All-wheel steering vehicle parallel parking path planning method for narrow parking space

Abstract

The invention relates to an all-wheel steering vehicle parallel parking path planning method for a narrow parking space. According to vehicle parameters and the parallel parking space size and road information detected by a sensor, a parallel parking scene of the all-wheel steering vehicle is established, initial parking point information of vehicle positioning is input, a designed algorithm is executed, and an existing safe parking path is detected in real time. Distance information is sampled in real time to generate a series of parking paths, the paths are dispersed, road constraints and vehicle dynamics constraints needing to be met in the parking process are defined, a path evaluation function is designed, the paths meeting the constraints are evaluated, and the paths are optimized. According to the method, the optimal design of the parallel parking path of the multi-section type multi-steering mode vehicle is completed, the more flexible parking path design of the all-wheel steering vehicle is achieved, the parking space occupied in the parking process is reduced, and the parking efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of all-wheel steering vehicle path planning, in particular to a parallel parking path planning method for all-wheel steering vehicles in narrow parking spaces. BACKGROUND

[0002] With the continuous development and popularization of intelligent vehicle technology, all-wheel steering vehicle path planning has become a research hotspot of the times. All-wheel steering vehicle path planning technology enables intelligent vehicles to safely, efficiently and comfortably navigate obstacles in different traffic scenarios, which is of great value in reducing traffic accidents and improving vehicle ride comfort and safety.

[0003] In recent years, with the increasing per capita vehicle ownership in cities, urban road congestion has become increasingly severe, and the space available for vehicles during parking operations is becoming increasingly narrow. Due to the size of the narrow parking space, the vehicle cannot complete the parking operation in one go, and the driver often needs to make multiple adjustments to successfully park the vehicle, which not only increases the parking time but also easily causes traffic congestion. Therefore, a parallel parking path planning method for all-wheel steering vehicles in narrow parking spaces is needed to improve parking efficiency while ensuring safety.

[0004] Most existing parallel parking path planning methods use geometric curve methods for path planning, including two-arc, arc-line, polynomial curve, spiral curve, B-spline curve, and sine function curve, or a combination of the above methods to design a multi-segment parking path to find a collision-free path for parking. As described in patent 202011594837.5, A* algorithm is used to plan a continuous curve in two-arc mode, considering the minimum parking distance and minimum turning radius, and combining collision conditions to obtain the optimal path. A parking model is established and a polynomial function relationship is established based on the vehicle parallel parking path, and a polynomial function path planning curve is obtained based on vehicle motion constraints and collision constraints, as described in patent 202310109858.0, a hyperbolic tangent tanh function is used to obtain a hyperbolic tangent path planning curve, and a parallel parking path planning curve is obtained by combining the polynomial function path planning curve and the hyperbolic tangent path planning curve. These parallel parking path planning methods help improve the efficiency of autonomous vehicle parking, but still have some deficiencies and limitations, mainly in that:

[0005] 1. The existing parallel parking path planning method mainly takes the automatic driving front axle steering vehicle as the research object, but fails to combine the path planning algorithm with the advantages of multiple steering modes of the all-wheel steering vehicle, and the existing parallel parking path planning method is poor in adaptability to the all-wheel steering vehicle with long body and heavy load, has low parking efficiency, and cannot meet the parking situation facing narrow parking spaces. Therefore, it is necessary to face narrow parking spaces, combine the advantages of multiple steering modes, adopt the parallel parking path planning suitable for the multiple steering mode combination of the all-wheel steering vehicle, and improve the parking efficiency.

[0006] 2. The existing parallel parking path planning method only considers the geometric relationship between the motion of the whole vehicle and the regular parking space, and lacks transformation processing of scattered and dynamic obstacle environment. In the urban parking space environment, the vehicle parking is not only affected by static obstacles, but also affected by dynamic obstacles. If the existing parallel parking path planning method is still used, dynamic obstacle avoidance cannot be effectively realized, the vehicle parking failure is easy to cause, and the safety of the parking path quality is poor. Therefore, it is necessary to face the all-wheel steering vehicle, introduce the influence of dynamic obstacles, realize the parallel parking path planning of dynamic obstacle avoidance, and improve the safety of the parking path quality SUMMARY

[0007] The purpose of the present application is to provide a parallel parking path planning method for an all-wheel steering vehicle facing a narrow parking space, so as to improve the safety and efficiency of the parallel parking of the all-wheel steering vehicle.

[0008] To achieve the above purpose, the technical scheme of the present application is: a parallel parking path planning method for an all-wheel steering vehicle facing a narrow parking space, comprising the following steps:

[0009] Step S1, according to the all-wheel steering vehicle parameters, a vehicle shape model is designed, and the parallel parking space size and road information detected by the sensor are set, the boundary of the parking space and the dynamic obstacle are set, and the parallel parking scene of the all-wheel steering vehicle is established;

[0010] Step S2, in the way of directly giving the starting parking point, the vehicle is driven to the parking position, the pre-parking preparation is completed, then the parking space size is detected, and the designed parallel parking path planning algorithm is executed, and it is detected in real time whether there is a safe parallel parking path with the current vehicle pose as the starting parking point;

[0011] Step S3: For ease of path design, parking is considered the reverse process of vehicle exiting the parking space. A segmented approach is used to divide the parking path into three segments: The first segment involves the vehicle reversing to a safe distance behind the parking space; the second segment uses a crab-steering mode, where all wheels turn at the same angle to allow the vehicle to move laterally without collision; the third segment first uses a stationary turning mode followed by a DWA obstacle avoidance mode; the vehicle completes the exit movement using a combination of these three paths.

[0012] Step S4: Generate the second path segment using the crab steering mode, combine it with the first path segment, and design the third path segment based on this. Extract a series of discrete points in the designed path, calculate the X and Y positions of each point, the path slope parameters, the current vehicle speed, the distance to the obstacle, and the navigation angle of the vehicle relative to the given starting parking point, and store the data.

[0013] Step S5: Establish the conditions for road constraints and vehicle kinematic constraints, process the obtained series of path clusters, and filter out the paths that satisfy the constraints.

[0014] Step S6: Design a path evaluation function to evaluate and select the best path that meets the constraints. This path is then used as the optimal parking path. Calculate the vehicle's heading angle, navigation angle from the target point, turning angle of each wheel, vehicle speed, and distance from obstacles for each process. Input these values ​​into the established vehicle shape module model to complete the planning of the parallel parking motion.

[0015] Furthermore, S1 specifically involves: designing a road environment, parallel parking spaces, and three dynamic obstacles, and adding a vehicle shape program module model to this design to establish a parallel parking scenario for a three-axle vehicle with all-wheel steering. The methods for establishing the obstacle model, the vehicle shape model including the vehicle body and tires, and the vehicle kinematic parameter settings are as follows:

[0016] Vehicle body shape model creation:

[0017]

[0018] The steering center of the i-th axis tire and the shapes of the left and right tires are established:

[0019]

[0020] Establish a dynamic obstacle model:

[0021]

[0022] Vehicle kinematic parameter settings:

[0023]

[0024] In the formula, A, B, C, and D represent four external shape points of the vehicle, x and y are the coordinates of the vehicle's center of mass, θ is the vehicle's heading angle, L1 and L3 are the vector distances between the center of mass and the first and third axes, respectively, and L... f L r Here, represents the front and rear overhang distances of the vehicle, W is the vehicle width, tyre_il(x,y) is the steering center of the left tire on the i-th axis, [tyre_ir(x,y)] is the steering center of the right tire on the i-th axis, tyre_ilA(x,y), tyre_ilB(x,y), tyre_ilC(x,y), tyre_ilD(x,y) are the corresponding outline points of the left tire on the i-th axis, tyre_irA(x,y), tyre_irB(x,y), tyre_irC(x,y), tyre_irD(x,y) are the corresponding outline points of the right tire on the i-th axis, l is the tire length, w is the tire width, and L... i Let δ be the vector distance between the centroid and the i-th axis. i Let be the turning angle of the tire on the i-th axis. i Here are the obstacle coordinates, `obstacle_R` is the obstacle radius, `vel_obstacle` is the obstacle's movement speed, and `flag_obstacle` is the obstacle's movement time. i (Y i () represents the vertical coordinate of the obstacle. Maximum permissible speed for vehicles The maximum permissible angular velocity of the vehicle, a max The maximum permissible acceleration value of the vehicle, β max The maximum permissible angular acceleration value for the vehicle.

[0025] Furthermore, step S2 specifically involves using the actual detected coordinate parameters as input to a multi-segment path planning algorithm at the vehicle pre-parking point. Parking is equivalent to the reverse process of vehicle exiting the parking space and is divided into three parking paths for design. The algorithm detects whether a safe parking path exists. If no safe parking path is found, the evaluation parameter weights are adjusted until a safe parallel parking path is planned.

[0026] Furthermore, the first segment of the path completes the task of reversing the vehicle to a predetermined safe distance behind the parking space, leaving more room for subsequent path design. Its path planning equation is:

[0027]

[0028] In the formula, P0(X0,Y0) is the expected endpoint after safe parking, P1(X1,Y1) is the endpoint coordinate of the first path segment, and L is the vehicle length.safe To maintain a safe distance from the back of the parking space, L p This refers to the length of the parking space.

[0029] Furthermore, the second path segment employs a crab-like turning mode. A pre-set path slope parameter is used, and distance parameters under crab-like turning mode are sampled in real-time to obtain different second path segments. The path planning equation is as follows:

[0030]

[0031] In the formula, P1(X1,Y1) is the starting coordinate of the second path segment, P2(X2,Y2) is the ending coordinate of the second path segment, and K2 and S2 are the slope and driving distance of the crab steering mode path.

[0032] Furthermore, the third segment of the path first employs a stationary turning mode, where the vehicle performs a zero-radius turn around its geometric center at P3(X3,Y3), thereby changing the vehicle's heading angle. This provides suitable heading angle parameters for the DWA obstacle avoidance path design and prepares the vehicle accordingly.

[0033]

[0034] In the formula, P2(X2,Y2) is the coordinate information of the end point of the second path, P3(X3,Y3) is the coordinate information of the vehicle after turning around in place, θ3 is the heading angle of the vehicle after turning around in place, θ2 is the heading angle of the vehicle before turning around in place, and θ is the given turning angle in place.

[0035] Furthermore, after the third path segment, the DWA obstacle avoidance mode is adopted. Without considering the constraints on the vehicle's own speed and angular velocity, the theoretical minimum value of the vehicle's instantaneous speed is calculated within time dt1, combining the current actual vehicle speed V3, angular velocity ω3, and vehicle kinematic parameters a3 and β3. and maximum value and the theoretical minimum value of angular velocity and maximum value The following is a detailed calculation process for the vehicle parameters:

[0036]

[0037] In the formula, dt1 is the time parameter for calculating the vehicle speed and angular velocity values. It is the theoretical minimum value of the vehicle's instantaneous speed. This is the theoretical maximum value of the vehicle's instantaneous speed. The vehicle's current acceleration value a3 is between 0 and a max Get the winning number, a max This is the maximum permissible acceleration value for the vehicle; It is the theoretical minimum value of the vehicle's instantaneous angular velocity. This is the theoretical maximum value of the vehicle's instantaneous angular velocity, and the vehicle's current angular acceleration value β3 is between 0 and β. max Take from the middle, β max This is the maximum permissible angular acceleration value for the vehicle;

[0038] Furthermore, after the third path segment, the DWA obstacle avoidance mode is adopted. Considering the constraints of the vehicle's own speed and angular velocity, all possible actual values ​​of the vehicle's instantaneous speed V4 and angular velocity ω4 are calculated. The total time t for forward simulation of local path planning and the time dt for forward simulation of local path point planning are set. Several simulated local paths are planned forward for each vehicle motion parameter V4 and ω4. The detailed planning process for each forward simulated local path is as follows:

[0039]

[0040] In the formula, V 4min It is the actual minimum value of the vehicle's speed at the next moment, V. 4max It is the actual maximum speed of the vehicle at the next moment, and it is set in increments of 0.01 in V. 4min and V 4max The vehicle obtains all possible instantaneous speed values ​​V4. ω represents the maximum permissible speed of the vehicle. 4min It is the actual minimum value of the vehicle's instantaneous angular velocity, ω. 4max It is the actual maximum value of the vehicle's instantaneous angular velocity, expressed in steps of 1 at ω. 4min and ω 4max The instantaneous angular velocity values ​​ω4 that the vehicle can obtain are obtained in the middle. n1 is the maximum allowable angular velocity of the vehicle, n2 is the number of iterations for simulating local path points forward, P3(X3) and P3(Y3) are the x and y coordinates of the starting point of the third segment of the vehicle's path, P4(X4) and P4(Y4) are the x and y coordinates of the ending point of the third segment of the vehicle's path, θ3 and θ4 are the heading angles of the vehicle at the starting and ending points of the third segment of the path, V4 is the instantaneous vehicle speed, and ω4 is the instantaneous angular velocity of the vehicle.

[0041] Furthermore, weight values ​​are set for each evaluation parameter, and the simulated local paths obtained after each planning iteration are evaluated and analyzed. The local path with the highest score is the optimal local path, and the combination of all optimal local paths forms the final path to the vehicle's parking starting point. The evaluation process is as follows:

[0042] The evaluation parameters are:

[0043] E_part=[Heading,dist,Vpoint]

[0044] The evaluation function is:

[0045]

[0046] In the formula, Heading is the heading score for the predicted destination, dist is the score for the nearest obstacle to the predicted destination, Vpoint is the speed score for the predicted destination, Epoint is the total score for the predicted destination, θ4 is the vehicle heading angle at the end of the third path, P5(X5) and P5(Y5) are the x and y coordinates of the vehicle's parking starting point, P4(X4) and P4(Y4) are the x and y coordinates of the end of the third path, and obstacle_R is the radius of the obstacle. i (X i ,Y i () represents the coordinate information of the obstacle.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1) This invention solves the problem of low parking efficiency and inability to meet the requirements of narrow parking spaces for all-wheel-steering vehicles in parallel parking scenarios. Addressing the shortcomings of existing parallel parking path planning methods, this invention sets up a parallel parking scenario with narrow parking spaces. Leveraging the advantages of multiple steering modes of all-wheel-steering vehicles and considering safety requirements such as collision constraints, a combination of crab steering and stationary turning modes is selected to design a parallel parking planning method for all-wheel-steering vehicles in narrow parking spaces. This method enables parallel parking path planning that meets the requirements of narrow parking spaces, improving parking efficiency while ensuring safety during the parking process.

[0049] 2) This invention addresses the problem of reduced parking efficiency or even failure caused by the safety requirements of dynamic obstacle avoidance in parallel parking path planning methods. To address the shortcomings of existing parallel parking path planning methods, this invention introduces the influence of dynamic obstacles on the vehicle in parallel parking scenarios. Considering various parameters such as the distance between the vehicle and the dynamic obstacle, and the vehicle's speed, a method suitable for dynamic parking path planning of all-wheel-steering vehicles is designed. This enables parallel parking planning with dynamic obstacle avoidance, improves dynamic parking efficiency, and ensures the safety of the parking path quality. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of a parallel parking scenario for a three-axle vehicle with all-wheel steering, oriented towards a narrow parking space, according to an embodiment of the present invention.

[0051] Figure 2 This is a schematic diagram of the parallel parking path design process for all-wheel steering vehicles in narrow parking spaces according to an embodiment of the present invention.

[0052] Figure 3This is a schematic diagram showing the relationship between the turning radius and the turning angle in the DWA obstacle avoidance mode of this invention.

[0053] Figure 4 This is a schematic diagram of the upper and lower road boundary constraints in an embodiment of the present invention.

[0054] Figure 5 This is a schematic diagram illustrating the generation and extraction of local path clusters that satisfy road and vehicle kinematic constraints in an embodiment of the present invention.

[0055] Figure 6 This is a diagram illustrating the optimal parking path that enables dynamic obstacle avoidance in an embodiment of the present invention. Detailed Implementation

[0056] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings.

[0057] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, 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 pertains. It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application.

[0058] Figure 1 This is a schematic diagram illustrating a parallel parking scenario for a three-axle vehicle with all-wheel steering in a narrow parking space, as described in this embodiment. Based on road conditions, parking space dimensions, and dynamic obstacle information, a road with a width of d and a length of L is designed. p Width is W p We identified parking spaces and added a program model of the vehicle's shape to them, creating a parallel parking scenario for a three-axle vehicle with all-wheel steering, designed for narrow parking spaces.

[0059] Figure 2 This is a schematic diagram of the parallel parking path design process for all-wheel steering vehicles in narrow parking spaces according to an embodiment of the present invention. It shows the overall multi-segment path design method and the overall implementation process of selecting the optimal local planning path and generating the parking path.

[0060] Figure 3 This is a schematic diagram illustrating the relationship between the turning radius and the turning angle in the DWA obstacle avoidance mode of this invention embodiment. In particular, it shows the geometric relationship between the turning radius and the first axis turning angle, and the connection between the heading angle and the DWA local path planning equations, thereby enabling local path planning.

[0061] Figure 4 This is a schematic diagram illustrating the upper and lower road boundary constraints in an embodiment of the present invention. Vehicles should avoid collisions with road boundaries and parking space boundaries during parking.

[0062] Figure 5 This is a schematic diagram illustrating the generation and extraction of local path clusters that satisfy road and vehicle kinematic constraints in an embodiment of the present invention. Based on this, an evaluation function evaluates all local path clusters, selecting the path with the highest evaluation score as the optimal local path, thus preparing for the subsequent optimal local path combination curve.

[0063] Figure 6 This is a diagram illustrating the optimal parking path for dynamic obstacle avoidance in an embodiment of the present invention. The first path completes the vehicle's reverse maneuver. The second path uses given parameters K2, S2, etc., to perform crab steering and park in a suitable position in the lateral direction. The third path first uses a stationary turning mode and then a DWA obstacle avoidance mode, combining all the optimal local paths to obtain the optimal parking path to drive to the given starting parking point, thus achieving dynamic obstacle avoidance.

[0064] In this embodiment, all parameters, including those for road parking scenarios and all-wheel steering vehicles, are as follows:

[0065]

[0066]

[0067] This embodiment provides a parallel parking path planning method for all-wheel steering vehicles in narrow parking spaces that enables dynamic obstacle avoidance, including the following steps:

[0068] Step S1: Design the vehicle shape model based on the parameters of the all-wheel steering vehicle, as well as the parallel parking space size and road information detected by the sensors, set the boundary of the parking space and dynamic obstacles, and establish the parallel parking scenario of the all-wheel steering vehicle.

[0069] Step S2: Using the method of directly giving the starting parking point, drive the vehicle to the parking position to complete the pre-parking preparation. Then, detect the parking space size and execute the designed parallel parking path planning algorithm. In real time, detect whether there is a safe parallel parking path with the current vehicle pose as the starting parking point. If not, adjust the evaluation parameter weight value and re-plan the parking path.

[0070] Step S3: Considering the convenience of the designed path, the parking process is equivalent to the reverse process of vehicle exiting the parking space. The parking path is divided into a combination of two straight lines and one curve. The first straight line provides more space for subsequent parking path planning; the second straight line allows for collision-free parking in a suitable position in the lateral direction; the third curve first inputs a given in-situ turning angle parameter, and then performs an in-situ turn based on this parameter to change the vehicle's heading angle, preparing for better use of the DWA obstacle avoidance mode for parking path planning later.

[0071] Step S4: Consider the boundary collision constraints of the vehicle with the parallel parking space and the surrounding road, and establish the road constraint conditions to avoid collisions between the vehicle and the road boundary and the parallel parking space boundary.

[0072] Step S5: First, generate local path clusters for the vehicle. Then, combine the collision constraints to generate and extract local path clusters that satisfy the road and vehicle kinematic constraints, in order to prepare for the subsequent evaluation of the local path clusters.

[0073] Step S6: Design a path evaluation function to evaluate and select the best path that satisfies the constraints. This best path is then used as the optimal local parking path. The heading angle, navigation angle from the target point, turning angle of each wheel, vehicle speed, and distance from obstacles are calculated for each process. These values ​​are then input into the established vehicle shape model. Finally, all optimal local parking paths are combined to complete the planning of the parallel parking motion.

[0074] In step S1, based on the road conditions, parking space dimensions, and dynamic obstacle information, a road with a width of d and a length of L is designed. p Width is W p The parking space has initial coordinates of obstacle. i (X i ,Y i To address dynamic obstacles, a three-axle vehicle's external shape model was designed, and a parallel parking scenario for the all-wheel-steering three-axle vehicle was established. The methods for establishing the obstacle model, the vehicle's external shape model including the vehicle body and tire shapes, and the vehicle's kinematic parameters are as follows:

[0075] Vehicle body shape model creation:

[0076]

[0077] The steering center of the i-th axis tire and the shapes of the left and right tires are established:

[0078]

[0079] Establishing a dynamic obstacle model:

[0080]

[0081] Vehicle kinematic parameter settings:

[0082]

[0083] In the formula, A, B, C, and D represent four external shape points of the vehicle, x and y are the coordinates of the vehicle's center of mass, θ is the vehicle's heading angle, L1 and L3 are the vector distances between the center of mass and the first and third axes, respectively, and L... f L rHere, represents the front and rear overhang distances of the vehicle, W is the vehicle width, tyre_il(x,y) is the steering center of the left tire on the i-th axis, [tyre_ir(x,y)] is the steering center of the right tire on the i-th axis, tyre_ilA(x,y), tyre_ilB(x,y), tyre_ilC(x,y), tyre_ilD(x,y) are the corresponding outline points of the left tire on the i-th axis, tyre_irA(x,y), tyre_irB(x,y), tyre_irC(x,y), tyre_irD(x,y) are the corresponding outline points of the right tire on the i-th axis, l is the tire length, w is the tire width, and L... i Let δ be the vector distance between the centroid and the i-th axis. i Let be the turning angle of the tire on the i-th axis. i Here are the obstacle coordinates, `obstacle_R` is the obstacle radius, `vel_obstacle` is the obstacle's movement speed, and `flag_obstacle` is the obstacle's movement time. i (Y i () represents the vertical coordinate of the obstacle. Maximum permissible speed for vehicles The maximum permissible angular velocity of the vehicle, a max The maximum permissible acceleration value of the vehicle, β max The maximum permissible angular acceleration value for the vehicle.

[0084] In step S2, the vehicle starts from the given actual parking point, and the path planning algorithm detects whether there is a safe parking path. If it shows that there is no safe parking path, the evaluation parameter weights are adjusted until a safe parallel parking path is planned.

[0085] In step S3, for the convenience of designing the path, parking is equivalent to the reverse process of a vehicle leaving the parking space. The parking path is divided into three segments using a segmented method: the first segment completes the task of reversing the vehicle to the set safe distance behind the parking space; the second segment uses a crab steering mode, where all wheels turn at the same angle to enable the vehicle to move a distance laterally without collision; the third segment first uses a stationary turning mode and then a DWA obstacle avoidance mode.

[0086]

[0087] In the formula, P0(X0,Y0) is the expected endpoint after safe parking, P1(X1,Y1) are the endpoint coordinates of the first path segment and the starting coordinates of the second path segment, and L is the vehicle length. safe To maintain a safe distance from the back of the parking space, L pP2(X2,Y2) represents the length of the parking space, P2(X2,Y2) represents the coordinates of the end point of the second path segment, K2 and S2 represent the slope and travel distance of the crab steering mode path, P3(X3,Y3) represents the coordinate information of the vehicle after turning in place, θ3 represents the heading angle of the vehicle after turning in place, θ2 represents the heading angle of the vehicle before turning in place, and θ represents the given turning angle.

[0088] In step S4, since the third path segment in step S3 adopts the DWA obstacle avoidance mode, combined with the real-time vehicle motion parameters, the real-time local path planning is obtained. The real-time local path consists of a series of discrete points, including the X and Y positions of the discrete points, the real-time vehicle speed, the distance to dynamic obstacles, and other information. The following is a detailed process of path discretization.

[0089]

[0090] Where dt1 is the time parameter for calculating the vehicle speed and angular velocity values. It is the theoretical minimum value of the vehicle's instantaneous speed. This is the theoretical maximum value of the vehicle's instantaneous speed. The vehicle's current acceleration value a3 is between 0 and a max win; It is the theoretical minimum value of the vehicle's instantaneous angular velocity. This is the theoretical maximum value of the vehicle's instantaneous angular velocity, and the vehicle's current angular acceleration value β3 is between 0 and β. max Take from the middle; V 4min It is the actual minimum value of the vehicle's instantaneous speed, V. 4max It is the actual maximum instantaneous speed of the vehicle, expressed in increments of 0.01 in V. 4min and V 4max Take the vehicle's instantaneous speed value V4; ω 4min It is the actual minimum value of the vehicle's instantaneous angular velocity, ω. 4max It is the actual maximum value of the vehicle's instantaneous angular velocity, expressed in steps of 1 at ω. 4min and ω 4max The instantaneous angular velocity value of the vehicle is ω4; n1 is the number of iterations for simulating local path points forward, n2 is the number of local paths for simulating forward, P3(X3) and P3(Y3) are the x and y coordinates of the starting point of the third segment of the vehicle's path, P4(X4) and P4(Y4) are the x and y coordinates of the ending point of the third segment of the vehicle's path, θ3 and θ4 are the heading angles of the vehicle at the starting and ending points of the third segment of the path, V4 is the instantaneous vehicle speed value, and ω4 is the instantaneous angular velocity value of the vehicle.

[0091] In step S5, considering the boundary collision constraints between the vehicle and the parallel parking space, as well as the surrounding roads, road constraint conditions are established to prevent collisions between the vehicle and the road boundaries and the boundaries of the parallel parking space. The constraint conditions are:

[0092] Road boundary constraints:

[0093]

[0094] Parallel parking space boundary constraints:

[0095]

[0096] Among them, X A Y A X represents the x and y coordinates of point A, the right front boundary point of the vehicle; B Y B X represents the x and y coordinates of point B, the right rear boundary point of the vehicle; C Y C X represents the x and y coordinates of the left rear boundary point C of the vehicle; D Y D The x and y coordinates of the left front boundary point D of the vehicle are respectively.

[0097] In step S6, the weight values ​​of each evaluation parameter are set, and the forward simulated local paths obtained after each planning are evaluated and analyzed. The optimal local path with the highest score is selected, and all optimal local paths are combined to form the final path to the vehicle's parking starting point. The evaluation process is as follows:

[0098] The evaluation parameters are:

[0099] E_part=[Heading,dist,Vpoint]

[0100] The evaluation function is:

[0101]

[0102] Where Heading is the heading score for the predicted destination, dist is the score for the nearest obstacle to the predicted destination, Vpoint is the speed score for the predicted destination, Epoint is the total score for the predicted destination, θ4 is the vehicle heading angle at the end of the third path, P5(X5) and P5(Y5) are the x and y coordinates of the vehicle's parking starting point, P4(X4) and P4(Y4) are the x and y coordinates of the end of the third path, and obstacle_R is the radius of the obstacle. i (X i ,Y i () represents the coordinate information of the obstacle.

[0103] This method leverages the advantages of multiple steering modes in all-wheel steering vehicles, selecting a combination of crab steering mode, stationary turning mode, and DWA obstacle avoidance mode to design parking paths. This enables dynamic obstacle avoidance parallel parking planning for vehicles in narrow parking spaces, improving dynamic parking efficiency and ensuring the safety of parking path quality.

[0104] The above description, including numerical values ​​and definitions, is merely a preferred embodiment of the present invention and is not intended to limit the invention in any other way. Any person skilled in the art may make changes or modifications to the disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention, without departing from the scope of the present invention, shall still fall within the protection scope of the present invention.

[0105] The above description, including numerical values ​​and definitions, is merely a preferred embodiment of the present invention and is not intended to limit the invention in any other way. Any person skilled in the art may make changes or modifications to the disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention, without departing from the scope of the present invention, shall still fall within the protection scope of the present invention.

Claims

1. A parallel parking path planning method for a four-wheel steering vehicle facing a narrow parking space, characterized by, Comprise the following steps: Step S1, according to the all-wheel steering vehicle parameter design vehicle shape model, and the sensor detects the parallel parking space size and road information, the boundary of the parking space and the dynamic obstacle is set, the all-wheel steering vehicle parallel parking scene is established; Step S2, in the way of directly giving the starting parking point, the vehicle is driven to the parking position, the pre-parking preparation is completed, then the parking space size is detected, and the designed parallel parking path planning algorithm is executed, whether there is a safe parallel parking path with the current vehicle pose as the starting parking point is detected in real time; Step S3, the parking is equivalent to the reverse process of vehicle out of the warehouse, the parking path is divided into three sections for design: the first section path, the task of vehicle rear to the rear of the parking space set safety distance is completed, the second section path adopts crab steering mode, all wheels are turned with the same angle, the vehicle is moved in the lateral direction without collision for a distance;The third section path adopts the original place rotation mode first and then adopts the DWA obstacle avoidance mode;The vehicle completes the out-of-warehouse movement under the combination of the three section paths; Step S4, the second section path is generated through the crab steering mode, combined with the first section path, and on this basis, the subsequent path is designed, a series of discrete points in the designed path are extracted, the X, Y position, path slope parameter, current vehicle speed, distance from the obstacle, and vehicle relative to the given starting parking point navigation angle are calculated, and the data is stored; Step S5, the conditions of road constraint and vehicle kinematics constraint are established, and a series of path clusters are processed to select the path meeting the constraint; Step S6, a path evaluation function is designed, the path meeting the constraint is evaluated and optimized, the path is taken as the optimal parking path, the heading angle of the vehicle in each process and the turning angle data of each wheel are calculated, which are input into the established vehicle shape model to complete the planning of parallel parking motion.

2. The method of claim 1, wherein, The S1 is specifically: a road environment, a parallel parking space and a dynamic obstacle are designed, and on this basis, a vehicle shape model is added, and a parallel parking scene of a full-wheel steering three-axle vehicle is established, the vehicle shape model includes a vehicle body and a tire shape model, and a dynamic obstacle model, and a method for establishing the vehicle body and the tire shape model is: Vehicle body shape model establishment: The steering center of the i-th axle tire, and the shape of the left and right tires are established: Dynamic obstacle model establishment: In the formula, A, B, C, D represent four shape points of the vehicle, x and y are the coordinates of the vehicle mass center, θ is the vehicle heading angle, L1 and L3 are the vector distances of the mass center to the first and third axes respectively, L f , L r are the front and rear suspension distances of the vehicle respectively, W is the vehicle width, tyre_il(x, y) is the turning center of the left tire of the ith axis, [tyre_ir(x, y)] is the turning center of the right tire of the ith axis, tyre_ilA(x, y), tyre_ilB(x, y), tyre_ilC(x, y), tyre_ilD(x, y) are the corresponding shape points of the left tire of the ith axis, tyre_irA(x, y), tyre_irB(x, y), tyre_irC(x, y), tyre_irD(x, y) are the corresponding shape points of the right tire of the ith axis, l is the length of the tire, w is the width of the tire, L i is the vector distance of the mass center to the ith axis, δ i is the turning angle of the tire of the ith axis, obstacle i is the obstacle coordinate, obstacle_R is the obstacle radius, vel_obstacle is the obstacle moving speed, flag_obstacle is the obstacle moving time, obstacle i (Y i ) is the longitudinal coordinate of the obstacle.

3. The method of claim 1, wherein, The step S2 is specifically: at the vehicle pre-parking point, the actually measured coordinate information is taken as the input of the path planning algorithm, the parking is equivalent to the reverse process of vehicle out of the warehouse and is divided into three section parking paths for design, whether there is a safe parking path is detected, if it is shown that there is no safe parking path, the evaluation parameter weight is adjusted until a safe parallel parking path is planned.

4. The method of claim 3, wherein, The first section path completes the task of vehicle rear to the rear of the parking space set safety distance, and leaves more space for subsequent path design, and the path planning equation is: In the formula, P0(X0, Y0) is the expected end point after safe parking, P1(X1, Y1) is the end point coordinate of the first path, L is the length of the vehicle, L safe is the safe distance from the rear of the parking space, p is the length of the parking space.

5. The method of claim 3, wherein, The second segment path adopts crab turning mode, and the slope and the driving distance of the path in this mode are sampled to obtain different second segment paths, and the path planning equation is: In the formula, P1(X1, Y1) is the starting point coordinate of the second segment path, P2(X2, Y2) is the end point coordinate of the second segment path, K2 and S2 are the slope and the driving distance of the crab turning mode path.

6. The method of claim 3, wherein, The third segment path first adopts the original turning mode, and the vehicle realizes zero radius turning around the geometric center at the original P2(X2, Y2), so as to change the vehicle heading angle and prepare for subsequent path design; In the formula, P2(X2, Y2) is the end point coordinate information of the second segment path, P3(X3, Y3) is the coordinate information after the original turning of the vehicle, θ3 is the vehicle heading angle after the original turning of the vehicle, θ2 is the vehicle heading angle before the original turning of the vehicle, and θ is the given original turning angle.

7. The method of claim 3, wherein, The third section path is followed by DWA obstacle avoidance mode, without considering the vehicle speed and angular velocity constraints, combining the current actual vehicle speed V3 and angular velocity ω3 and the vehicle kinematics parameters a3 and β3, the theoretical minimum value of the instantaneous vehicle speed value within dt1 time is calculated and the maximum value and the theoretical minimum value of the angular velocity value and the maximum value The calculation equation is: where dt1 is a time parameter of calculating the vehicle speed value and the angular speed value, is a theoretical minimum value of the vehicle instantaneous speed value, is a theoretical maximum value of the vehicle instantaneous speed value, the vehicle current acceleration value a3 is taken in the range of 0 and a max , the vehicle allowed maximum acceleration value a max is determined by the vehicle kinematic parameters; is a theoretical minimum value of the vehicle instantaneous angular speed value, is a theoretical maximum value of the vehicle instantaneous angular speed value, the vehicle current angular acceleration value β3 is taken in the range of 0 and β max , the vehicle allowed maximum angular acceleration value β max is determined by the vehicle kinematic parameters.

8. The method of claim 3, wherein, The third segment path adopts DWA obstacle avoidance mode, and all actual possible values of the instantaneous vehicle speed V4 and the angular velocity ω4 of the vehicle are calculated under the constraint conditions of the vehicle speed and the angular velocity of the vehicle, and the calculation equation is: where V 4min is the actual minimum value of the vehicle instantaneous speed value, V 4max is the actual maximum value of the vehicle instantaneous speed, taking the value of the vehicle instantaneous speed value V4 in steps of 0.01 between V 4min and V 4max ; the vehicle allowed maximum speed value is determined by the vehicle kinematic parameters; ω 4min is the actual minimum value of the vehicle instantaneous angular velocity value, ω 4max is the actual maximum value of the vehicle instantaneous angular velocity, taken in steps of 1 in the range ω 4min and ω 4max ; the vehicle allowed maximum angular velocity value is determined by the vehicle kinematic parameters.

9. The method of claim 3, wherein, The third segment path adopts DWA obstacle avoidance mode, the total time t of forward simulation local path planning is set, the time dt of forward simulation local path point planning is set, and the forward simulation local path planning equation of each path is: In the formula, n1 is the number of forward simulation local path point planning cycles, n2 is the number of forward simulation local paths, P3(X3) and P3(Y3) are the starting point horizontal and vertical coordinates of the third segment path of the vehicle, P4(X4) and P4(Y4) are the end point horizontal and vertical coordinates of the third segment path of the vehicle, θ3 and θ4 are the vehicle heading angles at the starting point and the end point of the third segment path, V4 is the instantaneous vehicle speed, and ω4 is the instantaneous angular velocity of the vehicle.

10. The method of claim 3, wherein, The third segment path adopts DWA obstacle avoidance mode, the optimal local path is screened out by evaluating all the forward simulation local paths obtained by the evaluation function, and the road boundary constraint and the parallel parking space boundary constraint are established; the evaluation function and the constraint condition equation are: The evaluation equation is: The road boundary constraint condition equation is: The parallel parking space boundary constraint condition equation is: In the formula, Heading is a heading score of the forward predicted end point, dist is a distance nearest obstacle score of the forward predicted end point, Vpoint is a speed score of the forward predicted end point, Epoint is a total score of the forward predicted end point, θ4 is a vehicle heading angle at the end point of the third segment path, P5X5, P5Y5 are horizontal and vertical coordinates of a parking starting point of the vehicle, P4(X4), P4(Y4) are horizontal and vertical coordinates of the end point of the third segment path of the vehicle, obstacle_R is a radius length of an obstacle, obstacle i X i ,Y i is coordinate information of the obstacle, X A , Y A are respectively horizontal and vertical coordinates of a right front boundary point A of the vehicle; X B , Y B are respectively horizontal and vertical coordinates of a right rear boundary point B of the vehicle; XC, YC are respectively horizontal and vertical coordinates of a left rear boundary point C of the vehicle; XD, YD are respectively horizontal and vertical coordinates of a left front boundary point D of the vehicle.

Images