An active obstacle avoidance control method for intelligent semi-trailer train

Abstract

The application discloses a kind of active obstacle avoidance control methods of intelligent semi-trailer train, the method includes: based on the sensor of being arranged on semi-trailer train obtains the information of surrounding environment;Surrounding environment information is preprocessed, and based on the surrounding environment information after preprocessing constructs the obstacle profile area around semi-trailer train;Based on the obstacle profile area constructs the drivable safety area with semi-trailer train as center;Based on the drivable safety area establishes the active obstacle avoidance decision model of semi-trailer train;Based on the active obstacle avoidance decision model obtains the corresponding control strategy of semi-trailer train in longitudinal and lateral direction;Based on the corresponding control strategy of semi-trailer train in longitudinal and lateral direction, the longitudinal speed and lateral moving direction of semi-trailer train are autonomously controlled.In the embodiment of the application, it can help intelligent semi-trailer train to autonomously avoid obstacles in complex traffic environment, improve driving safety and efficiency.

Description

Technical Field

[0001] The application relates to the field of intelligent vehicle control methods, and in particular to an active obstacle avoidance control method for intelligent semi-trailer trucks. Background Technology

[0002] Because of their longer wheelbase, semi-trailer trucks are prone to hitting surrounding obstacles when turning or changing lanes, affecting their driving. Their longer body also makes finding suitable lane positions difficult at high speeds. Furthermore, semi-trailer trucks are more difficult to handle, requiring advanced driving skills to avoid obstacles. However, current intelligent active obstacle avoidance systems and methods are only suitable for small cars and cannot support the long wheelbase, long body, and difficult-to-maneuver nature of semi-trailer trucks. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides an intelligent semi-trailer truck train active obstacle avoidance method that can promptly identify and avoid obstacles and road conditions on the road, helping the intelligent semi-trailer truck train to autonomously avoid obstacles in complex traffic environments, thereby reducing the probability of traffic accidents and improving driving safety.

[0004] To address the aforementioned technical problems, embodiments of the present invention provide an active obstacle avoidance control method for intelligent semi-trailer trucks, the method comprising:

[0005] Information about the surrounding environment is obtained based on sensors installed on the semi-trailer truck train;

[0006] The surrounding environment information is preprocessed, and the obstacle outline region around the semi-trailer truck is constructed based on the preprocessed surrounding environment information.

[0007] A safe driving zone centered on the semi-trailer truck is constructed based on the outline region of the obstacle.

[0008] Based on the drivable safe zone, an active obstacle avoidance decision model is established for the semi-trailer truck.

[0009] Based on the active obstacle avoidance decision model, the corresponding control strategies for the semi-trailer truck train in the longitudinal and lateral directions are obtained.

[0010] Based on the corresponding control strategies for the semi-trailer train in the longitudinal and lateral directions, the system autonomously controls the longitudinal speed and lateral movement direction of the semi-trailer train.

[0011] Preferably, the sensors include a camera, a lidar, a millimeter-wave radar, and an ultrasonic radar; wherein the lidar is installed on the windshield of the semi-trailer tractor; the millimeter-wave radar is installed on the semi-trailer tractor, on the sides and rear of the semi-trailer; the ultrasonic radar is installed around the bottom of the semi-trailer tractor and the semi-trailer; and the camera is installed at the bottom of the windshield of the semi-trailer tractor, on the left and right rearview mirrors, on the top of the driver's cab, and on the door of the semi-trailer tractor at the rear of the semi-trailer.

[0012] The camera is used to acquire a pixel array of RGB images around the semi-trailer truck train;

[0013] The lidar is used to acquire 3D point cloud distance information around the semi-trailer truck train;

[0014] The millimeter-wave radar is used to acquire point data of objects around the semi-trailer truck train.

[0015] The ultrasonic radar is used to acquire radar data of objects around the semi-trailer truck train.

[0016] Preferably, the data preprocessing of the surrounding environment information includes:

[0017] The pixel arrays of the RGB images obtained from each camera are sequentially stitched, denoised, smoothed, and transformed to obtain the pixel arrays of the processed RGB image.

[0018] Extract and select pixel features from the pixel array of the processed RGB image;

[0019] Based on the pixel features, obstacle recognition processing is performed to obtain a first set of recognized obstacles corresponding to the pixel features;

[0020] The 3-D point cloud distance information acquired by each set lidar is processed by removing the ground point cloud based on the depth data to obtain the processed 3-D point cloud distance information;

[0021] A clustering algorithm based on Euclidean distance is used to cluster the distance information of the processed 3-D point cloud, and a kd-tree is constructed based on the clustering results.

[0022] Search each node of the kd-tree and obtain the set of point clouds where the number of nodes is less than a preset threshold;

[0023] Based on the aforementioned point cloud, a second set of obstacles for LiDAR identification is obtained;

[0024] The point data of surrounding objects acquired by each millimeter-wave radar is subjected to point filtering and noise reduction processing to obtain the noise-reduced point data.

[0025] The noise-reduced dot data is processed to obtain the third set of recognized obstacles corresponding to the noise-reduced dot data.

[0026] The radar data obtained from each ultrasonic radar is denoised to obtain denoised radar data.

[0027] The distance between the semi-trailer train and the obstacle is obtained based on the noise-reduced radar data.

[0028] Preferably, the obstacle outline region around the semi-trailer truck is constructed based on the preprocessed surrounding environment information, including:

[0029] The preprocessed surrounding environment information is comprehensively analyzed using OR and Boolean operations to obtain obstacle information around the semi-trailer train. The obstacle information includes obstacle type, obstacle position relative to the semi-trailer train, and obstacle distance relative to the semi-trailer train.

[0030] The obstacle outline region around the semi-trailer truck is constructed based on the obstacle information around the semi-trailer truck.

[0031] Preferably, constructing a safe drivable zone centered on the semi-trailer truck based on the obstacle contour region includes:

[0032] Based on the actual dimensions and outlines of the semi-trailer tractor and the semi-trailer, the intersection angle between the semi-trailer tractor and the semi-trailer is determined, and the intersection angle θ between the semi-trailer tractor and the semi-trailer is calculated as follows:

[0033]

[0034] In the above formula, θ is the articulation angle between the intelligent semi-trailer truck tractor and the semi-trailer; V B δ(t) represents the tractor's speed, obtained from a speed sensor; δ(t) represents the tractor's instantaneous turning angle, obtained from a turning angle sensor; L1 represents the wheelbase of the semi-trailer tractor; L2 represents the wheelbase of the semi-trailer.

[0035] The safe distance between the semi-trailer and the obstacle is determined based on the intersection angle between the semi-trailer tractor and the semi-trailer and the outline area of ​​the obstacle around the semi-trailer.

[0036] Based on the size and safety spacing of the semi-trailer truck, a safe driving zone is constructed centered on the semi-trailer truck, extending outwards from the obstacle outline area.

[0037] The safe driving area is updated in real time based on changes in the position or shape of obstacles and changes in the angle of intersection between the semi-trailer tractor and the semi-trailer.

[0038] Preferably, the safety distance includes the semi-trailer's critical obstacle avoidance distance and the desired safety distance;

[0039]

[0040] In the above formula, D br The critical obstacle avoidance distance D for trailer braking q Let v be the desired safe distance, and v be the speed of the semi-trailer train. p a is the instantaneous velocity of the obstacle; max For the maximum permissible deceleration of a semi-trailer truck train, t d For the braking system delay time, t h To set the workshop time distance, μ is the road surface adhesion coefficient, d0 is the reserved expected safety distance, and f(μ) is the proportional function of the road surface adhesion coefficient.

[0041] Preferably, the proportional function f(μ) of the road adhesion coefficient is:

[0042]

[0043] In the above formula, μ min μ is the coefficient of adhesion on icy and snowy roads. norm This is the adhesion coefficient of the asphalt pavement.

[0044] Preferably, the step of establishing an active obstacle avoidance decision model for the semi-trailer truck based on the drivable safe area includes:

[0045] The obstacle avoidance strategies include, but are not limited to: when there is only an obstacle in front of the semi-trailer truck and the distance to the obstacle reaches the critical obstacle avoidance distance for the semi-trailer truck to brake, and emergency braking is not timely enough to take a double-sided lane change strategy.

[0046] When there are obstacles in front of the semi-trailer truck and obstacles on both sides, and the distance between the obstacles and the semi-trailer truck is within the expected safe distance, a following strategy should be adopted.

[0047] When there are obstacles in front of the semi-trailer truck and obstacles on both sides, and the distance between the obstacles is close and the semi-trailer truck reaches the critical obstacle avoidance distance for braking, an emergency braking strategy should be adopted.

[0048] Preferably, the control strategies for the semi-trailer truck train in the longitudinal and lateral directions are obtained based on the active obstacle avoidance decision model, including:

[0049] The lateral control strategy for the semi-trailer truck is derived by maintaining the longitudinal speed at a low speed (3 km / h or 5 km / h), and using a fifth-order polynomial for the lane-changing trajectory model, with the analytical function y(t) as follows:

[0050]

[0051] In the above formula, t is the time for the semi-trailer truck train to change lanes. tol The time required for a semi-trailer truck train to change lanes; y la This refers to the lateral displacement of the intelligent vehicle train during lane changing.

[0052] Differentiating the above equation yields the required lateral acceleration 'a' for the semi-trailer truck train. v ():

[0053]

[0054] In the above formula, t is the time for the semi-trailer truck train to change lanes. tol The time required for a semi-trailer truck train to change lanes; y la For the lateral displacement of the intelligent vehicle train during the lane change process, a y (t) represents the lateral acceleration required for the semi-trailer truck train.

[0055] Preferably, the control strategies for the semi-trailer truck train in the longitudinal and lateral directions are obtained based on the active obstacle avoidance decision model, including:

[0056] The longitudinal control strategy of the semi-trailer truck is derived by using the speed difference to establish the driving and braking torque equations of the semi-trailer truck, as follows:

[0057]

[0058] In the above formula, P m To solve for the proportional coefficient of the driving torque, I m To solve for the integral time coefficient of the driving torque, D m To solve for the differential time coefficient of the driving torque, T f To overcome rolling friction torque, the function v diff (t) is a preset function, the v diff (t) Taking the safe driving zone centered on the semi-trailer truck as a reference, the speed difference between the speed of the semi-trailer truck and the desired speed of the semi-trailer truck is calculated according to the preset rules.

[0059] The proportional coefficient P for solving the driving torque m for:

[0060] P m =0+ pmΔP m + pv ΔP v + ps ΔP s ;

[0061] In the above formula, P0 is the initial proportional value, and k pm ΔP is the proportional load mass factor. m k represents the proportional load capacity value. pv ΔP is the proportional velocity coefficient. v This is the proportional velocity value; k ps ΔP is the proportional rotation coefficient. s This is the proportional angle value;

[0062] The integral time coefficient I for solving the driving torque m for:

[0063] I m =0+ Im ΔI m + Iv ΔI v + Is ΔI s ;

[0064] In the above formula, I0 is the initial integral value, and k Im The integral load mass coefficient, ΔI m k represents the integral load capacity value. Iv The integral velocity coefficient, ΔI v k is the integral velocity value. Is Let ΔI be the integral rotation coefficient. s This is the integral angle value;

[0065] The differential time coefficient D of the driving torque solution m for:

[0066] D m =0+ Dm ΔD m + Dv ΔD v + Ds ΔD s ;

[0067] In the above formula, D0 is the initial differential value, and k Dm Let ΔD be the differential load mass coefficient. m k is the differential load mass value. Dv Let ΔD be the differential velocity coefficient. v k is the differential velocity value. Ds Let ΔD be the differential rotation coefficient. s This is the differential angle value.

[0068] In this embodiment of the invention, the active obstacle avoidance control method of the intelligent semi-trailer truck can identify and avoid obstacles and road conditions in a timely manner, helping the intelligent semi-trailer truck to autonomously avoid obstacles in complex traffic environments, thereby reducing the probability of traffic accidents and improving driving safety. Attached Figure Description

[0069] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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.

[0070] Figure 1 This is a flowchart illustrating an active obstacle avoidance control method for an intelligent semi-trailer truck train according to an embodiment of the present invention.

[0071] Figure 2 This is an obstacle recognition flowchart of an active obstacle avoidance control method for an intelligent semi-trailer truck train according to an embodiment of the present invention;

[0072] Figure 3 This is a schematic diagram of the outline of a semi-trailer truck and the obstacle recognition results of an active obstacle avoidance control method for an intelligent semi-trailer truck in an embodiment of the present invention.

[0073] Figure 4 This is a schematic diagram of the dual-lane-changing strategy of an active obstacle avoidance control method for an intelligent semi-trailer truck in an embodiment of the present invention.

[0074] Figure 5 This is a schematic diagram of a single-sided lane-changing strategy for an active obstacle avoidance control method for an intelligent semi-trailer truck in an embodiment of the present invention.

[0075] Figure 6 This is a schematic diagram of the following strategy of an active obstacle avoidance control method for an intelligent semi-trailer truck in an embodiment of the present invention.

[0076] Figure 7 This is a schematic diagram of emergency braking for an active obstacle avoidance control method for an intelligent semi-trailer truck according to an embodiment of the present invention. Detailed Implementation

[0077] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0078] Figure 1 This diagram illustrates a flowchart of an active obstacle avoidance control method for an intelligent semi-trailer truck according to an embodiment of the present invention.

[0079] like Figure 1 As shown, an active obstacle avoidance control method for an intelligent semi-trailer truck is disclosed, the method comprising:

[0080] S11: Acquire information about the surrounding environment based on sensors installed on the semi-trailer truck;

[0081] In the specific implementation of this invention, the sensors include a camera, a lidar, a millimeter-wave radar, and an ultrasonic radar; wherein the lidar is installed on the windshield of the semi-trailer truck tractor; the millimeter-wave radar is installed on the semi-trailer truck tractor, on the side and rear of the semi-trailer truck; the ultrasonic radar is installed around the bottom of the semi-trailer truck tractor and the semi-trailer truck; the camera is installed at the bottom of the windshield of the semi-trailer truck tractor, on the left and right rearview mirrors of the semi-trailer truck, on the top of the driver's cab of the semi-trailer truck, and on the door of the semi-trailer truck tractor at the rear of the semi-trailer truck tractor.

[0082] Furthermore, the camera is used to acquire pixel arrays of RGB images around the semi-trailer train; the lidar is used to acquire 3-D point cloud distance information around the semi-trailer train; the millimeter-wave radar is used to acquire point data of objects around the semi-trailer train; and the ultrasonic radar is used to acquire radar data of objects around the semi-trailer train.

[0083] Preferably, the tractor unit of the semi-trailer truck is equipped with a combined navigation and positioning system. This system provides precise vehicle positioning and navigation services, helping drivers operate the vehicle more safely and efficiently. The combined navigation and positioning system typically consists of multiple navigation sensors and algorithms, including a Global Positioning System (GPS), an Inertial Navigation System (INS), and map data. These sensors and algorithms work together to provide accurate vehicle positioning, speed, attitude, and navigation information. In a semi-trailer truck, the navigation and positioning system helps the driver understand the vehicle's current position, speed, and direction, and provides navigation guidance based on a pre-planned route. Furthermore, the system provides real-time traffic and road information, helping drivers avoid congestion and dangerous areas. In emergencies, the system can also provide accurate vehicle location and status information, helping rescue personnel quickly reach the scene and conduct rescue operations.

[0084] Semi-trailers are equipped with inertial elements to reduce vehicle impact and vibration, improving ride smoothness and safety. Inertial elements are typically rotating mass components, such as flywheels or gyroscopes, possessing high inertia and stability. When a semi-trailer traverses uneven roads or turns, the vehicle's center of gravity shifts, causing swaying and vibration. At this time, the inertial element functions, stabilizing the vehicle through rotation and reducing swaying and vibration. Inertial elements also help the semi-trailer to start, accelerate, and brake more smoothly, thus reducing impact and damage to the driver and cargo. Therefore, installing inertial elements on semi-trailers improves ride smoothness and safety, reducing vehicle damage and maintenance costs.

[0085] S12: Perform data preprocessing on the surrounding environment information, and construct the obstacle outline area around the semi-trailer truck based on the preprocessed surrounding environment information.

[0086] Figure 2 This paper illustrates an obstacle recognition flowchart of an active obstacle avoidance control method for an intelligent semi-trailer truck according to an embodiment of the present invention. In the specific implementation of the present invention, the pixel arrays of the RGB images obtained by each camera are sequentially stitched, denoised, smoothed, and transformed to obtain the pixel array of the processed RGB image; pixel features of the pixel array of the processed RGB image are extracted and selected; obstacle recognition processing is performed based on the pixel features to obtain a first set of recognized obstacles corresponding to the pixel features;

[0087] Furthermore, the 3D point cloud distance information acquired by each LiDAR is processed by removing ground point clouds based on depth data to obtain processed 3D point cloud distance information; the processed 3D point cloud distance information is clustered using a clustering algorithm based on Euclidean distance, and a kd-tree is constructed based on the clustering results; each node of the kd-tree is searched and a set of point clouds with nodes less than a preset threshold is obtained; a second set of obstacles for LiDAR recognition is obtained based on the set of point clouds.

[0088] Furthermore, the point data of surrounding objects acquired by each millimeter-wave radar is subjected to point filtering and noise reduction processing to obtain noise-reduced point data; the noise-reduced point data is then subjected to recognition processing to obtain the third set of recognized obstacles corresponding to the noise-reduced point data.

[0089] Furthermore, the radar data obtained by each ultrasonic radar is subjected to noise reduction processing to obtain noise-reduced radar data; the distance between the semi-trailer train and the obstacle is obtained based on the noise-reduced radar data.

[0090] Furthermore, the preprocessed surrounding environment information is comprehensively analyzed using OR and Boolean operations to obtain obstacle information around the semi-trailer train. The obstacle information includes obstacle type, obstacle position relative to the semi-trailer train, and obstacle distance relative to the semi-trailer train.

[0091] The obstacle outline region around the semi-trailer truck is constructed based on the obstacle information around the semi-trailer truck.

[0092] S13: Constructing a safe drivable zone centered on the semi-trailer truck based on the obstacle outline region includes:

[0093] For reference Figure 3 The diagram showing the obstacle recognition results illustrates that, in the specific implementation of this invention, after obtaining the outline area of ​​the obstacle, based on the actual size and outline of the semi-trailer tractor and the semi-trailer, the intersection angle between the semi-trailer tractor and the semi-trailer is determined by combining the navigation controller and inertial elements, based on the actual size and outline of the semi-trailer tractor and the semi-trailer. The intersection angle θ between the semi-trailer tractor and the semi-trailer is calculated as follows:

[0094]

[0095] In the above formula, θ is the articulation angle between the intelligent semi-trailer truck tractor and the semi-trailer; V B δ(t) represents the tractor's speed, obtained from a speed sensor; δ(t) represents the tractor's instantaneous turning angle, obtained from a turning angle sensor; L1 represents the wheelbase of the semi-trailer tractor; L2 represents the wheelbase of the semi-trailer.

[0096] Furthermore, based on the intersection angle between the semi-trailer tractor and the semi-trailer and the outline area of ​​the obstacles around the semi-trailer, a safe distance between the semi-trailer and the obstacles is determined; based on the size of the semi-trailer and the safe distance, a safe driving area centered on the semi-trailer is constructed by expanding outward from the outline area of ​​the obstacles.

[0097] Preferably, the safe driving area is updated in real time based on changes in the position or shape of obstacles and changes in the junction angle between the semi-trailer tractor and the semi-trailer.

[0098] Furthermore, the safety distance includes the semi-trailer's critical obstacle avoidance distance and the desired safety distance;

[0099]

[0100] In the above formula, D br The critical obstacle avoidance distance D for trailer braking qLet v be the desired safe distance, and v be the speed of the semi-trailer train. p a is the instantaneous velocity of the obstacle; max For the maximum permissible deceleration of a semi-trailer truck train, t d For the braking system delay time, t h To set the workshop time distance, μ is the road surface adhesion coefficient, d0 is the reserved expected safety distance, and f(μ) is the proportional function of the road surface adhesion coefficient.

[0101] Specifically, the proportional function f(μ) of the road adhesion coefficient is:

[0102]

[0103] In the above formula, μ min μ is the coefficient of adhesion on icy and snowy roads. norm This is the adhesion coefficient of the asphalt pavement.

[0104] S14: Based on the drivable safe area, establish an active obstacle avoidance decision model for the semi-trailer truck, including:

[0105] The obstacle avoidance strategy includes, but is not limited to: reference Figure 4 A diagram illustrating a dual-lane-changing strategy is provided. This strategy is applicable when a semi-trailer truck encounters an obstacle only ahead and the distance to the obstacle reaches the semi-trailer's critical braking distance. (Refer to...) Figure 5 A diagram illustrating a single-sided lane-changing strategy is shown. If emergency braking is not timely, a double-sided lane-changing strategy may be employed.

[0106] refer to Figure 6 The diagram illustrates the following strategy: when there are obstacles in front of the semi-trailer truck and obstacles on both sides, and the distance between the obstacles and the semi-trailer truck meets the expected safe distance, the following strategy is adopted.

[0107] refer to Figure 7 The diagram illustrates an emergency braking strategy employed when a semi-trailer truck encounters obstacles in front and on both sides, with the obstacles being close enough to reach the semi-trailer's critical obstacle avoidance distance.

[0108] Specifically, the obstacle avoidance strategy uses information about the vehicle and surrounding obstacles as input, and aims at vehicle safety and stability, designing an intelligent semi-trailer truck train decision-making and control flow. Taking the longitudinal direction of the intelligent semi-trailer truck train as the object, if there are obstacles on both sides, and the longitudinal distance is long, the train maintains its current course; if a warning distance is reached, the train must not exceed the speed of the vehicle in front; if the distance is short, emergency braking is initiated. When there are no obstacles on either side or on one side, if the longitudinal distance meets the safe lane-changing distance, a lane-changing action is taken; if the safe lane-changing distance is not met, the longitudinal distance needs to be reassessed. If the distance is safe, the train maintains its current course; if a warning is issued, the train follows the vehicle in front at a speed not exceeding that of the vehicle in front; if a dangerous distance is reached, emergency braking is executed.

[0109] S15: Based on the active obstacle avoidance decision model, obtain the corresponding control strategies for the semi-trailer truck in the longitudinal and lateral directions, including:

[0110] The lateral control strategy for the semi-trailer truck is derived by maintaining the longitudinal speed at a low speed (3 km / h or 5 km / h), and using a fifth-order polynomial for the lane-changing trajectory model, with the analytical function y(t) as follows:

[0111]

[0112] In the above formula, t is the time for the semi-trailer truck train to change lanes. tol The time required for a semi-trailer truck train to change lanes; y la This refers to the lateral displacement of the intelligent vehicle train during lane changing.

[0113] Differentiating the above equation yields the required lateral acceleration 'a' for the semi-trailer truck train. v ():

[0114]

[0115] In the above formula, t is the time for the semi-trailer truck train to change lanes. tol The time required for a semi-trailer truck train to change lanes; y la For the lateral displacement of the intelligent vehicle train during the lane change process, a y (t) represents the lateral acceleration required for the semi-trailer truck train.

[0116] Furthermore, the MPC control algorithm is used for dynamic optimization, and the state-space equations of the semi-trailer train containing yaw rate are established as follows:

[0117]

[0118] In the above formula, β t ω is the lateral deflection angle of the tractor. t β is the yaw angle of the tractor unit. s For the side slip angle of the semi-trailer, ω s δ is the yaw angle of the semi-trailer, and δ is the front wheel steering angle. Find the derivative of the tractor's sideslip angle. Find the derivative of the yaw angle of the tractor. Find the derivative of the semi-trailer sideslip angle. This refers to the sway angle of the semi-trailer.

[0119] Predicted state deviation value:

[0120] x(k+1|k)=A co x(k)+B co u(k)

[0121] In the above formula, x(k) is the state deviation value, and u(k) is the control deviation value; A co B co The state-space matrix can be obtained through the kinematics of the intelligent semi-trailer truck train.

[0122] The optimization objective equation is established as follows:

[0123]

[0124] In the above formula, J(k) is the target value, Q and R are diagonal matrices, X is the state matrix, and U is the control matrix; the data dynamically optimized by the MPC control algorithm and the lateral acceleration a required by the semi-trailer truck are used. y ()Weighted analysis is performed to derive the corresponding control strategies in the horizontal direction.

[0125] Preferably, the control strategies for the semi-trailer truck train in the longitudinal and lateral directions are obtained based on the active obstacle avoidance decision model, including:

[0126] The longitudinal control strategy of the semi-trailer truck is derived by using the speed difference to establish the driving and braking torque equations of the semi-trailer truck, as follows:

[0127]

[0128] In the above formula, P m To solve for the proportional coefficient of the driving torque, I m To solve for the integral time coefficient of the driving torque, D m To solve for the differential time coefficient of the driving torque, T f To overcome rolling friction torque, the function v diff (t) is a preset function, the v diff (t) Using the safe driving zone centered on the semi-trailer truck as a reference, the speed difference between the speed of the semi-trailer truck and the desired speed of the semi-trailer truck is calculated according to the preset rules.

[0129] Specifically, the proportional coefficient P for solving the driving torque... m for:

[0130] P m =0+ pm ΔP m + pv ΔP v + ps ΔP s ;

[0131] In the above formula, P0 is the initial proportional value, and k pm ΔP is the proportional load mass factor. mk represents the proportional load capacity value. pv ΔP is the proportional velocity coefficient. v This is the proportional velocity value; k ps ΔP is the proportional rotation coefficient. s This is the proportional angle value;

[0132] The integral time coefficient I for solving the driving torque m for:

[0133] I m =0+ Im ΔI m + Iv ΔI v + Is ΔI s ;

[0134] In the above formula, I0 is the initial integral value, and k Im The integral load mass coefficient, ΔI m k represents the integral load capacity value. Iv The integral velocity coefficient, ΔI v k is the integral velocity value. Is Let ΔI be the integral rotation coefficient. s This is the integral angle value;

[0135] The differential time coefficient D of the driving torque solution m for:

[0136] D m =0+ Dm ΔD m + Dv ΔD v + Ds ΔD s ;

[0137] In the above formula, D0 is the initial differential value, and k Dm Let ΔD be the differential load mass coefficient. m k is the differential load mass value. Dv Let ΔD be the differential velocity coefficient. v k is the differential velocity value. Ds Let ΔD be the differential rotation coefficient. s This is the differential angle value.

[0138] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0139] Furthermore, the embodiments of the present invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An active obstacle avoidance control method for intelligent semi-trailer trucks, characterized in that, The method includes: Information about the surrounding environment is obtained based on sensors installed on the semi-trailer truck train; The surrounding environment information is preprocessed, and the obstacle outline region around the semi-trailer truck is constructed based on the preprocessed surrounding environment information. A safe driving zone centered on the semi-trailer truck is constructed based on the outline region of the obstacle. Based on the drivable safe zone, an active obstacle avoidance decision model is established for the semi-trailer truck. Based on the active obstacle avoidance decision model, the corresponding control strategies for the semi-trailer truck train in the longitudinal and lateral directions are obtained. Based on the corresponding control strategies of the semi-trailer train in the longitudinal and lateral directions, the longitudinal speed and lateral movement direction of the semi-trailer train are autonomously controlled. Based on the outline region of the obstacle, a safe driving zone centered on the semi-trailer truck is constructed, including: Based on the actual dimensions and outlines of the semi-trailer tractor and semi-trailer, the junction angle between the semi-trailer tractor and semi-trailer is determined. Calculated as; ； In the above formula, The articulation angle between the tractor and the semi-trailer in an intelligent semi-trailer truck train; The tractor's speed is determined by a speed sensor. The instantaneous turning angle of the tractor unit is obtained from the turning angle sensor; This refers to the wheelbase of the semi-trailer truck tractor. This refers to the wheelbase of the semi-trailer. The safe distance between the semi-trailer and the obstacle is determined based on the junction angle between the semi-trailer tractor and the semi-trailer and the outline area of ​​the obstacle around the semi-trailer. Based on the size and safety interval of the semi-trailer truck, a safe driving zone is constructed by extending outward from the obstacle outline area with the semi-trailer truck as the center. The safe driving area is updated in real time based on changes in the position or shape of obstacles and changes in the angle of intersection between the semi-trailer tractor and the semi-trailer.

2. The intelligent semi-trailer truck train active obstacle avoidance method as described in claim 1, characterized in that, The sensors include a camera, a lidar, a millimeter-wave radar, and an ultrasonic radar; wherein the lidar is installed on the windshield of the semi-trailer tractor; the millimeter-wave radar is installed on the semi-trailer tractor, on the sides and rear of the semi-trailer; the ultrasonic radar is installed around the bottom of the semi-trailer tractor and the semi-trailer; and the camera is installed at the bottom of the windshield of the semi-trailer tractor, on the left and right rearview mirrors, on the top of the driver's cab, and on the door of the semi-trailer tractor at the rear of the semi-trailer. The camera is used to acquire a pixel array of RGB images around the semi-trailer truck train; The lidar is used to acquire 3D point cloud distance information around the semi-trailer truck train; The millimeter-wave radar is used to acquire point data of objects around the semi-trailer truck train. The ultrasonic radar is used to acquire radar data of objects around the semi-trailer truck train.

3. The intelligent semi-trailer truck train active obstacle avoidance method as described in claim 2, characterized in that, The data preprocessing of the surrounding environment information includes: The pixel arrays of the RGB images obtained from each camera are sequentially stitched, denoised, smoothed, and transformed to obtain the pixel arrays of the processed RGB image. Extract and select pixel features from the pixel array of the processed RGB image; Based on the pixel features, obstacle recognition processing is performed to obtain a first set of recognized obstacles corresponding to the pixel features; The 3-D point cloud distance information acquired by each set lidar is processed by removing the ground point cloud based on the depth data to obtain the processed 3-D point cloud distance information; A clustering algorithm based on Euclidean distance is used to cluster the distance information of the processed 3-D point cloud, and a kd-tree is constructed based on the clustering results. Search each node of the kd-tree and obtain the set of point clouds where the number of nodes is less than a preset threshold; Based on the aforementioned point cloud, a second set of obstacles for LiDAR identification is obtained; The point data of surrounding objects acquired by each millimeter-wave radar is subjected to point filtering and noise reduction processing to obtain noise-reduced point data. The noise-reduced dot data is processed to obtain the third set of recognized obstacles corresponding to the noise-reduced dot data. The radar data obtained from each ultrasonic radar is denoised to obtain denoised radar data. The distance between the semi-trailer train and the obstacle is obtained based on the noise-reduced radar data.

4. The intelligent semi-trailer truck train active obstacle avoidance method as described in claim 1, characterized in that, Based on the preprocessed surrounding environment information, the obstacle outline region around the semi-trailer truck train is constructed, including: The preprocessed surrounding environment information is comprehensively analyzed using OR and Boolean operations to obtain obstacle information around the semi-trailer train. The obstacle information includes obstacle type, obstacle position relative to the semi-trailer train, and obstacle distance relative to the semi-trailer train. The obstacle outline region around the semi-trailer truck is constructed based on the obstacle information around the semi-trailer truck.

5. The intelligent semi-trailer truck train active obstacle avoidance method as described in claim 1, characterized in that, The safety interval includes the critical obstacle avoidance distance for semi-trailer braking and the expected safety distance; ； In the above formula, Critical obstacle avoidance distance for trailer braking To achieve the desired safe distance, For the speed of semi-trailer truck trains, The instantaneous velocity of the obstacle; The maximum permissible deceleration for semi-trailer truck trains, For braking system delay time, To set the workshop time interval, The road surface adhesion coefficient, To reserve the desired safe distance, It is a proportional function of the road adhesion coefficient.

6. The active obstacle avoidance method for intelligent semi-trailer trucks as described in claim 5, characterized in that... The proportional function of the road adhesion coefficient for ； In the above formula, The coefficient of adhesion for icy and snowy roads; This is the adhesion coefficient of the asphalt pavement.

7. The intelligent semi-trailer truck train active obstacle avoidance method as described in claim 5, characterized in that, The active obstacle avoidance decision model established based on the drivable safe area for the semi-trailer truck includes: The obstacle avoidance strategies include, but are not limited to: when there is only an obstacle in front of the semi-trailer truck and the distance to the obstacle reaches the critical obstacle avoidance distance for the semi-trailer truck to brake, and emergency braking is not timely enough to take a double-sided lane change strategy. When there are obstacles in front of the semi-trailer truck and obstacles on both sides, and the distance between the obstacles and the semi-trailer truck is within the expected safe distance, a following strategy should be adopted. When there are obstacles in front of the semi-trailer truck and obstacles on both sides, and the distance between the obstacles is close and the semi-trailer truck reaches the critical obstacle avoidance distance for braking, an emergency braking strategy should be adopted.

8. The intelligent semi-trailer truck train active obstacle avoidance method as described in claim 1, characterized in that, Based on the active obstacle avoidance decision model, the corresponding control strategies for the semi-trailer truck in the longitudinal and lateral directions are obtained, including: The method for deriving the corresponding control strategy for the semi-trailer truck in the lateral direction includes maintaining the longitudinal speed at a low speed, using a fifth-order polynomial for the lane-changing trajectory model, and its analytical function... for: ； In the above formula, t is the time it takes for the semi-trailer truck train to change lanes. The time required for a semi-trailer truck train to change lanes; This refers to the lateral displacement of the intelligent vehicle train during lane changing. Differentiating the above equation yields the lateral acceleration required for the semi-trailer truck train. : ； In the above formula, t is the time it takes for the semi-trailer truck train to change lanes. The time required for a semi-trailer truck train to change lanes; For the lateral displacement of intelligent vehicle trains during lane changing, This refers to the lateral acceleration required for a semi-trailer truck train.

9. The intelligent semi-trailer truck train active obstacle avoidance method as described in claim 1, characterized in that, Based on the active obstacle avoidance decision model, the corresponding control strategies for the semi-trailer truck in the longitudinal and lateral directions are obtained, including: The longitudinal control strategy of the semi-trailer truck is derived by using the speed difference to establish the driving and braking torque equations of the semi-trailer truck, as follows: ； In the above formula, To determine the proportional coefficient for the driving torque, Solve for the integral time coefficient of the driving torque. Solve for the differential time coefficients of the driving torque. To overcome rolling friction torque, the function For a preset function, the Using the safe driving zone centered on the semi-trailer truck as a reference, and setting it according to preset rules, the speed difference between the speed of the semi-trailer truck and the desired speed of the semi-trailer truck is calculated. The proportional coefficient for solving the driving torque for: ； In the above formula, This is the initial ratio value. This is the proportional load mass factor. This represents the proportional load capacity value. The proportional velocity coefficient, This is the proportional speed value; This is the proportional rotation coefficient. This is the proportional angle value; The integral time coefficient for solving the driving torque for: ; In the above formula, The initial integral value, This is the integral load capacity coefficient. This represents the integral load capacity value. The integral velocity coefficient, This is the integral speed value; The integral rotation coefficient, This is the integral angle value; The differential time coefficient of the driving torque solution for: ; In the above formula, The initial differential value, This is the differential load mass coefficient. This is the differential load mass value; The differential velocity coefficient, The value is the differential velocity. The differential rotation coefficient, This is the differential angle value.

Images