifth-degree polynomial function is constructed as shown in the following equation (24):

[0139] (twenty four)

[0140] Next, regarding the stability of longitudinal and lateral vehicle speeds and the efficiency of obstacle avoidance during lane changes, this embodiment of the invention constructs a trajectory objective function related to longitudinal acceleration, lateral acceleration, and longitudinal distance for obstacle avoidance, as shown in the following equation (25):

[0141] (25)

[0142] Where, in the formula These are the weighting coefficients. These are the vehicle's longitudinal acceleration, vehicle's lateral acceleration, and obstacle avoidance longitudinal distance, respectively.

[0143] Furthermore, embodiments of the present invention modify the above equation (24) by adjusting the time... Calculate the second derivative Substituting into equation (25), and simplifying, we get the following equation (26):

[0144] (26)

[0145] Next, this embodiment of the invention solves the limit using the Lagrange multiplier method, focusing on the longitudinal obstacle avoidance distance. and obstacle avoidance time Find the extreme points of a bivariate function. The optimal solution corresponding to the minimum value , Satisfy the following equation (27):

[0146] (27)

[0147] It is easy to understand that, according to the above formula (27), in the embodiments of the present invention, when the initial velocity and weight coefficient are known, the total time required for obstacle avoidance can be calculated. and longitudinal driving distance .right , , Different weighting coefficients result in different obstacle avoidance times at high speeds. Longer obstacle avoidance times require greater longitudinal obstacle avoidance distances, leading to higher collision risks. When the optimal obstacle avoidance curve meets the obstacle avoidance requirements and the prediction model's safety margin requirements, the weighting coefficient of lateral acceleration can be increased to maintain vehicle ride comfort when the obstacle avoidance risk is low. Conversely, when the obstacle avoidance risk is high, the weighting coefficient of lateral acceleration should be decreased, and the proportion of longitudinal obstacle avoidance distance should be increased to improve lane-changing efficiency.

[0148] In some embodiments of the present invention, an optimized integral slip controller is constructed based on a vehicle dynamics model of a hub motor, and the desired additional yaw moment is calculated by optimizing the integral slip controller, including but not limited to:

[0149] Construct the desired yaw rate function and the centroid sideslip angle function.

[0150] The system's sliding surface function is constructed based on the desired yaw rate function and the center of mass sideslip angle function.

[0151] The system's synovial surface function is optimized by integrating terms and synovial factors to obtain the optimized synovial surface function. The first derivative of the synovial surface is then obtained by differentiating the system's synovial surface function and the optimized synovial surface function.

[0152] The objective function of the preset control system is constructed based on the seven-degree-of-freedom dynamics model of the vehicle.

[0153] An optimized integral sliding controller is constructed based on the first derivative of the sliding surface and the preset objective function of the control system.

[0154] The additional yaw moment is predicted by optimizing the integral sliding controller.

[0155] In this specific embodiment, the present invention first constructs the desired yaw rate function and the centroid sideslip angle function, and constructs the system slick surface function based on the desired yaw rate function and the centroid sideslip angle function. Next, the present invention optimizes the system slick surface function by using integral terms and slick surface factors to obtain the slick surface optimization function, and differentiates the system slick surface function and the slick surface optimization function to obtain the corresponding slick surface first derivative. Further, the present invention constructs a preset control system objective function based on the vehicle's seven-degree-of-freedom dynamics model, and constructs an optimized integral slick surface controller based on the slick surface first derivative and the preset control system objective function, thereby predicting the additional yaw moment through the optimized integral slick surface controller. Specifically, in the present invention, the optimized integral slick surface controller uses the steering wheel angle as the input quantity to calculate the desired yaw rate and centroid sideslip angle of steady-state steering. The general form of the steady-state steering angle in the present invention is shown in the following formula (28):

[0156] (28)

[0157] Where, in the formula This indicates insufficient steering ratio. , representing centripetal acceleration, Wheelbase The steady-state turning radius, The component of the vehicle's mass acting on the front axle. The component of the vehicle's mass acting on the rear axle. These are the wheelbases from the center of mass to the front and rear axles, respectively. These are the lateral stiffness of the front and rear single wheels, respectively.

[0158] Accordingly, the desired yaw rate in this embodiment of the invention is obtained by the following equation (29):

[0159] (29)

[0160] Accordingly, the ideal centroid deflection angle of the vehicle in steady state in this embodiment of the invention is shown in the following equation (30):

[0161] (30)

[0162] It should be noted that in this embodiment of the invention, the yaw stability control process determines the additional yaw moment through a control algorithm, and the final execution point is allocated to the torque distribution of the vehicle's four wheels. Therefore, in this embodiment of the invention, the torque balance equation established based on the seven-degree-of-freedom dynamic model, when the required additional yaw moment is... The objective function of the preset control system can be obtained as shown in equation (31):

[0163] (31)

[0164] Furthermore, in the design of the sliding surface in this embodiment of the invention, the deviations between the two steady-state targets and the steady-state desired parameters are used as the system sliding surface function, as shown in the following equation (32):

[0165] (32)

[0166] Where, in the formula To control the weighting coefficients.

[0167] Meanwhile, based on the above system sliding surface function, due to system disturbances and chattering phenomena in the calculation of sliding control, this embodiment of the invention introduces an integral term and a sliding factor to optimize the sliding surface, thereby enhancing the robustness of the system. Accordingly, the optimized sliding surface function obtained after optimization in this embodiment of the invention is shown in the following equation (33):

[0168] (33)

[0169] Where, in the formula For positive gain, The value of will affect the system's steady-state time. The larger the value, the faster the system response, but... If the value is too high, the system will experience chattering. Let be a constant, representing the synovial factor, and .

[0170] Furthermore, in this embodiment of the invention, the derivatives of equations (32) and (33) are taken to obtain the first derivative of the system's sliding surface. That is, the first derivative of the synovial surface, as shown in equation (34):

[0171] (34)

[0172] Meanwhile, the Lyapunov function stability judgment function is constructed in this embodiment of the invention, as shown in the following equation (35):

[0173] (35)

[0174] It is easy to understand that, by It can be seen that the optimized integral sliding controller constructed in this embodiment of the invention is convergent and stable. Next, this embodiment of the invention rearranges the above equation (34) to obtain the system's equivalent control law. As shown in equation (36):

[0175] (36)

[0176] Accordingly, in the embodiments of the present invention when At times, the system is prone to fluctuations, and the system parameters will deviate from the state trajectory. Therefore, it is necessary to choose an appropriate reaching law. To reduce system chattering while increasing system convergence speed, this embodiment of the invention introduces a saturation function. The results are shown in equations (37) and (38) respectively:

[0177] (37)

[0178] (38)

[0179] Where, in the formula , For controller parameters, the embodiments of the present invention use larger values. Improve convergence speed, smaller Reduce vibration, For boundary layer.

[0180] Furthermore, in this embodiment of the invention, the above equations (36) and (37) are added to construct the objective function of the optimized integral slippage controller, namely the additional yaw moment of the vehicle. The function is shown in equation (39) below:

[0181] (39)

[0182] Accordingly, embodiments of the present invention predict the additional yaw moment of the vehicle through an optimized integral slippage controller. .

[0183] In some embodiments of the present invention, a longitudinal speed adaptive fuzzy controller is constructed, and the desired longitudinal acceleration is calculated by the longitudinal speed adaptive fuzzy controller, including but not limited to:

[0184] Construct a universe-of-discourse scaling control function.

[0185] The scaling factor is constructed based on the domain scaling control function.

[0186] Construct a fuzzy proportional-integral control function.

[0187] A longitudinal speed adaptive fuzzy controller is constructed based on the fuzzy proportional-integral control function and the scaling factor.

[0188] The desired longitudinal acceleration is predicted by an adaptive fuzzy controller based on longitudinal vehicle speed.

[0189] In this specific embodiment, the present invention first constructs a universe-of-discourse scaling control function, and then constructs a scaling factor based on the universe-of-discourse scaling control function. Simultaneously, the present invention constructs a fuzzy proportional-integral (PI) control function, and then constructs a longitudinal speed adaptive fuzzy controller based on the fuzzy PPI control function and the scaling factor, thereby enabling the prediction of the corresponding desired longitudinal acceleration through the longitudinal speed adaptive fuzzy controller. Specifically, in this embodiment, the longitudinal speed adaptive fuzzy controller uses the deviation between the actual vehicle speed driven by the hub motor and the desired speed calculated using a fifth-order polynomial. and its deviation change rate As input to the system, the increments of the proportional and integral coefficients serve as outputs, correcting the PI controller parameters. The PI controller inputs the speed deviation and outputs the desired longitudinal acceleration of the vehicle. For example, to enable dynamic adjustment of the input and output of the fuzzy system based on the input error and rate of change, this embodiment of the invention establishes a corresponding universe of discourse scaling system. The new input and output universes of discourse after universe of discourse scaling control... and It can be expressed by the following formula (40):

[0190] (40)

[0191] Where, in the formula , These are the scaling factors for the input and output universes of discourse, respectively. , . and These represent the initial input universe of discourse and the output universe of discourse of the fuzzy controller, respectively.

[0192] Accordingly, the scaling factor constructed in this embodiment of the invention is shown in equation (41):

[0193] (41)

[0194] Where, in the formula and For the domain adjustment factor, and , Its value is limited by the system's accuracy. and This represents the sensitivity coefficient, which determines the response speed of the system's universe of discourse. , >0. In some embodiments of the present invention, the scaling factor is specifically... , Accordingly, the output universe value in this embodiment of the invention is shown in equation (42):

[0195] (42)

[0196] Where, in the formula This is the scaling factor. This is the weight coefficient vector. , This is the initial value for the output universe of discourse. It's easy to understand that the output universe of discourse should follow... The larger, The larger the scaling factor, The smaller the scaling factor, the better. In this embodiment of the invention, the scaling factor is set... scaling factor , scaling factor .

[0197] Furthermore, the fuzzy proportional-integral control function constructed in this embodiment of the invention is shown in equation (43) below:

[0198] (43)

[0199] Where, in the formula , These are the initial parameters for the fuzzy proportional-integral controller. , To adjust the amount in real time.

[0200] Meanwhile, the embodiments of the present invention define vehicle speed error. The fundamental universe of discourse is [-0.6, 0.6], and the rate of change of vehicle speed deviation is... The fundamental domain is [-30, 30]. and The basic universes of discourse are [-120, 120] and [-30, 30]. Correspondingly, in this embodiment of the invention, both the input and output universes of discourse are [-6, -5, -4, -3, -2, -1, 0, 1, 2, 3, 4, 5, 6]. Furthermore, the fuzzy subsets are all... , representing negative large, negative medium, negative small, zero, positive small, positive medium, and positive large, respectively. In this embodiment of the invention, the membership function of the fuzzy algorithm is triangular. In the established fuzzy rules, when the longitudinal vehicle speed deviation... When it is large, it should be increased. , reduce Quickly adjust system parameters to stabilize vehicle speed, when longitudinal vehicle speed deviation and When it is not large, it should be reduced slightly. Increase Maintaining the system control state when longitudinal vehicle speed deviation When it is small, it should be reduced slightly. Increase To maintain system stability, therefore, in the embodiments of the present invention , The specific fuzzy rule table is shown in Table 4 below:

[0201] Table 4

[0202]

[0203] Furthermore, the desired acceleration expression of the longitudinal vehicle speed adaptive fuzzy controller constructed in this embodiment of the invention is shown in the following equation (44):

[0204] (44)

[0205] It should be noted that in some embodiments of the present invention, during the process of adjusting the wheel hub motor torque based on the optimal obstacle avoidance trajectory, the desired additional yaw moment, and the desired longitudinal acceleration combined with preset motion data, the wheel hub motor torque is readjusted based on longitudinal vehicle speed maintenance and lateral steady-state steering. For example, in order to simultaneously meet the requirements of yaw stability and vehicle speed stability, the incremental driving force of each wheel in the embodiments of the present invention meets the longitudinal acceleration constraint and the additional yaw moment constraint, while also ensuring that the motor working torque is within the maximum torque constraint and meeting the road surface adhesion conditions. According to the fluctuation of vehicle speed during obstacle avoidance, the vertical load of the front and rear axles of the vehicle will be redistributed. Therefore, according to the vehicle's seven-degree-of-freedom dynamics model, the requirements it needs to meet are as shown in the following equation (45):

[0206] (45)

[0207] Where, in the formula These are the additional driving forces for the front left, front right, rear left, and rear wheels, respectively. and These are the front and rear track widths, respectively. For longitudinal acceleration increment, This is a vertical load on the front and rear axles. Accordingly, in the embodiments of the present invention... The calculation process is shown in the following formula (46):

[0208] (46)

[0209] Where, in the formula , The radius is the wheel radius.

[0210] It should be noted that, in some embodiments of the present invention, when constructing a lateral displacement tracking controller based on a wheel hub motor vehicle dynamics model, the wheel hub motor vehicle dynamics model is first simplified, considering two degrees of freedom: lateral motion and yaw motion. That is, it is assumed that the slip angle and slip stiffness of the left and right wheels of the vehicle are the same, and the lateral acceleration is within the range of... Within the range, the front wheel steering angle is relatively small. Accordingly, the total lateral force of the vehicle in this embodiment of the invention can be represented by the following equation (47):

[0211] (47)

[0212] Where, in the formula , These represent the total lateral forces acting on the front and rear wheels, respectively. , The lateral stiffness of the front and rear wheels of the vehicle. This is the sideslip angle of the centroid. Accordingly, in this embodiment of the invention, the sideslip angle of the centroid... It can be expressed as the following formula (48):

[0213] (48)

[0214] Based on the above formulas, the relevant state differential equations of the prediction system can be expressed as follows (49) to (53):

[0215] (49)

[0216] (50)

[0217] (51)

[0218] (52)

[0219] (53)

[0220] Where, in the formula This is a lateral displacement. This is the heading angle.

[0221] Furthermore, before performing linearization, this embodiment of the invention first sets the state variables of the prediction model as follows: The control quantity is The output is As can be seen from equation (53) above, there is a significant nonlinear relationship between the state variables of the system. Accordingly, in this embodiment of the invention, the state equation is... Performing a Taylor expansion at the given point, ignoring higher-order terms and retaining only first-order terms, yields the linearizable state equation and output equation as shown in equation (54):

[0222] (54)

[0223] Where, in the formula , , , , The partial derivatives of the Jacobian matrix are shown in equations (55) and (56):

[0224] (55)

[0225] (56)

[0226] It is easy to understand that the state equation after linearization in this embodiment of the invention has continuity. Therefore, it is also necessary to perform forward Euler discretization on the state equation, as shown in the following equation (57):

[0227] (57)

[0228] Where, in the formula , , This represents the deviation between system linearization and discretization. The number of state variables. , , All are linear matrices. Next, this embodiment of the invention further optimizes the system by setting the control quantity from the previous moment. and the state quantity at the current moment To form new state variables, its augmented matrix As shown in equation (58):

[0229] (58)

[0230] Accordingly, the state equation and output equation of the new state variable in this embodiment of the invention are shown in the following equation (59):

[0231] (59)

[0232] Where, in the formula , , , , To control the quantity.

[0233] Next, in this embodiment of the invention, the output equation is iterated over a rolling process. To predict the step size, To control the step size, the output equation is expressed in matrix form as shown in equation (60):

[0234] (60)

[0235] Where, in the formula To output column vectors, For control quantity column vectors, , , These are the state variables, control variables, and error rolling coefficient matrices, respectively, and their specific expressions are shown in equations (61) to (66) below:

[0236] (61)

[0237] (62)

[0238] (63)

[0239] (64)

[0240] (65)

[0241] (66)

[0242] Furthermore, in this embodiment of the invention, the desired control quantity that minimizes the tracking error of the system is calculated through rolling optimization, and an objective function is constructed concerning the output deviation, the control increment within the prediction step size range, and the control quantity limit value, as shown in the following equation (67):

[0243] (67)

[0244] Where, in the formula , , , To characterize the weight matrix for different target parameters, It is a relaxation factor.

[0245] Next, in this embodiment of the invention, the dynamic constraint equations are substituted into the objective function for calculation, and the optimal control increment is corrected to achieve the current desired result. This serves as the input to the controller. By repeatedly performing the above steps within the control cycle, real-time tracking and control of the obstacle avoidance trajectory can be achieved.

[0246] An embodiment of the present invention also provides a vehicle lane-changing obstacle avoidance system, comprising:

[0247] The first building block is used to build the vehicle dynamics model of the hub motor.

[0248] The second construction module is used to build a safe distance model based on the motion relationship between the vehicle and the vehicle in front.

[0249] The third building module is used to construct a lane-change risk prediction model based on the safe distance model.

[0250] The trajectory prediction module is used to determine the need for lane change and obstacle avoidance intervention through the lane change risk prediction model, and obtain the optimal obstacle avoidance trajectory by fitting the lane change path with a preset polynomial.

[0251] The fourth module is used to construct a lateral displacement tracking controller based on the vehicle dynamics model of the hub motor, and to predict the vehicle's preset motion data through the lateral displacement tracking controller. The preset motion data includes lateral displacement and heading angle.

[0252] The first calculation module is used to construct an optimized integral slip controller based on the vehicle dynamics model of the hub motor, and to calculate the desired additional yaw moment through the optimized integral slip controller. The desired additional yaw moment is the additional yaw moment required for the vehicle to perform steady-state steering.

[0253] The second calculation module is used to construct a longitudinal speed adaptive fuzzy controller, which calculates the desired longitudinal acceleration.

[0254] The control module is used to adjust the torque of the hub motor based on the optimal obstacle avoidance trajectory, the desired additional yaw moment, and the desired longitudinal acceleration, combined with preset motion data.

[0255] Reference Figure 5 An embodiment of the present invention also provides a vehicle lane-changing obstacle avoidance device, comprising:

[0256] At least one processor 210.

[0257] At least one memory 220 is used to store at least one program.

[0258] When at least one program is executed by at least one processor 210, the at least one processor 210 implements a vehicle lane-changing obstacle avoidance method as described in the above embodiments.

[0259] An embodiment of the present invention also provides a computer-readable storage medium storing computer-executable instructions that are executed by one or more control processors, for example, performing the steps described in the above embodiments.

[0260] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0261] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A method for vehicle lane-changing obstacle avoidance, characterized in that, Includes the following steps: Construct a vehicle dynamics model for hub motors; Based on the motion relationship between this vehicle and the vehicle in front, construct a safe distance model; Construct a lane-change risk prediction model based on the aforementioned safe distance model; When lane change risk prediction model determines that lane change obstacle avoidance intervention is required, the optimal obstacle avoidance trajectory is obtained by fitting the lane change path with a preset polynomial. A lateral displacement tracking controller is constructed based on the in-wheel motor vehicle dynamics model, and the lateral displacement tracking controller predicts the preset motion data of the vehicle; wherein, the preset motion data includes lateral displacement and heading angle; An optimized integral slip film controller is constructed based on the in-wheel motor vehicle dynamics model, and the desired additional yaw moment is calculated through the optimized integral slip film controller; wherein, the desired additional yaw moment is the additional yaw moment required for the vehicle to perform steady-state steering; A longitudinal speed adaptive fuzzy controller is constructed, and the desired longitudinal acceleration is calculated through the longitudinal speed adaptive fuzzy controller. The torque of the hub motor is adjusted based on the optimal obstacle avoidance trajectory, the desired additional yaw moment, and the desired longitudinal acceleration, combined with the preset motion data. The step of constructing a lane-changing risk prediction model based on the safe distance model includes: Based on the safety distance model, a dual-fuzzy dynamic identification risk prediction model is constructed using a dual-fuzzy inference algorithm. The dual-fuzzy dynamic identification risk prediction model includes an upper fuzzy identification layer and a lower fuzzy identification layer. The upper fuzzy identification layer is used to identify the motion relationship between the vehicle in the first lane and the vehicle in front to obtain the upper fuzzy output. The lower fuzzy identification layer is used to identify the preset state parameters of the vehicle in front and the vehicle in front. Based on the upper fuzzy output and the preset state parameters, the lane change obstacle avoidance intervention factor is predicted.

2. The vehicle lane-changing obstacle avoidance method according to claim 1, characterized in that, The construction of the in-wheel motor vehicle dynamics model includes: A seven-degree-of-freedom (DOF) dynamic model of a vehicle is constructed; wherein, the seven-DOF dynamic model of a vehicle includes the longitudinal motion dynamic equilibrium equations of lane changing and obstacle avoidance, the lateral motion dynamic equilibrium equations of lane changing and obstacle avoidance, and the yaw motion dynamic equilibrium equations of lane changing.

3. The vehicle lane-changing obstacle avoidance method according to claim 1, characterized in that, The process of constructing a safe distance model based on the motion relationship between the vehicle and the vehicle in front includes: Based on the first travel distance of the vehicle and the second travel distance of the vehicle in front, a safe braking and obstacle avoidance distance function is constructed. The safe braking obstacle avoidance distance function is subjected to vehicle acceleration boundary processing to obtain the safe distance model.

4. The vehicle lane-changing obstacle avoidance method according to claim 3, characterized in that, When lane-change obstacle avoidance intervention is determined through the lane-change risk prediction model, the optimal obstacle avoidance trajectory is obtained by fitting the lane-change path using a preset polynomial, including: Based on the lane change and obstacle avoidance intervention factor, lane change and obstacle avoidance intervention is determined, and a fifth-degree polynomial function is constructed by a preset time coefficient matrix and a boundary condition output matrix. Construct a trajectory objective function based on the vehicle's longitudinal acceleration, lateral acceleration, and obstacle avoidance longitudinal distance; The optimal obstacle avoidance trajectory is obtained by solving the objective function of the trajectory using the Lagrange multiplier method based on the fifth-order polynomial function.

5. A vehicle lane-changing obstacle avoidance method according to claim 2, characterized in that, The step of constructing an optimized integral slip controller based on the in-wheel motor vehicle dynamics model, and calculating the desired additional yaw moment through the optimized integral slip controller, includes: Construct the desired yaw rate function and the centroid sideslip angle function; Construct the system's sliding surface function based on the desired yaw rate function and the centroid sideslip angle function; The system's synovial surface function is optimized using an integral term and a synovial factor to obtain an optimized synovial surface function. The first derivative of the synovial surface function is then obtained by differentiating the system's synovial surface function and the optimized synovial surface function. A preset control system objective function is constructed based on the aforementioned seven-degree-of-freedom vehicle dynamics model; The optimized integral sliding controller is constructed based on the first derivative of the sliding surface and the preset objective function of the control system. The additional yaw moment is predicted by the optimized integral sliding controller.

6. A vehicle lane-changing obstacle avoidance method according to claim 2, characterized in that, The construction of a longitudinal speed adaptive fuzzy controller, through which the desired longitudinal acceleration is calculated, includes: Construct a universe-of-discourse scaling control function; The scaling factor is constructed based on the universe scaling control function. Construct a fuzzy proportional-integral control function; The longitudinal vehicle speed adaptive fuzzy controller is constructed based on the fuzzy proportional-integral control function and the scaling factor. The desired longitudinal acceleration is predicted by the longitudinal speed adaptive fuzzy controller.

7. A vehicle lane-changing obstacle avoidance system, characterized in that, include: The first building module is used to build the vehicle dynamics model of the hub motor; The second construction module is used to build a safe distance model based on the motion relationship between the vehicle and the vehicle in front; The third construction module is used to construct a lane-changing risk prediction model based on the safe distance model. The trajectory prediction module is used to obtain the optimal obstacle avoidance trajectory by fitting the lane change path through a preset polynomial when lane change risk prediction model determines that lane change and obstacle avoidance intervention is required. The fourth construction module is used to construct a lateral displacement tracking controller based on the in-wheel motor vehicle dynamics model, and to predict the preset motion data of the vehicle through the lateral displacement tracking controller; wherein, the preset motion data includes lateral displacement and heading angle; The first calculation module is used to construct an optimized integral slip film controller based on the vehicle dynamics model of the hub motor, and to calculate the desired additional yaw moment through the optimized integral slip film controller; wherein, the desired additional yaw moment is the additional yaw moment required for the vehicle to perform steady-state steering; The second calculation module is used to construct a longitudinal speed adaptive fuzzy controller, and to calculate the desired longitudinal acceleration through the longitudinal speed adaptive fuzzy controller. The control module is used to adjust the torque of the hub motor based on the optimal obstacle avoidance trajectory, the desired additional yaw moment, and the desired longitudinal acceleration, combined with the preset motion data. The third building module is specifically used to perform the following steps: Based on the safety distance model, a dual-fuzzy dynamic identification risk prediction model is constructed using a dual-fuzzy inference algorithm. The dual-fuzzy dynamic identification risk prediction model includes an upper fuzzy identification layer and a lower fuzzy identification layer. The upper fuzzy identification layer is used to identify the motion relationship between the vehicle in the first lane and the vehicle in front to obtain the upper fuzzy output. The lower fuzzy identification layer is used to identify the preset state parameters of the vehicle in front and the vehicle in front. Based on the upper fuzzy output and the preset state parameters, the lane change obstacle avoidance intervention factor is predicted.

8. A vehicle lane-changing obstacle avoidance device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a vehicle lane-changing obstacle avoidance method as described in any one of claims 1 to 6.

9. A computer storage medium storing a processor-executable program, characterized in that, The program executable by the processor, when executed by the processor, is used to implement a vehicle lane-changing obstacle avoidance method as described in any one of claims 1 to 6.

Images