A method and system for selecting a target parking space based on optimal path planning

Abstract

The present invention discloses a method and system for selecting a target parking space based on optimal path planning, and relates to the technical field of automatic parking space selection. By adding the trajectory prediction influence of dynamic obstacles, adjusting the planning of parking trajectories and the selection of optimal trajectories and optimal target parking spaces, the safety, efficiency and predictability of parking are further improved. Situational judgment is also added to comprehensively consider the parking planning of fixed parking spaces and public parking spaces, which is more in line with the parking habits and convenience of drivers.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic parking space selection, and in particular to a method and system for selecting a target parking space based on optimal path planning. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the rapid growth of the number of motor vehicles in my country, problems such as urban congestion and tight transportation infrastructure resources have become increasingly severe. In particular, the demand for parking spaces has increased rapidly, resulting in small parking spaces and compact parking spaces. This has caused many parking problems such as frequent parking safety accidents, high technical difficulty, and long time consumption, which have seriously affected people's daily travel and driving experience.

[0004] As parking problems become increasingly prominent, related automatic parking technologies are also developing rapidly. The automatic parking system has greatly improved the safety and efficiency of parking, reduced the driving difficulty and intensity of the driver, improved the driving experience, and achieved safe, efficient and convenient parking.

[0005] The inventors have found that the current automatic parking system has the following problems: the traditional parking system mainly controls the vehicle to park by identifying a parking space around the vehicle body and sensing the surrounding environment, planning the parking trajectory, but less consideration is given to the optimal trajectory planning and optimal target parking space selection when multiple parking spaces are identified around the vehicle body, and the comprehensive optimal planning in the real parking environment is not achieved. At the same time, the current automatic parking system only senses and identifies parking spaces, obstacles, etc., and obtains environmental information, but does not consider the prediction of the future movement trajectory of dynamic obstacles, resulting in the vehicle being unable to adjust the path in real time according to the movement trajectory of the obstacle. Therefore, there is an urgent need for a method to achieve intelligent planning of the optimal path and optimal parking space selection in a real parking scenario to reduce the safety risks caused by dynamic obstacles and overcome the limitations of parking in a single target parking space. Summary of the invention

[0006] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a method and system for selecting a target parking space based on optimal path planning. By adding the trajectory prediction influence of dynamic obstacles, adjusting the planning of parking trajectories and the selection of optimal trajectories and optimal target parking spaces, the safety, efficiency and predictability of parking are further improved. Situational judgment is also added to comprehensively consider the parking planning of fixed parking spaces and public parking spaces, which is more in line with the driver's parking habits and convenience.

[0007] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0008] The first aspect of the present disclosure provides a method for selecting a target parking space based on optimal path planning, comprising the following steps:

[0009] Collect vehicle parameters and determine the vehicle's current environment based on the destination selection; after determination, identify the parking space and establish a spatial coordinate system;

[0010] Perceive obstacles in the environment and predict short-term trajectories of dynamic obstacles in real time;

[0011] Using the perception data, determine the initial posture of the vehicle and the relative position of the vehicle to each identified parking space;

[0012] Determine whether the parking conditions are met and determine the range of the first-level target parking space;

[0013] Plan the longitudinal speed, acceleration, and jerk curves of the vehicle during parking. Based on the planning results, select the parking trajectory strategy according to the initial posture of the vehicle and the relative position of the vehicle to each identified parking space;

[0014] Considering the impact of obstacles, select parking spaces that can successfully plan a safe path in the first-level target parking space range to form the second-level target parking space range;

[0015] In the range of the secondary target parking spaces, the optimal target parking space and parking trajectory are comprehensively selected based on the parking trajectory efficiency of each target parking space and its distance from the destination, and the vehicle is parked in the optimal target parking space according to the optimal parking trajectory.

[0016] Furthermore, the parking space recognition process is to use a camera to identify parking spaces with parking space lines, and to use an ultrasonic radar to identify parking spaces without parking space lines but with parked vehicles on one or both sides of the parking space; determine the occupancy status of all identified parking spaces, screen all identified empty parking spaces around the vehicle, and upload all empty parking space information.

[0017] Furthermore, the spatial coordinate system takes the center of the parking space entrance as the origin, the lane direction as the x-axis, and the direction perpendicular to the lane as the y-axis, and calculates the position coordinates of each point on the vehicle body through distance and direction angle information.

[0018] Furthermore, the process of sensing obstacles in the environment is to use the fusion of cameras and ultrasonic radars to sense obstacles, that is, use the camera to identify obstacles, measure the size of obstacles and measure the distance, and use the ultrasonic radar to measure the distance and size of obstacles. The data of the two sensors complement and correct each other.

[0019] Furthermore, after the camera and ultrasonic radar are integrated to perceive dynamic obstacles, they track their movement trajectory and predict their movement trajectory for a period of time in the future based on their type and historical movement trajectory. At the same time, they continuously update their movement trajectory and correct the predicted trajectory in real time.

[0020] Furthermore, the process of obtaining the initial posture of the vehicle and its relative position to the parking space is to use cameras at various locations on the vehicle body and ultrasonic radar to measure the angular distance to the parking space in the established spatial coordinate system to calculate the position and posture of the vehicle body, and obtain the relative position relationship between the vehicle and the parking space and the initial posture parameters of the vehicle.

[0021] Furthermore, the process of planning a safe path is to determine whether the position and size of the static obstacle conflict with the planned trajectory. If a conflict occurs, the parking trajectory strategy is downgraded to re-plan the trajectory and judge again until a conflict-free trajectory or a safe trajectory is planned.

[0022] A second aspect of the present disclosure provides a target parking space selection system based on optimal path planning, comprising:

[0023] The parking space recognition module is configured to collect vehicle parameters and determine the current environment of the vehicle based on the destination selection; after the determination, the parking space is recognized and a spatial coordinate system is established;

[0024] An obstacle prediction module is configured to sense obstacles in the environment and predict the short-term trajectory of dynamic obstacles in real time;

[0025] The first-level target parking space module is configured to obtain the initial posture of the vehicle and the relative position of the vehicle to the parking space, determine the initial posture of the vehicle and the relative position of the vehicle to each identified parking space using the sensing data, determine whether the parking conditions are met, and determine the range of the first-level target parking space;

[0026] Plan the longitudinal speed, acceleration, and jerk curves of the vehicle during parking. Based on the planning results, select the parking trajectory strategy according to the initial posture of the vehicle and the relative position of the vehicle to each identified parking space;

[0027] The secondary target parking space module is configured to consider the influence of obstacles and select parking spaces that can successfully plan a safe path in the primary target parking space range to form the secondary target parking space range;

[0028] The optimal trajectory and parking space selection module is configured to comprehensively select the optimal target parking space and parking trajectory according to the parking trajectory efficiency of each target parking space and the distance between each target parking space and the destination within the secondary target parking space range;

[0029] The vehicle entry module is configured to park the vehicle into the optimal target parking space according to the optimal parking trajectory.

[0030] A third aspect of the present disclosure provides a medium on which a program is stored. When the program is executed by a processor, the steps in the method for selecting a target parking space based on optimal path planning as described in the first aspect of the present disclosure are implemented.

[0031] The fourth aspect of the present disclosure provides a device, including a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in the method for selecting a target parking space based on optimal path planning as described in the first aspect of the present disclosure are implemented.

[0032] The beneficial effects of the above embodiments of the present invention are as follows:

[0033] The present invention further improves the safety, efficiency and predictability of parking by adding the trajectory prediction influence of dynamic obstacles, adjusting the planning of parking trajectory and the selection of optimal trajectory and optimal target parking space. It also adds scenario judgment, comprehensively considers the parking planning of fixed parking spaces and public parking spaces, and is more in line with the driver's parking habits and convenience.

[0034] During the parking process, the present invention first identifies multiple parking spaces around the vehicle body and obtains information about each parking space. Then, a parking trajectory is planned for each available parking space through different parking trajectory strategies, and the safest and most efficient trajectory, i.e., the optimal trajectory, is selected, thereby achieving dual optimal selection of the optimal trajectory and the optimal target parking space.

[0035] The present invention incorporates the influence of the predicted trajectory of dynamic obstacles and the influence of the distance between the parking space and the destination into the selection of the optimal trajectory and the optimal target parking space, thereby achieving more optimized, safe, efficient, and predictable parking trajectory planning and more efficient and convenient target parking space selection.

[0036] The present invention incorporates scene classification, and differentiates and plans public parking space scenes and fixed parking space scenes. When there is a fixed parking space nearby, the vehicle can automatically go to the fixed parking space for parking, which is more in line with the driver's parking habits. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0038] Figure 1 This is a flow chart of a method for selecting a target parking space based on optimal path planning in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0039] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0040] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "include" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or their combinations;

[0041] Embodiment 1:

[0042] The first embodiment of the present disclosure provides a method for selecting a target parking space based on optimal path planning. Figure 1 As shown, the following steps are included:

[0043] S1: Collect vehicle parameters and determine the vehicle's current environment based on the destination selection; after determination, identify the parking space and establish a spatial coordinate system.

[0044] Preferably, the vehicle parameters include vehicle size, driver's fixed parking space position, etc.

[0045] S1.1: Determine the vehicle’s current environment and obtain environmental information such as entrance and exit locations.

[0046] Preferably, the current environment of the vehicle includes a roadside parking environment, a public parking lot, and a parking lot with fixed parking spaces in a residential area, a company, etc.

[0047] S1.1.1: Using the vehicle location information obtained by the vehicle-mounted GPS, first determine the distance to the fixed parking space entered by the owner. When the distance is close, go directly to the fixed parking space for parking. When the distance is far: determine the environment in the map based on the vehicle location information, and further confirm the vehicle's environment based on different environmental characteristics through cameras and ultrasonic radars. When the vehicle is located on the road and there are multiple lane lines in the environment, it is a roadside parking environment; when the vehicle is located in a parking lot and there are no multiple lane lines in the environment, it is a parking lot environment.

[0048] S1.2: When the vehicle is located in a parking lot with fixed parking spaces in a residential area, company, etc., there is no need to select a parking space. The vehicle only needs to be controlled to go to a fixed parking space and plan the optimal parking trajectory for parking.

[0049] S1.3: Use cameras and ultrasonic radars to identify parking spaces within a certain range around the vehicle, determine specific information such as the occupancy status, location, size, etc. of the parking spaces, and establish a spatial coordinate system.

[0050] S1.3.1: Use cameras to identify parking spaces with parking space lines, and use ultrasonic radar to identify parking spaces without parking space lines but with vehicles parked on one or both sides of the parking space. Determine the occupancy status of all identified parking spaces, filter all identified empty parking spaces around the vehicle, and upload all empty parking space information.

[0051] Preferably, parking space information such as parking space position and size is determined by the position of the parking space angle, and the parking space angle position information is acquired by camera and ultrasonic radar perception.

[0052] S1.3.2: The spatial coordinate system is established with the center of the parking space entrance as the origin, the lane direction as the x-axis, and the direction perpendicular to the lane as the y-axis. The position coordinates of each point on the vehicle body can be calculated through information such as distance and direction angle.

[0053] S1.4: If there are no empty parking spaces around the vehicle, the driver will be reminded that there are no available parking spaces around, and the vehicle will be controlled to maintain a low speed in the current lane. If it reaches a turning point, it will automatically turn to the side and continue to look for a parking space.

[0054] S2: Perceive obstacles in the environment and predict short-term trajectories of dynamic obstacles in real time.

[0055] S2.1: Use the camera and ultrasonic radar to sense obstacles within a certain range around the vehicle in real time, and obtain information such as obstacle type, location, size and dynamic obstacle movement trajectory.

[0056] S2.1.1: Use camera and ultrasonic radar to sense obstacles, that is, use camera to identify obstacles, measure obstacle size and distance, and use ultrasonic radar to measure obstacle distance and size. The data of the two sensors complement and correct each other. The obstacles are represented by polygons:

[0057]

[0058] In the formula, Z i represents the i-th obstacle, Z ij represents the jth vertex angle of the i-th polygonal obstacle, i represents the i-th obstacle, j represents the j-th vertex angle of the obstacle, and there are j vertex angles in total. represents the Z of the i-th obstacle i1 , Z i2 The connecting line segments of vertex corners.

[0059] S2.2: Real-time short-term trajectory prediction of dynamic obstacles.

[0060] S2.2.1: After the camera and ultrasonic radar are integrated to perceive dynamic obstacles, they track their movement trajectory and predict their movement trajectory for the next 5S based on their type and historical movement trajectory. At the same time, they continuously update their movement trajectory and correct the predicted trajectory in real time.

[0061] Preferably, the categories of dynamic obstacles in parking lot scenes mainly include: vehicles, pedestrians, children, pets, motorcycles and non-motor vehicles.

[0062] The specific prediction curve expression is:

[0063] y=k1x 5 +k2x 4 +k3x 3 +k4x 2 +k5x

[0064]

[0065] In the formula, y represents the ordinate value of the point on the curve, x represents the abscissa value of the point on the curve, k1, k2, k3, k4, k5 are the influence coefficients of the obstacle prediction trajectory curve, which can be substituted into the historical trajectory points to obtain specific values; v represents the instantaneous speed at time t of the obstacle prediction, t represents the obstacle trajectory prediction duration, and v e It represents the instantaneous velocity of the starting point of the obstacle trajectory prediction, that is, the instantaneous velocity of the last point of the obstacle trajectory, and a represents the historical average acceleration of the obstacle.

[0066] S3: Obtain the initial posture of the vehicle and its relative position to the parking space, and use the perception data to determine the initial posture of the vehicle and the relative position of the vehicle to each identified parking space.

[0067] S3.1: In the established spatial coordinate system, use cameras at various locations on the vehicle body and ultrasonic radar to measure the angular distance to the parking space to calculate the vehicle body position and posture, obtain the relative position relationship between the vehicle and the parking space and the vehicle initial posture parameters. The specific calculation formula is:

[0068]

[0069] Where α is the angle between the vehicle body and the horizontal line, and its positive or negative value indicates that it is above or below the horizontal line, and y A ,y C Respectively represent the ordinate values ​​of points A and C, x A 、x C They represent the horizontal coordinate values ​​of points A and C respectively. Point A is the left front corner of the vehicle, and point C represents the right rear corner of the vehicle.

[0070] S3.2: The formula for calculating the coordinates of the center position of the vehicle's rear axle is:

[0071]

[0072] In the formula, x E ,y E They represent the x and y values ​​of the center of the vehicle's rear axle respectively, d is the length of the vehicle's rear overhang, point C represents the right rear corner of the vehicle, point D is the left rear corner of the vehicle, and θ is the vehicle's attitude angle.

[0073] S4: Determine whether the parking condition is met and determine the range of the first-level target parking space.

[0074] S4.1: Determine the parking conditions based on the size and space constraints of all the identified empty parking spaces in the surrounding area, and use the parking spaces that meet the parking conditions as the first-level target parking space range.

[0075] The specific conditions for parking are:

[0076]

[0077] Among them, condition (1) is to ensure that the parking space size meets the requirements for vehicle parking, condition (2) is to ensure that the lane width meets the requirements for vehicle turning and parking, and condition (3) is to ensure that there are no static obstacles blocking the parking space and the parking space entrance. 障碍物 represents the static obstacle area, Q 车位 represents the interior and exit area of ​​the parking space, a and b represent the length and width of the vehicle and the parking space respectively, S R Indicates the width of the lane next to the parking space, S a is the lane width threshold.

[0078] S5: Planning the longitudinal speed, acceleration, and jerk curve of the vehicle during the parking process, and based on the planning results, selecting a parking trajectory strategy according to the initial posture of the vehicle and the relative position of the vehicle to each identified parking space.

[0079] S5.1: Set the maximum values ​​of speed v, acceleration a and jerk j, and plan their change curves so that the car can run stably according to acceleration-constant speed-deceleration. The speed and acceleration expressions are:

[0080]

[0081]

[0082] Where, t x It represents the time node when the acceleration change causes different speed changes, x∈[1,7], and t represents a certain instant in the parking process of the vehicle.

[0083] S5.2: Calculate parking trajectory strategies in different relative position areas between the vehicle and the parking space based on vehicle basic parameters, parking space type and size, parking space position, kinematic constraints, spatial constraints and other parameters.

[0084] Preferably, the vehicle kinematic constraints, space constraints and size constraints are mainly:

[0085]

[0086] Among them, formula (1) is the vehicle kinematic constraint, formula (2) is the space constraint, and formula (3) is the parking space size constraint. In the formula, δ is the maximum turning angle of the vehicle's front wheels, is the maximum turning speed of the front wheels of the vehicle, S R is the lane width, S a represents the minimum lane width threshold, a and b represent the length and width of the vehicle and parking space respectively.

[0087] S5.3: For different types of parking spaces, select the corresponding optimal parking strategy according to the relative position of the vehicle and the parking space and the initial posture of the vehicle body.

[0088] Preferably, the parking trajectory strategy includes one-time reverse parking, forward adjustment followed by reverse parking, and direct reverse multiple adjustment parking. The priority is one-time reverse parking as the best level, forward adjustment followed by reverse parking as the second best level, and direct reverse adjustment parking as the worst level.

[0089] Under the constraints, the optimal parking strategy is selected according to the relationship between the initial position of the vehicle and the parking space and the initial posture of the vehicle. The main selection conditions are:

[0090]

[0091] In the formula, w x 、w y They are respectively the distances in the x-axis and y-axis directions between the initial position of the vehicle and the center position of the parking space, w xa 、w ya They are respectively the distance thresholds in the x-axis and y-axis directions between the initial position of the vehicle and the center position of the parking space, R min represents the minimum turning radius of the vehicle, θ is the initial posture angle of the vehicle body, S R is the lane width, S a Indicates the second lane width threshold.

[0092] Preferably, when the initial position of the vehicle and the lane width all meet the conditions, that is, when equations (1), (2), and (3) are all satisfied, a one-time reverse parking strategy is preferred; when equation (3) is satisfied and one or none of equations (1) and (2) is satisfied, a forward adjustment followed by reverse parking strategy is preferred; when equation (3) is not satisfied, a direct reverse adjustment parking strategy is selected.

[0093] S5.4: Substitute the obtained vehicle initial position and initial posture parameters to determine the parking strategy, and use the straight line-arc-clotone curve combination method to plan the vehicle parking trajectory for each parking space in the first-level target parking space separately under the vehicle kinematic constraints and space constraints.

[0094] Preferably, the parking trajectory is based on a straight line, a circular arc curve, and a clothoid curve, and a curve is planned for a vehicle with a certain initial posture from an initial position point to a successfully parked parking space. The expressions of the straight line, circular arc curve, and clothoid curve are respectively:

[0095]

[0096]

[0097]

[0098]

[0099]

[0100] Wherein, formula (1) is a straight line expression, formula (2) is a circular curve expression, formula (3) is a clothoid curve expression, formula (4) is an expression for the lateral and longitudinal distance traveled when the car is traveling on a circular curve, and formula (5) is an expression for the lateral and longitudinal distance traveled when the car is traveling on a clothoid curve; x a ,y a ,θ a Indicates the horizontal and vertical coordinates of the center point of the rear axle and the body posture of the vehicle when the vehicle is traveling in a straight line. b ,y b ,θ b Indicates the horizontal and vertical coordinates of the center point of the rear axle and the body posture of the vehicle when the vehicle is traveling on an arc curve. c ,y c ,θ cIt represents the horizontal and vertical coordinates of the center point of the rear axle of the vehicle and the body posture when the vehicle is traveling in a spiral curve; x1, y1, θ1 represent the horizontal and vertical coordinates of the center point of the rear axle of the vehicle and the body posture when the vehicle just starts to travel in a straight line; x2, y2, θ2 represent the horizontal and vertical coordinates of the center point of the rear axle of the vehicle and the body posture when the vehicle just starts to travel in a circular curve; x3, y3, θ3 represent the horizontal and vertical coordinates of the center point of the rear axle of the vehicle and the body posture when the vehicle just starts to travel in a spiral curve; d1 represents the length of the straight-line travel distance of the vehicle; dx2, dy2, dθ2 represent the horizontal and vertical travel distances of the circular curve and the change in the body posture angle of the vehicle; dx3, dy3, dθ3 represent the horizontal and vertical travel distances of the circular curve and the change in the body posture angle of the vehicle; l2, l3, l4 represent the circular curve and spiral curve travel distances of the vehicle and the total length of the spiral curve respectively; R2 and R3 represent the curvature radius of the circular curve of the vehicle and the curvature radius at the end point of the spiral curve respectively.

[0101] S6: Considering the influence of obstacles, select parking spaces that can successfully plan a safe path in the first-level target parking space range to form a second-level target parking space range.

[0102] S6.1: Determine whether the position and size of the static obstacle conflict with the planned trajectory. If a conflict occurs, downgrade the parking trajectory strategy, re-plan the trajectory, and make another determination until a conflict-free trajectory or a safe trajectory is planned.

[0103] S6.2: Consider the impact of dynamic obstacles, mainly considering the short-term predicted trajectory of dynamic obstacles. The conflict risk value is mainly determined by the area size and time overlap between the predicted trajectory of the dynamic obstacle and the planned trajectory of the vehicle and the conflict point within the safe distance around the trajectory.

[0104] S6.3: Consider the impact of dynamic obstacles and add additional adjustment strategies for trajectory planning such as secondary parking and parking waiting. According to the position, size, motion trajectory, and short-term predicted trajectory of dynamic obstacles, the dynamic obstacle conflict risk value is calculated and a safety threshold is set. If the safety threshold is exceeded, the parking trajectory strategy is downgraded and the trajectory is replanned and judged again until a conflict-free trajectory or a safe trajectory is planned.

[0105] S6.3.1: Stop and wait is a dynamic parking strategy adopted when the predicted trajectory conflict risk value of the dynamic obstacle exceeds the safety threshold after the vehicle starts to park. It is a secondary parking trajectory planning strategy, that is, an adjustment strategy that may be adopted based on the dynamic environment based on the original trajectory planning strategy. When the predicted obstacle trajectory does not conflict with the current position of the vehicle and has a certain safety distance, the stop and wait strategy is adopted; when the predicted obstacle trajectory conflicts with the current position of the vehicle or the distance is very close, the secondary planning strategy is adopted.

[0106] S6.3.2: The secondary planning strategy is mainly for vehicle "avoidance", that is, the vehicle also uses the straight line-arc-clotron curve combination method to plan the vehicle to move forward to the roadside and avoid other obstacles. When the vehicle "ends the avoidance", the parking trajectory is replanned based on the current position.

[0107] S7: In the range of the secondary target parking spaces, the optimal target parking space and the parking trajectory are comprehensively selected according to the parking trajectory efficiency of each target parking space and the distance between the target parking space and the destination, and the vehicle is parked in the optimal target parking space according to the optimal parking trajectory.

[0108] Preferably, efficiency mainly considers track length, number of vehicle turns, and vehicle waiting time.

[0109] S7.1: Calculate the distance between each target parking space and the destination: When in a public parking environment: Use the positional relationship between the vehicle and the parking lot pedestrian entrance and exit and the positional relationship between the vehicle and each target parking space to calculate the distance between each target parking space and the parking lot pedestrian entrance and exit; When in a roadside parking environment, use the positional relationship between the vehicle and the destination and the positional relationship between the vehicle and each target parking space to calculate the distance between each target parking space and the destination.

[0110] S7.2: Select the optimal trajectory based on the comprehensive values ​​under different weight coefficients, and then select the optimal target parking space. The selection formula is as follows:

[0111] M=k1(k 11 L+k 12 N+k 13 T)+k2S

[0112] Where M is a quantitative evaluation index, k1 and k2 are weight coefficients of trajectory efficiency and the distance between the parking space and the destination, k1+k2=1, k 11 , k 12 , k 13 is the weight coefficient affecting efficiency, k 11 +k 12 +k 13 =1, L is the total path distance, N is the number of turns, T is the waiting time for the vehicle to park, and S is the distance between the parking space and the destination.

[0113] S7.3: After selecting the optimal target parking space and planning the optimal parking trajectory, the timing of the passengers getting off the vehicle is determined, that is, the corresponding threshold is set according to the size of the vehicle door, etc. After the vehicle is parked in the parking space, if the calculated distance between the left and right sides of the vehicle and the obstacle is less than the threshold, it will cause inconvenience for the passengers to get off the vehicle, and the passengers are advised to get off in advance. The distance between the left and right sides of the vehicle and the obstacle is calculated during the entire parking process. When the distance between the left and right sides of the vehicle is large, a reminder is issued. After the driver agrees, the vehicle stops at the recommended position. When all doors are closed again, the vehicle continues to park.

[0114] Embodiment 2:

[0115] Embodiment 2 of the present disclosure provides a target parking space selection system based on optimal path planning, including:

[0116] The parking space recognition module is configured to collect vehicle parameters and determine the current environment of the vehicle based on the destination selection; after the determination, the parking space is recognized and a spatial coordinate system is established;

[0117] An obstacle prediction module is configured to sense obstacles in the environment and predict the short-term trajectory of dynamic obstacles in real time;

[0118] The first-level target parking space module is configured to obtain the initial posture of the vehicle and the relative position of the vehicle to the parking space, determine the initial posture of the vehicle and the relative position of the vehicle to each identified parking space using the sensing data, determine whether the parking conditions are met, and determine the range of the first-level target parking space;

[0119] Plan the longitudinal speed, acceleration, and jerk curves of the vehicle during parking. Based on the planning results, select the parking trajectory strategy according to the initial posture of the vehicle and the relative position of the vehicle to each identified parking space;

[0120] The secondary target parking space module is configured to consider the influence of obstacles and select parking spaces that can successfully plan a safe path in the primary target parking space range to form the secondary target parking space range;

[0121] The optimal trajectory and parking space selection module is configured to comprehensively select the optimal target parking space and parking trajectory according to the parking trajectory efficiency of each target parking space and the distance between each target parking space and the destination within the secondary target parking space range;

[0122] The vehicle entry module is configured to park the vehicle into the optimal target parking space according to the optimal parking trajectory.

[0123] Embodiment three:

[0124] Embodiment 3 of the present disclosure provides a medium on which a program is stored. When the program is executed by a processor, the steps in the method for selecting a target parking space based on optimal path planning as described in Embodiment 1 of the present disclosure are implemented. The steps are:

[0125] S1: Collect vehicle parameters and determine the current environment of the vehicle based on the destination selection; after determination, identify the parking space and establish a spatial coordinate system;

[0126] S2: Perceive obstacles in the environment and predict short-term trajectories of dynamic obstacles in real time;

[0127] S3: Obtaining the initial posture of the vehicle and its relative position to the parking space, and using the perception data to determine the initial posture of the vehicle and the relative position of the vehicle to each identified parking space;

[0128] S4: Determine whether the parking conditions are met and determine the range of the first-level target parking space;

[0129] S5: planning the longitudinal speed, acceleration, and jerk curve of the vehicle during the parking process, and selecting a parking trajectory strategy based on the planning results according to the initial posture of the vehicle and the relative position of the vehicle to each identified parking space;

[0130] S6: Considering the impact of obstacles, select parking spaces that can successfully plan a safe path in the first-level target parking space range to form a second-level target parking space range;

[0131] S7: In the range of the secondary target parking spaces, the optimal target parking space and the parking trajectory are comprehensively selected according to the parking trajectory efficiency of each target parking space and the distance between the target parking space and the destination, and the vehicle is parked in the optimal target parking space according to the optimal parking trajectory.

[0132] The detailed steps are the same as those of the target parking space selection method based on optimal path planning provided in Example 1, and will not be repeated here.

[0133] Embodiment 4:

[0134] Embodiment 4 of the present disclosure provides a device, including a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in the method for selecting a target parking space based on optimal path planning as described in Embodiment 1 of the present disclosure are implemented. The steps are:

[0135] S1: Collect vehicle parameters and determine the current environment of the vehicle based on the destination selection; after determination, identify the parking space and establish a spatial coordinate system;

[0136] S2: Perceive obstacles in the environment and predict short-term trajectories of dynamic obstacles in real time;

[0137] S3: Obtaining the initial posture of the vehicle and its relative position to the parking space, and using the perception data to determine the initial posture of the vehicle and the relative position of the vehicle to each identified parking space;

[0138] S4: Determine whether the parking conditions are met and determine the range of the first-level target parking space;

[0139] S5: planning the longitudinal speed, acceleration, and jerk curve of the vehicle during the parking process, and selecting a parking trajectory strategy based on the planning results according to the initial posture of the vehicle and the relative position of the vehicle to each identified parking space;

[0140] S6: Considering the impact of obstacles, select parking spaces that can successfully plan a safe path in the first-level target parking space range to form a second-level target parking space range;

[0141] S7: In the range of the secondary target parking spaces, the optimal target parking space and the parking trajectory are comprehensively selected according to the parking trajectory efficiency of each target parking space and the distance between the target parking space and the destination, and the vehicle is parked in the optimal target parking space according to the optimal parking trajectory.

[0142] The detailed steps are the same as those of the target parking space selection method based on optimal path planning provided in Example 1, and will not be repeated here.

[0143] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0144] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for selecting a target parking space based on optimal path planning, characterized in that: The following steps are involved: Collect vehicle parameters and determine the vehicle's current environment based on the destination selection; After the determination, the parking space is identified and a spatial coordinate system is established; Perceive obstacles in the environment and predict short-term trajectories of dynamic obstacles in real time; Using the perception data, determine the initial posture of the vehicle and the relative position of the vehicle to each identified parking space; Determine whether the parking conditions are met and determine the range of the first-level target parking space; Plan the longitudinal speed, acceleration, and jerk curves of the vehicle during parking. Based on the planning results, select the parking trajectory strategy according to the initial posture of the vehicle and the relative position of the vehicle to each identified parking space; Considering the impact of obstacles, select parking spaces that can successfully plan a safe path in the first-level target parking space range to form the second-level target parking space range; In the range of the secondary target parking spaces, the optimal target parking space and parking trajectory are comprehensively selected based on the parking trajectory efficiency of each target parking space and its distance from the destination, and the vehicle is parked in the optimal target parking space according to the optimal parking trajectory; Among them, the process of planning a safe path is to determine whether the position and size of the static obstacle conflicts with the planned trajectory. If a conflict occurs, the parking trajectory strategy is downgraded to re-plan the trajectory and judge again until a conflict-free trajectory or an unsafe trajectory is planned; considering the impact of dynamic obstacles, the short-term predicted trajectory of dynamic obstacles is mainly considered. The conflict risk value is mainly determined by the area size and time overlap between the predicted trajectory of the dynamic obstacle and the vehicle parking planned trajectory and the conflict point within the safe distance around the trajectory. Considering the impact of dynamic obstacles, additional adjustment strategies for secondary parking and parking waiting trajectory planning are added. According to the position, size, motion trajectory and short-term predicted trajectory of the dynamic obstacle, the dynamic obstacle conflict risk value is calculated, and a safety threshold is set. If the safety threshold is exceeded, the parking trajectory strategy is downgraded to re-plan the trajectory and judge again until a conflict-free trajectory or an unsafe trajectory is planned.

2. The method for selecting a target parking space based on optimal path planning according to claim 1, characterized in that: The parking space recognition process is to use cameras to identify parking spaces with parking space lines, and use ultrasonic radars to identify parking spaces without parking space lines but with parked vehicles on one or both sides of the parking space; determine the occupancy status of all identified parking spaces, screen all identified empty parking spaces around the vehicle, and upload all empty parking space information.

3. The method for selecting a target parking space based on optimal path planning according to claim 1, characterized in that: The spatial coordinate system takes the center of the parking space entrance as the origin, the lane direction as the x-axis, and the direction perpendicular to the lane as the y-axis, and calculates the position coordinates of each point on the vehicle body through distance and direction angle information.

4. The method for selecting a target parking space based on optimal path planning according to claim 1, characterized in that: The process of sensing obstacles in the environment is to use the fusion of camera and ultrasonic radar to sense obstacles, that is, use the camera to identify obstacles, measure the size of obstacles and measure the distance, and use ultrasonic radar to measure the distance and size of obstacles. The data of the two sensors complement and correct each other.

5. The method for selecting a target parking space based on optimal path planning according to claim 4, characterized in that: After the camera and ultrasonic radar are integrated to perceive dynamic obstacles, they track their movement trajectory and predict their movement trajectory for a period of time in the future based on their type and historical movement trajectory. At the same time, they continuously update their movement trajectory and correct the predicted trajectory in real time.

6. The method for selecting a target parking space based on optimal path planning according to claim 1, characterized in that: The process of obtaining the initial posture of the vehicle and its relative position to the parking space is to use cameras at various locations on the vehicle body and ultrasonic radar to measure the angular distance to the parking space in the established spatial coordinate system to calculate the vehicle body position and posture, and obtain the relative position relationship between the vehicle and the parking space and the vehicle's initial posture parameters.

7. A target parking space selection system based on optimal path planning, characterized in that: include: The parking space recognition module is configured to collect vehicle parameters and determine the current environment of the vehicle based on the destination selection; After the determination, the parking space is identified and a spatial coordinate system is established; An obstacle prediction module is configured to sense obstacles in the environment and predict the short-term trajectory of dynamic obstacles in real time; The first-level target parking space module is configured to use the sensing data to determine the initial posture of the vehicle and the relative position of the vehicle and each identified parking space; determine whether the parking conditions are met, and determine the range of the first-level target parking space; Plan the longitudinal speed, acceleration, and jerk curves of the vehicle during parking. Based on the planning results, select the parking trajectory strategy according to the initial posture of the vehicle and the relative position of the vehicle to each identified parking space; The secondary target parking space module is configured to consider the influence of obstacles and select parking spaces that can successfully plan a safe path in the primary target parking space range to form the secondary target parking space range; The optimal trajectory and parking space selection module is configured to comprehensively select the optimal target parking space and parking trajectory according to the parking trajectory efficiency of each target parking space and the distance between each target parking space and the destination within the secondary target parking space range; The vehicle entry module is configured to park the vehicle into the optimal target parking space according to the optimal parking trajectory; Among them, the process of planning a safe path is to determine whether the position and size of the static obstacle conflicts with the planned trajectory. If a conflict occurs, the parking trajectory strategy is downgraded to re-plan the trajectory and judge again until a conflict-free trajectory or an unsafe trajectory is planned; considering the impact of dynamic obstacles, the short-term predicted trajectory of dynamic obstacles is mainly considered. The conflict risk value is mainly determined by the area size and time overlap between the predicted trajectory of the dynamic obstacle and the vehicle parking planned trajectory and the conflict point within the safe distance around the trajectory. Considering the impact of dynamic obstacles, additional adjustment strategies for secondary parking and parking waiting trajectory planning are added. According to the position, size, motion trajectory and short-term predicted trajectory of the dynamic obstacle, the dynamic obstacle conflict risk value is calculated, and a safety threshold is set. If the safety threshold is exceeded, the parking trajectory strategy is downgraded to re-plan the trajectory and judge again until a conflict-free trajectory or an unsafe trajectory is planned.

8. A computer-readable storage medium, characterized in that: A plurality of instructions are stored therein, and the instructions are suitable for being loaded by a processor of a terminal device and executing the method for selecting a target parking space based on optimal path planning as described in any one of claims 1-6.

9. A terminal device, characterized in that: It includes a processor and a computer-readable storage medium, the processor is used to implement various instructions; the computer-readable storage medium is used to store multiple instructions, and the instructions are suitable for being loaded by the processor and executing the target parking space selection method based on optimal path planning according to any one of claims 1-6.

Images