Intelligent steer-by-wire automatic parking method for a semi-trailer

Abstract

The application discloses a semi-trailer intelligent steer-by-wire automatic parking method, which comprises the following steps: S1, performing parking environment modeling, vehicle state definition and kinematics modeling; S2, generating a parking path based on a hybrid A* algorithm; and S3, using a model predictive control algorithm, designing a target function and performing optimization solving to realize rolling optimization of an automatic parking process. The semi-trailer intelligent steer-by-wire automatic parking method provided by the application realizes autonomous parking control of a semi-trailer in a complex environment, and realizes automatic collision avoidance, posture adjustment, path minimization and control cost in the parking process.

Description

Technical Field

[0001] This invention relates to the field of vehicle intelligent control technology, specifically to a method for intelligent drive-by-wire steering and automatic parking of a semi-trailer. Background Technology

[0002] When parking a semi-trailer, due to its length and large turning radius, the driver has difficulty accurately judging surrounding obstacles using the rearview mirror, resulting in low parking efficiency and a high risk of collisions. Existing semi-trailer assistance systems mostly rely on manual operation and lack automated steering and path coordination control. Furthermore, existing automatic parking systems for ordinary vehicles are not adapted to the articulated structure of semi-trailers, and traditional path planning algorithms do not consider the impact of dynamic changes in the folding angle on the rear axle trajectory.

[0003] Therefore, it is necessary to propose a method for intelligent drive-by-wire steering and automatic parking of semi-trailers. Summary of the Invention

[0004] The purpose of this invention is to provide a method for intelligent drive-by-wire steering and automatic parking of semi-trailers, in order to solve the technical problems existing in the background art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for intelligent drive-by-wire steering automatic parking of a semi-trailer includes the following steps:

[0007] S1, perform parking environment modeling, and define vehicle state and kinematic modeling;

[0008] S2, generating parking paths based on a hybrid A* algorithm;

[0009] S3 utilizes model predictive control algorithms to design and optimize the objective function, thereby achieving rolling optimization of the automatic parking process.

[0010] Furthermore, in step S1, when modeling the parking environment, a point cloud map of the parking area is generated by scanning with LiDAR, and information on road boundaries, target parking spaces, and obstacles is identified.

[0011] Furthermore, in step S1, when defining the vehicle state and modeling its kinematics, the semi-trailer state vector is used... It means that, among them:

[0012] x and y represent the coordinates of the rear axle center of the tractor. Indicates the heading angle of the tractor unit. Indicates the folding angle between the tractor and trailer. Indicates the trailer's heading angle, with ;

[0013] In the kinematic model of the vehicle, the differential equations are as follows:

[0014] ;

[0015] ;

[0016] ;

[0017] ;

[0018] ;

[0019] Where v represents the reversing speed, L1 represents the wheelbase of the tractor, and L2 represents the wheelbase of the trailer;

[0020] semi-trailer control variables express, Indicates the front axle steering angle of the tractor unit. The rear axle steering angle of the trailer;

[0021] After discretizing the semi-trailer's state vector, the state is updated using the Euler method, as shown below:

[0022] ;

[0023] in Let represent the state vector at step k. , This represents the semi-trailer control variable at step k.

[0024] Furthermore, in step S2, when generating the parking path based on the hybrid A* algorithm, specifically:

[0025] The following heuristic function is used for path planning:

[0026] ;

[0027] Where S represents the position portion of the current state vector [ ], that is, the coordinates of the rear axle center of the tractor, and H(S) represents the total cost estimate at the current position S. This is a heuristic term from the traditional A* algorithm, representing the Euclidean distance between the current position and the target position. λ is the attitude penalty term, and λ is the dynamic weight of the folding angle.

[0028] And a reference trajectory is generated by interpolation, based on the initial state. and target state Generate reference path points with uniform temporal distribution:

[0029] ;

[0030] in, This represents the total number of steps.

[0031] Furthermore, in step S3, the following objective function is established:

[0032] ;

[0033] in This indicates the state tracking error. Let Q and R represent the semi-trailer control variables at step k, where Q and R are both weight matrices.

[0034] And includes the following constraints:

[0035] Position constraints restrict the x and y coordinates of the rear axle center of the towing vehicle to be within a preset rectangle when parking is completed;

[0036] Obstacle avoidance constraints ensure that the trailer's rear coordinates always avoid identified obstacles. The conversion can be performed using the following formula:

[0037] ;

[0038] The consistency constraint ensures that the trailer and tractor are aligned when parking is completed, i.e., it satisfies... ;

[0039] Furthermore, during parking, the folding angle will be... The maximum value is limited to 30°, which will reduce the folding angle. The rate of change is limited to no more than 15° / s.

[0040] Furthermore, in step S3, when optimizing the objective function, the fmincon function is used to solve the constrained nonlinear optimization problem to output the following optimal control sequence. :

[0041] .

[0042] Compared with the prior art, the advantages of the present invention are as follows:

[0043] The intelligent drive-by-wire steering automatic parking method for semi-trailers provided by this invention enables autonomous parking control of semi-trailers in complex environments, and achieves automatic collision avoidance, posture adjustment, path minimization, and control costs during the parking process. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in this embodiment, the accompanying drawings used in the description of the embodiment will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart illustrating an intelligent drive-by-wire steering automatic parking method for semi-trailers provided by the present invention.

[0046] Figure 2 This is a simulation diagram of the parking process at the beginning;

[0047] Figure 3 This is a simulation diagram showing the parking process nearing its end. Detailed Implementation

[0048] To make the technical means, creative features, objectives and effects of this invention easier to understand, the following description, in conjunction with the accompanying drawings and specific embodiments, further explains how this invention is implemented.

[0049] In one specific embodiment, refer to Figure 1 As shown, the present invention provides a method for intelligent drive-by-wire steering automatic parking of a semi-trailer, comprising the following steps:

[0050] S1, perform parking environment modeling, and define vehicle state and kinematic modeling;

[0051] S2, generating parking paths based on a hybrid A* algorithm;

[0052] S3 utilizes model predictive control algorithms to design and optimize the objective function, thereby achieving rolling optimization of the automatic parking process.

[0053] In one specific embodiment, during parking environment modeling in step S1, a point cloud map of the parking area is generated by scanning with LiDAR, and road boundaries, target parking spaces, and obstacle information are identified. When generating the parking area point cloud map, devices such as cameras and ultrasonic sensors can also be used, and vehicle attitude angle data provided by an IMU device can be combined with high-precision positioning information obtained from a GPS device to generate the vehicle's initial pose.

[0054] In a simulation experiment, referring to Figure 2 and Figure 3 As shown, the target parking space is represented by a dashed line, and its left coordinates are [10, -7.5, 10, 4].

[0055] When defining the vehicle state and modeling its kinematics, the semi-trailer state vector is used. It means that, among them:

[0056] x and y represent the coordinates of the rear axle center of the tractor. Indicates the heading angle of the tractor unit. Indicates the folding angle between the tractor and trailer. Indicates the trailer's heading angle, with ;

[0057] In the kinematic model of the vehicle, the differential equations are as follows:

[0058] ;

[0059] ;

[0060] ;

[0061] ;

[0062] ;

[0063] Where v represents the reversing speed, L1 represents the wheelbase of the tractor, and L2 represents the wheelbase of the trailer.

[0064] semi-trailer control variables express, Indicates the front axle steering angle of the tractor unit. The rear axle steering angle of the trailer.

[0065] After discretizing the semi-trailer's state vector, the state is updated using the Euler method, as shown below:

[0066] ;

[0067] in Let represent the state vector at step k. , This represents the semi-trailer control variable at step k.

[0068] In the vehicle's kinematic model, the state is updated using the reversing speed v and the discrete time step Ts. In this embodiment, v = -1 m / s and Ts = 0.2 s can be set.

[0069] In step S2, when generating the parking path based on the hybrid A* algorithm, specifically:

[0070] The following heuristic function is used for path planning:

[0071] ;

[0072] Where S represents the position portion of the current state vector [ ], that is, the coordinates of the rear axle center of the tractor, and H(S) represents the total cost estimate at the current position S. This is a heuristic term from the traditional A* algorithm, representing the Euclidean distance between the current position and the target position. λ is the attitude penalty term, and λ is the dynamic weight of the folding angle.

[0073] Traditional A* algorithms only consider geometric paths. This invention introduces an attitude penalty term based on the Euclidean distance heuristic of the traditional A* algorithm to restrict folding behavior in the path and ensure that the path satisfies the motion coupling characteristics of the trailer and the tractor.

[0074] And a reference trajectory is generated by interpolation, based on the initial state. and target state λ is the dynamic weight of the folding angle, which avoids excessive bending and generates reference path points with a uniform temporal distribution.

[0075] ;

[0076] in, This represents the total number of steps.

[0077] In step S3, the following objective function is established:

[0078] ;

[0079] in This indicates the state tracking error. Let Q and R represent the control variables for the semi-trailer at step k. Both Q and R are weight matrices. Weight matrices Q and R correspond to the weights of the state variables and control variables, respectively. The weight matrix R is used to suppress excessive changes in the control variables and avoid abrupt changes in direction, i.e., to suppress... Mutations that cause changes in the system can help maintain its stability.

[0080] And includes the following constraints:

[0081] Position constraints restrict the rear axle center coordinates (x, y) of the tractor vehicle to within a preset rectangular frame when parking is completed. In this embodiment, 10 ≤ x ≤ 20, -7 ≤ y ≤ -3.

[0082] Obstacle avoidance constraints ensure that the trailer's rear coordinates always avoid identified obstacles. The conversion can be performed using the following formula:

[0083] ;

[0084] The consistency constraint ensures that the trailer and tractor are aligned when parking is completed, i.e., it satisfies... .

[0085] Furthermore, during parking, the folding angle will be... The maximum value is limited to 30°, which will reduce the folding angle. The rate of change is limited to no more than 15° / s.

[0086] Furthermore, in step S3, when optimizing the objective function, the fmincon function is used to solve the constrained nonlinear optimization problem to output the following optimal control sequence. :

[0087] .

[0088] Obtain the optimal control sequence Then, the front axle steering angle is automatically adjusted accordingly. Rear axle steering angle In this way, the precise parking of the semi-trailer can be achieved.

[0089] In summary, the intelligent drive-by-wire steering automatic parking method for semi-trailers provided by this invention enables autonomous parking control of semi-trailers in complex environments, and achieves automatic collision avoidance, posture adjustment, path minimization, and control costs during the parking process.

[0090] Finally, it should be noted that the above description is only an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for intelligent drive-by-wire steering automatic parking of a semi-trailer, characterized in that, Includes the following steps: S1, perform parking environment modeling, and define vehicle state and kinematic modeling; S2, generating parking paths based on a hybrid A* algorithm; S3 utilizes model predictive control algorithms to design objective functions and perform optimization solutions to achieve rolling optimization of the automatic parking process; In step S1, when defining the vehicle state and modeling the kinematics, the semi-trailer state vector is used... It means that, among them: x, y This indicates the coordinates of the rear axle center of the tractor unit. Indicates the heading angle of the tractor unit. Indicates the folding angle between the tractor and trailer. Indicates the trailer's heading angle, with ; In the kinematic model of the vehicle, the differential equations are as follows: ； ； ； ； ； in, v The L1 indicates the reversing speed, and the L2 indicates the wheelbase of the tractor unit. semi-trailer control variables express, Indicates the front axle steering angle of the tractor unit. Indicates the rear axle steering angle of the trailer; After discretizing the semi-trailer's state vector, the state is updated using the Euler method, as shown below: ； in Let represent the state vector at step k. , This represents the semi-trailer control variable at step k; In step S2, when generating the parking path based on the hybrid A* algorithm, specifically: The following heuristic function is used for path planning: ； in, S This represents the positional part of the current state vector. ], that is, the coordinates of the rear axle center of the tractor. H ( S () indicates the current position S Total cost estimate This is a heuristic term from the traditional A* algorithm, representing the Euclidean distance between the current position and the target position. For posture penalty items, λ Dynamic weights for folding angles; And a reference trajectory is generated by interpolation, based on the initial state. and target state Generate reference path points with uniform temporal distribution: ； in, This represents the total number of steps. In step S3, the following objective function is established: ； in , representing the state tracking error, Let Q and R represent the semi-trailer control variables at step k, where Q and R are both weight matrices. And includes the following constraints: Position constraints ensure that the coordinates of the rear axle center of the tractor are within the range when parking is complete. x, y Confined within a preset rectangular frame; Obstacle avoidance constraints ensure that the trailer's rear coordinates always avoid identified obstacles. The conversion can be performed using the following formula: ； The consistency constraint ensures that the trailer and tractor are aligned when parking is completed, i.e., it satisfies... ; Furthermore, during parking, the folding angle will be... The maximum value is limited to 30°, which will reduce the folding angle. The rate of change is limited to no more than 15° / s.

2. The intelligent drive-by-wire steering automatic parking method for semi-trailers according to claim 1, characterized in that, In step S1, when modeling the parking environment, a point cloud map of the parking area is generated by scanning with LiDAR, and information on road boundaries, target parking spaces, and obstacles is identified.

3. The intelligent drive-by-wire steering automatic parking method for semi-trailers according to claim 2, characterized in that, In step S3, when optimizing the objective function, the fmincon function is used to solve the constrained nonlinear optimization problem to output the following optimal control sequence. : 。

Images