A method, apparatus and device for determining a pull-out direction

Abstract

The application provides a parking-out direction determination method, device and equipment. The method comprises the following steps: obtaining candidate parking-out directions corresponding to a target vehicle; if there are at least two candidate parking-out directions, for each candidate parking-out direction, determining a minimum driving area and a maximum driving area of the candidate parking-out direction; if the maximum driving area has an obstacle, moving the edge of the maximum driving area to the direction of the minimum driving area to obtain a target driving area without the obstacle; determining a driving area score of the candidate parking-out direction based on the width and length of the target driving area; selecting a target parking-out direction from all the candidate parking-out directions based on the driving area score of each candidate parking-out direction to control the target vehicle to park out of the parking space from the target parking-out direction. Through the application, the most efficient parking-out direction can be obtained, the safety of the target parking-out direction is ensured, the safety and reliability of vehicle parking are improved, and the efficiency and success rate of vehicle pickup are improved.

Description

Technical Field

[0001] This application relates to the field of intelligent driving, and in particular to a method, apparatus and device for determining parking direction. Background Technology

[0002] With the continuous development of vehicle driver assistance technology and vehicle autonomous driving technology, many vehicles have automatic parking (automatic parking) function. Automatic parking refers to the vehicle automatically parking into a space without the need for manual control by the user. It can help users park automatically, avoid the need for manual parking, and improve the user experience.

[0003] In addition to automatic parking, the vehicle also has an automatic pick-up and exit function. Automatic pick-up and exit means that the vehicle automatically exits the parking space and drives to the designated location according to the pre-planned pick-up route. The user can wait for the vehicle at the designated location without having to go to the parking space to park the vehicle, thus improving the user experience.

[0004] However, there is no effective way to automatically park a vehicle out of a parking space. For example, users need to configure the parking direction, and the vehicle should automatically park out of the space based on this direction. However, this parking direction may be infeasible, meaning that if the vehicle tries to automatically park in this direction, it will collide with obstacles, preventing it from automatically parking and affecting the efficiency of vehicle pick-up and drop-off. Summary of the Invention

[0005] This application provides a method for determining the exit direction, the method comprising:

[0006] Obtain candidate parking directions corresponding to the target vehicle. The detection area corresponding to the candidate parking directions is free of obstacles. The detection area is the minimum space area required for the target vehicle to park out of the parking space.

[0007] If there are at least two candidate parking directions, for each candidate parking direction, determine the minimum driving area and the maximum driving area of ​​that candidate parking direction; wherein, the minimum driving area is determined based on the minimum envelope area of ​​the target vehicle, and the maximum driving area is obtained by expanding the minimum driving area.

[0008] If there are obstacles in the maximum driving area, then move the edge of the maximum driving area in the direction of the minimum driving area to obtain a target driving area without obstacles;

[0009] The driving area score of the candidate parking direction is determined based on the width and length of the target driving area; the target parking direction is selected from all candidate parking directions based on the driving area score of each candidate parking direction, so as to control the target vehicle to park out of the parking space from the target parking direction.

[0010] This application provides a device for determining the exit direction, the device comprising:

[0011] The acquisition module is used to acquire candidate parking directions corresponding to the target vehicle. The detection area corresponding to the candidate parking direction is free of obstacles, and the detection area is the minimum space area required for the target vehicle to exit the parking space. If there are at least two candidate parking directions, for each candidate parking direction, the minimum driving area and the maximum driving area are determined. The minimum driving area is determined based on the minimum envelope area of ​​the target vehicle, and the maximum driving area is obtained by expanding the minimum driving area.

[0012] The processing module is used to move the edge of the maximum driving area in the direction of the minimum driving area if there is an obstacle in the maximum driving area, so as to obtain a target driving area without obstacles.

[0013] The determination module is used to determine the driving area score of the candidate parking direction based on the width and length of the target driving area; and to select a target parking direction from all candidate parking directions based on the driving area score of each candidate parking direction, so as to control the target vehicle to park out of the parking space from the target parking direction.

[0014] This application provides an electronic device, including: a processor and a machine-readable storage medium, the machine-readable storage medium storing machine-executable instructions that can be executed by the processor; the processor is used to execute the machine-executable instructions to implement the docking direction determination method of the above example of this application.

[0015] This application provides a computer program product, which includes a computer program that, when executed by a processor, implements the mooring direction determination method of the above-described example of this application.

[0016] This application provides a machine-readable storage medium storing machine-executable instructions that can be executed by a processor; wherein the processor is used to execute the machine-executable instructions to implement the mooring direction determination method of the above example of this application.

[0017] As can be seen from the above technical solutions, in this embodiment, when there are at least two candidate parking directions for the target vehicle, it does not randomly select a candidate parking direction as the target parking direction, nor does it use a fixed parking direction as the target parking direction. Instead, based on the minimum and maximum driving areas of the candidate parking directions, the edge of the maximum driving area is moved towards the direction of the minimum driving area to obtain a target driving area free of obstacles. The driving area score is determined based on the width and length of the target driving area, and then the target parking direction is selected based on the driving area score of each candidate parking direction. In this way, the target parking direction can be obtained by taking into account both global pick-up efficiency and local parking efficiency, resulting in the parking direction with the highest overall efficiency, ensuring the safety of the target parking direction, and improving the safety and reliability of vehicle parking. For example, when the target vehicle automatically parks out from the target parking direction, it will not collide with obstacles, can make full use of the driving space around the parking space, improve the vehicle's pick-up efficiency, and increase the vehicle's pick-up success rate. It can take into account the distribution of obstacles around the parking space and select the parking direction that allows the driver to safely exit the parking space. It also takes into account the pick-up route and the size of the parking space and selects the parking direction that requires fewer gear shifts and does not take a long detour. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a method for determining the mooring direction in one embodiment of this application.

[0019] Figure 2A This is a schematic diagram of the target vehicle in one embodiment of this application;

[0020] Figure 2B This is a schematic diagram of a public parking space for picking up and dropping off passengers in one embodiment of this application;

[0021] Figure 2C This is a schematic diagram of a private parking space for picking up and dropping off passengers in one embodiment of this application;

[0022] Figure 3 This is a flowchart illustrating a method for determining the mooring direction in one embodiment of this application.

[0023] Figure 4A This is a schematic diagram of an obstacle detection area in one embodiment of this application;

[0024] Figure 4B This is a schematic diagram of the left detection area in one embodiment of this application;

[0025] Figure 4C This is a schematic diagram of the arc angle and the initial tilt angle of the vehicle in one embodiment of this application;

[0026] Figure 4D This is a schematic diagram of the front detection area in one embodiment of this application;

[0027] Figure 5 This is a flowchart illustrating the determination of relative position in one embodiment of this application;

[0028] Figure 6A This is a schematic diagram of a pick-up route starting from the parking space in one embodiment of this application;

[0029] Figure 6B This is a schematic diagram showing the absence of a second type of intersection point in one embodiment of this application;

[0030] Figure 6C This is a schematic diagram illustrating the existence of a second type of intersection point in one embodiment of this application;

[0031] Figure 6D This is a schematic diagram showing two second-type intersection points in one embodiment of this application;

[0032] Figure 7A This is a schematic diagram of a drivable area in one embodiment of this application;

[0033] Figure 7B This is a schematic diagram of the deadzone of a horizontal parking space in one embodiment of this application;

[0034] Figure 7C This is a schematic diagram of a dead zone for a vertical parking space in one embodiment of this application;

[0035] Figure 7D This is a schematic diagram of the deadzone of a tilted parking space in one embodiment of this application;

[0036] Figure 7E This is a schematic diagram of the screening obstacles in one embodiment of this application;

[0037] Figure 7F This is a schematic diagram illustrating the determination of the minimum distance in one embodiment of this application;

[0038] Figure 7G This is a schematic diagram of the moved driving area in one embodiment of this application;

[0039] Figure 7H This is a schematic diagram of obstacle removal in one embodiment of this application;

[0040] Figure 8A This is a schematic diagram illustrating a difficult point in one embodiment of this application;

[0041] Figure 8B This is a schematic diagram of the dividing line of a difficult point in one embodiment of this application;

[0042] Figure 8CThis is a schematic diagram of lateral movement when the reference angle is small in one embodiment of this application;

[0043] Figure 8D This is a schematic diagram showing the movement of the bottom edge when the reference angle is large in one embodiment of this application;

[0044] Figure 9A This is a schematic diagram of the mooring direction determining device in one embodiment of this application;

[0045] Figure 9B This is a hardware structure diagram of an electronic device according to one embodiment of this application. Detailed Implementation

[0046] This application proposes a parking direction determination method, which can be applied to autonomous driving devices or driver assistance devices, wherein the autonomous driving device or driver assistance device is deployed on the target vehicle. See also Figure 1 The diagram shown is a flowchart of the method, which may include:

[0047] Step 101: Obtain the candidate parking direction corresponding to the target vehicle. There are no obstacles in the detection area corresponding to the candidate parking direction. This detection area is the minimum space area required for the target vehicle to park out of the parking space.

[0048] Step 102: If there are at least two candidate parking directions, for each candidate parking direction, determine the minimum driving area and the maximum driving area of ​​that candidate parking direction; wherein, the minimum driving area is determined based on the minimum envelope area of ​​the target vehicle, and the maximum driving area is obtained by expanding the minimum driving area.

[0049] Step 103: If there are obstacles in the maximum driving area, move the edge of the maximum driving area in the direction of the minimum driving area to obtain the target driving area without obstacles.

[0050] Step 104: Determine the driving area score of the candidate parking direction based on the width and length of the target driving area; select the target parking direction from all candidate parking directions based on the driving area score of each candidate parking direction, so as to control the target vehicle to park out of the parking space from the target parking direction.

[0051] For example, moving the edge of the maximum driving area in the direction of the minimum driving area to obtain a target driving area without obstacles may include, but is not limited to: determining the maximum driving area as the current driving area; determining whether there are obstacles in the current driving area; if not, determining the current driving area as the target driving area; if so, determining the edge corresponding to the minimum distance between all obstacles in the current driving area and each edge of the minimum driving area as the reference edge; selecting an adjustment edge corresponding to the reference edge from the four edges of the current driving area, with the direction of the adjustment edge being the same as the direction of the reference edge (e.g., if the reference edge is the bottom edge, the adjustment edge is also the bottom edge). Based on the distance between the obstacle corresponding to the minimum distance and the adjustment edge, moving the adjustment edge of the current driving area in the direction of the reference edge, and determining the moved driving area as the current driving area. Then, returning to execute the operation of determining whether there are obstacles in the current driving area, until there are no obstacles in the current driving area, and determining the current driving area as the target driving area.

[0052] For example, determining the edge corresponding to the minimum distance as a reference edge based on the distances of all obstacles within the current driving area to each edge of the minimum driving area can include: if the distances of the obstacle to the first and second edges of the minimum driving area are the same, and this distance is the minimum distance, then a dividing line is drawn within the overlapping area of ​​the first and second edges based on a configured reference angle; wherein, the vertex of the overlapping area is the intersection of the first and second edges, the distance between the obstacle and the first edge is the distance between the obstacle and that vertex, and the distance between the obstacle and the second edge is the distance between the obstacle and that vertex; the reference angle is the angle between the first edge and the dividing line; if the obstacle is located in the area formed by the dividing line and the first edge, then the first edge is determined as the reference edge; if the obstacle is located in the area formed by the dividing line and the second edge, then the second edge is determined as the reference edge.

[0053] For example, determining the driving area score of the candidate parking direction based on the width and length of the target driving area may include, but is not limited to, using the following formula to determine the driving area score of the candidate parking direction: p e This represents the driving area score for the candidate parking direction, where D represents the width of the target driving area, L represents the length of the target driving area, and d represents the driving area score for the target parking direction. v Indicates the width of the target vehicle, l v w1 represents the width weight, and w2 represents the length weight.

[0054] For example, selecting a target parking direction from all candidate parking directions based on the driving area score of each candidate parking direction includes: determining the candidate parking direction with the largest driving area score as the target parking direction based on the driving area score of each candidate parking direction; or, for each candidate parking direction, determining a target score for that candidate parking direction based on its driving area score and priority score; and determining the candidate parking direction with the largest target score as the target parking direction based on the target score of each candidate parking direction. All candidate parking directions include at least two of the following: front parking direction, left parking direction, and right parking direction; based on the acquired pick-up route, which is the planned driving route of the target vehicle when parking out of the parking space, in the process of determining the priority score of the candidate parking directions, then:

[0055] If the pick-up route and the target vehicle are positioned in front, the priority score for parking from the front is higher than that from the left, and higher than that from the right. If the pick-up route is to the left, the priority score for parking from the left is higher than that from the right; if the pick-up route is to the right, the priority score for parking from the right is higher than that from the left. Alternatively, if the relative position is to the left, the priority score for parking from the left is higher than that from the front, and higher than that from the right. Or, if the relative position is to the right, the priority score for parking from the right is higher than that from the front, and higher than that from the left.

[0056] For example, obtaining candidate parking directions corresponding to the target vehicle may include, but is not limited to: obtaining the initial parking direction corresponding to the target vehicle, wherein there are no obstacles in the detection area corresponding to the initial parking direction; wherein all initial parking directions include at least one of the front parking direction, left parking direction, and right parking direction. If there is one initial parking direction, then that initial parking direction is determined as a candidate parking direction. If there are at least two initial parking directions, when the relative position of the pick-up route and the target vehicle is the front position, all initial parking directions are determined as candidate parking directions; when the relative position is the left position, if the initial parking direction includes the right parking direction, then the right parking direction is eliminated, and the remaining initial parking directions are determined as candidate parking directions; when the relative position is the right position, if the initial parking direction includes the left parking direction, then the left parking direction is eliminated, and the remaining initial parking directions are determined as candidate parking directions. The process of determining the relative position of the pick-up route and the target vehicle may include, but is not limited to:

[0057] The system determines whether the pick-up route and the parking space have a first-type intersection point. The edges to be compared can include the front, left, and right edges. If so, the first-type intersection point is designated as the target point. If not, a perpendicular line segment of a specified length is generated starting from the midpoint of each edge to be compared. The system then determines whether the pick-up route and the perpendicular line segment of each edge to be compared have a second-type intersection point. If no second-type intersection point exists, the midpoint corresponding to the minimum distance between the midpoint of each edge to be compared and the pick-up route is designated as the target point. If one second-type intersection point exists, the midpoint corresponding to that intersection point is designated as the target point. If at least two second-type intersection points exist, the midpoint corresponding to the furthest second-type intersection point along the pick-up route is designated as the target point. Then, the relative position is determined based on the target point. If the target point is located on the front edge, the relative position is the front position; if the target point is located on the left edge, the relative position is the left position; and if the target point is located on the right edge, the relative position is the right position.

[0058] For example, the candidate parking direction can be a front parking direction, a left parking direction, or a right parking direction. The detection area corresponding to the front parking direction can be a rectangular front detection area, the detection area corresponding to the left parking direction can be a trapezoidal left detection area, and the detection area corresponding to the right parking direction can be a trapezoidal right detection area. The process of determining the left or right detection area may include, but is not limited to: determining the inner diameter and outer diameter of the fan ring based on the configured minimum turning radius, the vehicle width, and the vehicle length; wherein the inner diameter of the fan ring is the radius of the inner circle formed by the target vehicle turning based on the minimum turning radius, and the outer diameter of the fan ring is the radius of the outer circle formed by the turn; determining the arc angle corresponding to the vehicle length based on the outer diameter of the fan ring and the vehicle length; determining the fan ring angle based on the arc angle and the configured expected initial tilt angle of the vehicle during parking; determining the positions of the four vertices of the trapezoid based on the inner diameter of the fan ring, the outer diameter of the fan ring, the fan ring angle, the initial corner position of the target vehicle, and the configured potential parking distance; and determining the left or right detection area based on the four vertices.

[0059] For example, the inner diameter of the fan ring can be determined using the following formula: The outer diameter of the fan ring can be determined using the following formula: The following formula can be used to determine this radian angle: The positions (i.e., vertex coordinates) of the four vertices can be determined using the following formula: x hz,1 =R max sinθ+Δd hz , Among them, R min Indicates the inner diameter of the fan ring, R represents the minimum turning radius, and l v d represents the length of the vehicle body.v R indicates the vehicle width. max The outer diameter of the fan ring is represented by α, the radian angle by θ, and the fan ring angle by Δd. hz Indicates the potential berthing distance; where x hz,1 The x-coordinate represents the horizontal coordinate of the first vertex position, and the vertical coordinate of the first vertex position is determined based on the vertical coordinate of the initial corner position of the target vehicle; furthermore, x hz,4 The y-coordinate of the fourth vertex is represented by the horizontal coordinate. hz,4 The vertical coordinate of the fourth vertex position is represented; in addition, the horizontal coordinate of the second vertex position is determined based on the horizontal coordinate of the initial corner position and the potential berthing distance, and the vertical coordinate of the second vertex position is determined based on the vertical coordinate of the initial corner position; in addition, the horizontal coordinate of the third vertex position is determined based on the horizontal coordinate of the first vertex position, and the vertical coordinate of the third vertex position is determined based on the vertical coordinate of the fourth vertex position.

[0060] As can be seen from the above technical solutions, in this embodiment, when there are at least two candidate parking directions for the target vehicle, it does not randomly select a candidate parking direction as the target parking direction, nor does it use a fixed parking direction as the target parking direction. Instead, based on the minimum and maximum driving areas of the candidate parking directions, the edge of the maximum driving area is moved towards the direction of the minimum driving area to obtain a target driving area free of obstacles. The driving area score is determined based on the width and length of the target driving area, and then the target parking direction is selected based on the driving area score of each candidate parking direction. In this way, the target parking direction can be obtained by taking into account both global pick-up efficiency and local parking efficiency, resulting in the parking direction with the highest overall efficiency, ensuring the safety of the target parking direction, and improving the safety and reliability of vehicle parking. For example, when the target vehicle automatically parks out from the target parking direction, it will not collide with obstacles, can make full use of the driving space around the parking space, improve the vehicle's pick-up efficiency, and increase the vehicle's pick-up success rate. It can take into account the distribution of obstacles around the parking space and select the parking direction that allows the driver to safely exit the parking space. It also takes into account the pick-up route and the size of the parking space and selects the parking direction that requires fewer gear shifts and does not take a long detour.

[0061] The following describes the method for determining the berthing direction in this embodiment, taking into account specific application scenarios.

[0062] This application proposes a parking direction determination method, which can be applied to autonomous driving devices or driver assistance devices deployed on a target vehicle. The autonomous driving device or driver assistance device needs to have valet parking functionality, such as AVP (Automated Valet Parking) or PAVP (Public Automated Valet Parking). Through the valet parking function, automatic pick-up and exit from the parking space can be achieved. Automatic pick-up and exit from the parking space means that the vehicle automatically exits the parking space and travels to a designated location according to a pre-planned pick-up route. The user can wait for the vehicle at the designated location without needing to physically go to the parking space to park the vehicle.

[0063] See Figure 2A The diagram illustrates the target vehicle, which is an intelligent vehicle capable of either assisted driving or autonomous driving. For example, if the target vehicle is equipped with autonomous driving equipment, it is an intelligent vehicle capable of autonomous driving, and the parking direction determination method is implemented through the autonomous driving equipment. Alternatively, if the target vehicle is equipped with assisted driving equipment, it is an intelligent vehicle capable of assisted driving, and the parking direction determination method is implemented through the assisted driving equipment.

[0064] The target vehicle can also be equipped with cameras and ultrasonic radar. Figure 2A In this example, we'll use four cameras and twelve ultrasonic radars, where V represents the camera and U represents the ultrasonic radar. Of course, besides cameras and ultrasonic radar, the target vehicle can also be equipped with millimeter-wave radar and / or lidar; there are no restrictions on this. The cameras and ultrasonic radars can detect obstacles around the target vehicle; there are no restrictions on the method of obstacle detection, as long as obstacles can be detected by the cameras and ultrasonic radars.

[0065] Based on the automatic pick-up and exit function, the target vehicle can automatically exit from the parking space and drive to the designated location according to the pick-up route. For example, depending on different application scenarios and mapping methods, the target vehicle can automatically exit from a public parking space based on a high-precision map, or it can automatically exit from a private parking space based on a high-precision map.

[0066] For example, high-precision maps of public parking areas, produced by professional map service providers, are widely used in automated pick-up and drop-off scenarios in public parking lots such as shopping malls, battery swapping stations, and train stations. In these public parking lots, pick-up and drop-off routes are dynamically generated based on the location of waiting vehicles and the pick-up point. See also Figure 2B The diagram shown illustrates the pick-up and drop-off functionality of public parking spaces. Based on this automatic pick-up and drop-off feature, the target vehicle can automatically park from a space in the public parking area and proceed along the designated pick-up route to the designated location (i.e., the pick-up point).

[0067] The high-precision map of the private parking area is created automatically by a private vehicle with mapping capabilities (i.e., the target vehicle), specifically for scenarios involving automated pick-up and drop-off of the target vehicle at designated parking spaces. Within the private parking area, the pick-up and drop-off route remains fixed. See also... Figure 2C The diagram shown illustrates a private parking space pick-up / drop-off system. Based on the automatic pick-up / drop-off function, the target vehicle can automatically park from the private parking space and drive to the pick-up point according to the designated route.

[0068] To maximize the intelligence level, success rate, and efficiency of parking direction determination while ensuring safety, this embodiment proposes a two-stage parking direction determination method. The first stage is the safety assurance stage, which primarily considers the distribution of obstacles around the parking space and filters out parking directions that allow for safe exit. The second stage is the efficiency assurance stage, which primarily considers the pick-up route and the size of the parking space, filtering out parking directions that require fewer gear shifts and avoid longer routes.

[0069] See Figure 3 The diagram shown is a flowchart illustrating a method for determining the berthing direction. This method may include:

[0070] Step 301: Determine the initial berthing direction based on the distribution of obstacles.

[0071] For example, based on high-precision maps, information collected by cameras, information collected by ultrasonic radar, and the size information of the target vehicle, the distribution of obstacles around the target vehicle can be calculated, and then the feasible parking direction in which the target vehicle can safely drive out of the parking space can be selected as the initial parking direction based on the distribution of obstacles.

[0072] Step 302: Determine if the initial berthing direction is unique. If so, that is, there is only one initial berthing direction, then the initial berthing direction can be used as the target berthing direction, and the process of determining the berthing direction ends.

[0073] If not, i.e., there are at least two initial berthing directions, then step 303 can be executed.

[0074] Step 303: Select candidate parking directions from all initial parking directions based on the pick-up route.

[0075] For example, based on the pick-up route on the high-precision map, the parking direction that does not take a detour is selected from all the initial parking directions, and the parking direction that does not take a detour is selected as the candidate parking direction.

[0076] Step 304: Determine if the candidate berthing direction is unique. If so, that is, if there is only one candidate berthing direction, then the candidate berthing direction can be taken as the target berthing direction, and the berthing direction determination process ends.

[0077] If not, i.e., there are at least two candidate berthing directions, then step 305 can be executed.

[0078] Step 305: Select the target parking direction from all candidate parking directions based on the drivable area.

[0079] For example, for each candidate parking direction, the size of the drivable area corresponding to that candidate parking direction can be calculated, and the candidate parking direction with the largest drivable area can be selected as the target parking direction.

[0080] First, regarding step 301, an initial parking direction can be determined based on the obstacle distribution. This initial parking direction may include at least one of a front parking direction, a left parking direction, and a right parking direction. The detection area corresponding to this initial parking direction is free of obstacles, and this detection area can be the minimum space required for the target vehicle to exit the parking space. For example, the detection area corresponding to the front parking direction can be a rectangular front detection area, the detection area corresponding to the left parking direction can be a trapezoidal left detection area, and the detection area corresponding to the right parking direction can be a trapezoidal right detection area.

[0081] For example, obstacle distribution is a factor to consider when choosing a parking direction. To ensure that no collision occurs during parking, obstacle collision detection zones are designed for the front, left, and right parking directions. See [link / reference] Figure 4A The diagram shown is a schematic of the obstacle detection area corresponding to the direction in which the vehicle leaves the parking space.

[0082] For the front parking direction, the front detection area is a rectangular area; for the left parking direction, the left detection area is a trapezoidal area; and for the right parking direction, the right detection area is a trapezoidal area.

[0083] Collision detection technology can detect whether there are obstacles in the front detection area, left detection area, and right detection area, such as obstacles marked on the map, obstacles acquired in real time by ultrasonic radar, or obstacles acquired in real time by camera, thereby determining the feasibility of the corresponding parking direction.

[0084] For example, the detection area (i.e., the obstacle collision detection area) is the minimum space required for a target vehicle to park from that direction. If an obstacle exists within the detection area, the parking direction corresponding to that detection area is not feasible, and it is not used as the initial parking direction. If no obstacle exists within the detection area, the parking direction corresponding to that detection area is used as the initial parking direction, and in subsequent processes, it can be further analyzed whether this initial parking direction is the target parking direction.

[0085] For example, regarding the left-side detection area corresponding to the left-side parking exit direction, see... Figure 4B The diagram shown illustrates how to determine the left detection area. The left detection area can be determined using the following method:

[0086] When the target vehicle is in the parking space, a coordinate system is established with the center position of the target vehicle as the origin. The initial corner positions of the target vehicle (4 initial corner positions) are denoted as ve,[1,2,3,4], where ve,1 represents the upper right corner position, ve,2 represents the upper left corner position, ve,3 represents the lower left corner position, and ve,4 represents the lower right corner position.

[0087] The potential berthing distance can be pre-configured and denoted as Δd. hz Δd hz This represents the potential forward and backward movement space of the target vehicle when it parks from the left, i.e., the movement Δd of the target vehicle. hz Then it exits from the left. For example, Δd hz It can be configured based on experience, such as the potential forward and backward movable distance Δd. hz It can be 0.5m.

[0088] Move the target vehicle forward by Δd hz At that time, the positions of the four corner points of the target vehicle are denoted as v'e, [1,2,3,4]. The target vehicle is then moved backward by Δd. hz At that time, the positions of the four corner points of the target vehicle are denoted as v”e, [1,2,3,4].

[0089] See Figure 4B As shown, the fan ring represents the area swept by the target vehicle when parking with the minimum turning radius R. The inner diameter of the fan ring is Rmin, the outer diameter is Rmax, and the corresponding angle is θ. The minimum turning radius R is an attribute of the target vehicle and can be pre-configured. The inner diameter Rmin is the radius of the inner circle formed by the target vehicle turning with the minimum turning radius R, the outer diameter Rmax is the radius of the outer circle formed by the target vehicle turning with the minimum turning radius R, and the angle θ is the angle of the fan ring formed by the turn.

[0090] from Figure 4B It can be seen that the initial corner point position ve, [1,2,3,4] of the target vehicle overlaps with the corner point of the corresponding fan-shaped area at the front and rear positions in the following region: [v'e,2,v'e,3,v'e,2 ... hz,2 ,v hz,4 Therefore, this overlapping area is the necessary passage for the target vehicle to exit from the left. For ease of calculation, this irregular area is simplified to a corner point of [v]. hz,1 ,v hz,3 ,v hz,4 The trapezoidal region of [v'e,3] is the left detection region corresponding to the left parking direction.

[0091] For example, in [v hz,1 ,v hz,3 ,v hz,4 In the left detection region composed of [v'e,3], v hz,1 v'e,3 represents the position of the first vertex in the left detection region, and v'e,3 represents the position of the second vertex in the left detection region. hz,3 This indicates the position of the third vertex in the left detection region, v hz,4 This indicates the position of the fourth vertex in the detection area on the left.

[0092] Based on the minimum turning radius R and the target vehicle's body width d v and vehicle length l v Determine the inner diameter R of the fan ring. min outer diameter R of the fan ring max For example, the inner diameter of the fan ring can be determined using the following formula: The outer diameter of the fan ring is determined using the following formula:

[0093] Based on the outer diameter R of the fan ring max and vehicle length l v Determine the radian angle α corresponding to the vehicle body length, and then determine the sector angle θ based on the radian angle α and the desired initial vehicle tilt angle β during parking. For example, the radian angle can be determined using the following formula: The angle of the fan ring is determined using the following formula: θ = α + β.

[0094] For example, see Figure 4C The diagram shows the radian angle α and the initial vehicle tilt angle β. α is the radian corresponding to the vehicle body length, and its magnitude is a fixed value determined by the minimum turning radius of the target vehicle and the vehicle size. β is the desired initial vehicle tilt angle when parking on the left. Figure 4C The dashed rectangle represents the initial position the vehicle is expected to reach when it exits from the left side, and β is the tilt angle of the vehicle at the expected initial position.

[0095] For example, based on the inner diameter R of the fan ring min , outer diameter R of the fan ring max The fan-ring angle θ, the initial corner position ve of the target vehicle [1,2,3,4], and the configured potential parking distance Δd hz Determine the positions of the four vertices of the trapezoid [vhz,1,vhz,3,vhz,4,v'e,3]; and form the left detection region by combining these four vertex positions.

[0096] The positions of the four vertices are determined using the following formula: x hz,1 =R max sinθ+Δd hz ,

[0097] x hz,1 Indicates the position v of the first vertex hz,1 The horizontal coordinates, the position of the first vertex v hz,1 The vertical coordinate is determined based on the vertical coordinate of the initial corner point position, such as the vertical coordinate of the first vertex position v. hz,1 The vertical coordinate is equal to the vertical coordinate of the initial corner point position ve,2. hz,4 This represents the lateral coordinate (y) of the fourth vertex position vhz,4. hz,4 This represents the longitudinal coordinate of the fourth vertex position, vhz,4. The lateral coordinate of the second vertex position, v'e,3, is determined based on the lateral coordinate of the initial corner position and the potential berthing distance, such as the initial corner position ve,3 and the potential berthing distance Δd. hz The difference, the vertical coordinate of the second vertex position v'e,3 is determined based on the vertical coordinate of the initial corner point position, such as the vertical coordinate of the second vertex position v'e,3 being equal to the vertical coordinate of the initial corner point position ve,2. The third vertex position v hz,3 The horizontal coordinate is based on the position v of the first vertex. hz,1 The horizontal coordinates are determined, such as the position v of the third vertex. hz,3 The horizontal coordinate can be equal to the position v of the first vertex. hz,1 The horizontal coordinate of the third vertex, v hz,3 The vertical coordinate is based on the position v of the fourth vertex. hz,4 The vertical coordinate is determined, such as the position v of the third vertex. hz,3 The vertical coordinate can be equal to the position v of the fourth vertex. hz,4 The vertical coordinate.

[0098] Assume the target vehicle's body length is l v The length is 4.8m, and the vehicle width is d. v The potential forward and backward movement distance Δd is 1.8m. hz Given a minimum turning radius R of 4.5m and a tilt angle β of 30°, the value of R is calculated. min It is 2.91m, R max The value is 6.72m, α is 45.6°, and θ is 75.6°. The angle is 3.17°. The distance between vhz,4 in the x and y directions can be approximated as: x hz,4 =R min sinθ, y hz,4 =R min -R min cosθ.

[0099] For example, Δdhz and β are both preset, and their values ​​are related to the size of the left-side detection area. The values ​​of these two parameters represent the intensity of the desire for the target vehicle to park from the left. The larger the Δdhz value and the smaller the β value, the smaller the left-side detection area, the lower the probability of an obstacle falling into the left-side detection area, and the higher the urgency to want the target vehicle to park from the left. Conversely, the smaller the Δdhz value and the smaller the β value, the lower the urgency to want the target vehicle to park from the left.

[0100] For example, the method for determining the right detection area corresponding to the right parking direction is similar to that for determining the left detection area, and will not be repeated here.

[0101] For example, for the front detection area corresponding to the front parking direction, see... Figure 4D The diagram shown illustrates the determination of the front detection area. The front detection area can be determined using the following method:

[0102] When the target vehicle is in the parking space, a coordinate system is established with the center of the target vehicle as the origin. The initial corner positions (4 initial corner positions) of the target vehicle are denoted as ve, [1,2,3,4]. The shaded rectangle represents potential obstacles on both sides of the target vehicle, and the lateral distance from the obstacle to the target vehicle is the safe parking distance d. sa Safe distance d sa Based on experience, the lateral distance should be configured to be less than the safe distance d. sa It is impossible to park at this time. The longitudinal distance from the front of the obstacle to the front of the target vehicle is d. ver Longitudinal distance d ver Configuration can be based on experience.

[0103] For the rectangle with corner points v'e, [1,2,3,4], this represents the position where the target vehicle parks with the minimum turning radius R. The fan rings represent the areas swept by the target vehicle when it parks to the left and right with the minimum turning radius R, respectively. The corner points of the overlapping area of ​​the two fan rings are [ve,1,v]. ver,1 For ease of calculation, this irregular region is simplified to have corner points [ve, 1, v]. ver,2 ,v ver,3 The rectangular area defined by [ve,2] represents the front detection area corresponding to the front parking direction. For example, ve,1 represents the lower right corner of the front detection area, ve,2 represents the lower left corner of the front detection area, and v... ver,3 This indicates the position of the upper right corner of the front detection area, v ver,2 This indicates the position of the top-left corner of the front detection area. Alternatively, the right side can be extended outwards by a certain distance to obtain the expanded ve,1 and v. ver,3 Alternatively, the left side can be expanded outward by a certain distance to obtain the expanded ve,2 and v. ver,2The rectangular area formed by the expanded points serves as the front detection area.

[0104] See Figure 4D As shown, from the geometric relationship, v ver,1 The lateral distance in the x-direction can be: R max and R min The method for determining d can be found in the detection area on the left. v Indicates the vehicle width, d sa d represents the lateral safety distance. ver This represents the longitudinal distance. In the above formula, the first two terms are fixed values, so the size of the front detection area is determined by the parameter d. ver The value of parameter d determines the value of the parameter d. ver This indicates the intensity of the desire for the target vehicle to park from the front. (d) ver The higher the value, the smaller the front detection area, the lower the probability of an obstacle falling into the front detection area, and the greater the urgency to want the target vehicle to park from the front. Conversely, the lower the value, the less urgency to want the target vehicle to park from the front.

[0105] from Figure 4D It can be seen that v ver,1 It is v ver,2 With v ver,3 The midpoint of v ver,1 The lateral distance in the x-direction is v. ver,2 The lateral distance in the x-direction is also v ver,3 Lateral distance in the x-direction. Additionally, v ver,2 The longitudinal distance in the y-direction is the same as the longitudinal distance in the y-direction of ve,2, v ver,3 The longitudinal distance in the y-direction is the same as the longitudinal distance in the y-direction of ve,1. In summary, we can obtain [ve,1,v...] ver,2 ,v ver,3 The horizontal and vertical coordinates of the four vertices of [ve,2] are used to obtain the front detection area.

[0106] For example, after obtaining the front detection area, left detection area, and right detection area, if there is no obstacle in the front detection area, the front parking direction is used as the initial parking direction; if there is an obstacle in the front detection area, the front parking direction is not used as the initial parking direction. If there is no obstacle in the left detection area, the left parking direction is used as the initial parking direction; if there is an obstacle in the left detection area, the left parking direction is not used as the initial parking direction. If there is no obstacle in the right detection area, the right parking direction is used as the initial parking direction; if there is an obstacle in the right detection area, the right parking direction is not used as the initial parking direction. Thus, the initial parking direction corresponding to the target vehicle can be obtained. The detection area corresponding to the initial parking direction is free of obstacles, and all initial parking directions can include at least one of the front parking direction, left parking direction, and right parking direction.

[0107] To determine whether an obstacle exists in a certain detection area, collision detection can be performed between the detection area and prior obstacle information in a high-precision map, between the detection area and obstacle information acquired in real time by a camera, or between the detection area and obstacle information acquired in real time by an ultrasonic radar. There are no restrictions on this collision detection, as long as it can determine whether an obstacle exists in the detection area.

[0108] In one possible implementation, after obtaining the initial parking direction, all initial parking directions can be output. Alternatively, after obtaining the initial parking direction, the number of initial parking directions can be determined. If there is one initial parking direction, all initial parking directions can be output. If there are at least two initial parking directions, for each initial parking direction, it is determined whether the lane lines of the adjacent parking spaces of the target vehicle are located within the detection area corresponding to that initial parking direction. If so, that initial parking direction is designated as a parking direction to be eliminated; otherwise, it is designated as a parking direction to be retained. Based on this, if only parking directions to be retained exist, all initial parking directions can be output; if only parking directions to be eliminated exist, all initial parking directions can be output; if both parking directions to be eliminated and parking directions to be retained exist simultaneously, all parking directions to be retained can be output, i.e., no parking directions to be eliminated are output.

[0109] The reason for adopting the above method is that obstacles must be resolutely avoided, while adjacent parking spaces can be used. Therefore, if there is no parking direction to be retained, a parking direction to be eliminated can be output, thereby using adjacent parking spaces. This method helps to ensure that the final target parking direction is both safe and reasonable.

[0110] Second, regarding step 303, candidate parking directions can be selected from all initial parking directions based on the pick-up route. These candidate parking directions can include at least one of the following: a front parking direction, a left parking direction, and a right parking direction. For example, if there is one initial parking direction, that initial parking direction is determined as a candidate parking direction. Alternatively, if there are at least two initial parking directions, when the relative position of the pick-up route and the target vehicle is in a front position, all initial parking directions are determined as candidate parking directions. When the relative position of the pick-up route and the target vehicle is in a left position, if the initial parking directions include a right parking direction, the right parking direction is eliminated, and the remaining initial parking directions are determined as candidate parking directions. When the relative position of the pick-up route and the target vehicle is in a right position, if the initial parking directions include a left parking direction, the left parking direction is eliminated, and the remaining initial parking directions are determined as candidate parking directions.

[0111] For example, a pick-up route is the path a target vehicle must take from its parking space to the pick-up point (i.e., a designated location). Due to differences in mapping methods, pick-up routes can be generated in various ways. For instance, pick-up routes for private parking spaces are generated during mapping and can start from within the parking space. Pick-up routes for public parking spaces can be generated in real-time from the cloud or on the vehicle, or they can be generated during mapping. Public parking space pick-up routes do not necessarily start from within the parking space. In a public battery swapping station scenario, the starting point of the mapping can be set as the pick-up start point, and the battery swapping station as the pick-up end point, facilitating vehicle startup from the parking space and automatic battery swapping. In this scenario, the pick-up route is relatively fixed.

[0112] For example, based on the acquired pick-up route, the relative position of the pick-up route to the target vehicle can be a front position, a left position, or a right position. See also Figure 5 The diagram illustrates the process of determining relative positions. The following steps can be used to determine the relative positions of the pick-up route and the target vehicle:

[0113] Step 501: Determine whether there is a first-type intersection between the pick-up route and the parking space.

[0114] If yes, then proceed to step 502; otherwise, proceed to step 503.

[0115] For example, the parking space where the target vehicle is located can be rectangular (quadrilateral), meaning the parking space has four sides: the front side, the left side, the right side, and the rear side. The side to be compared can include the front side, the left side, and the right side. Based on this, it can be determined whether the pick-up route and the side to be compared have a first-type intersection.

[0116] See Figure 6AThe diagram shows the pick-up route starting from the parking space. The solid line with arrows represents the pick-up route, and the arrows indicate the direction of travel. Clearly, the pick-up route intersects the front side (the side to be compared) at a first-type intersection point. Figure 6A The circle in the diagram can represent the first type of intersection.

[0117] Step 502: Determine the first type of intersection point as the target point, and then proceed to step 508.

[0118] Step 503: Generate a perpendicular line segment of a specified length starting from the midpoint of each edge to be compared.

[0119] For example, starting from the midpoint of the front side of the parking space, generate a perpendicular line segment of a specified length (e.g., 6m); starting from the midpoint of the left side of the parking space, generate a perpendicular line segment of a specified length; starting from the midpoint of the right side of the parking space, generate a perpendicular line segment of a specified length. See also... Figure 6B The diagram illustrates a perpendicular line segment of a specified length. Solid black dots represent the midpoints of the edges to be compared, and dashed lines with solid black dots represent perpendicular line segments of a specified length. Clearly, for each edge to be compared, there is a corresponding perpendicular line segment.

[0120] Step 504: Determine whether there is a second type of intersection between the pick-up route and the perpendicular segment of each side to be compared.

[0121] If there is no second type intersection, then step 505 can be executed; if there is one second type intersection, then step 506 can be executed; if there are at least two second type intersections, then step 507 can be executed.

[0122] Step 505: Based on the distance between the midpoint of each edge to be compared and the pick-up route, determine the midpoint corresponding to the minimum distance (i.e. the midpoint of the edge to be compared with the minimum distance) as the target point, and then execute step 508.

[0123] For example, see Figure 6B The diagram illustrates the absence of any second-type intersection points. The solid lines with arrows represent the pick-up route. Clearly, the perpendicular segments corresponding to the front side, left side, and right side do not have a second-type intersection with the pick-up route. Based on this, the distance d between the midpoint of the front side and the pick-up route can be calculated. f Distance d f This represents the shortest distance between the midpoint of the front side and the pick-up route. The distance d between the midpoint of the left side and the pick-up route can be calculated. l Distance d l This represents the shortest distance between the midpoint of the left side and the pick-up route. The distance d between the midpoint of the right side and the pick-up route can be calculated. r Distance dr This represents the shortest distance from the midpoint of the right side to the pick-up route. Then, we can start from distance d. f Distance d l and distance d r Select the minimum distance from the given information, such as distance d. f and distance d f The corresponding midpoint (i.e., the midpoint of the front side) is determined as the target point.

[0124] Step 506: Determine the midpoint corresponding to the second type of intersection as the target point, and then execute step 508.

[0125] See Figure 6C The diagram illustrates a second type of intersection. The solid line with arrows represents the pick-up route. The perpendicular segment corresponding to the front side does not intersect the pick-up route in the second type. The perpendicular segment corresponding to the left side does not intersect the pick-up route in the second type. The perpendicular segment corresponding to the right side intersects the pick-up route in the second type. The midpoint of this second type of intersection (i.e., the midpoint of the right side) can be determined as the target point.

[0126] Step 507: Along the direction of travel of the pick-up route, determine the relative position of each second type of intersection on the pick-up route, determine the midpoint corresponding to the farthest second type of intersection as the target point, and then execute step 508.

[0127] See Figure 6D The diagram illustrates a route with two intersection points of the second type. The solid line with arrows represents the reception route. The perpendicular segment corresponding to the front side intersects the route at intersection point 1 (second type), the perpendicular segment corresponding to the left side intersects the route at intersection point 2 (second type), and the perpendicular segment corresponding to the right side does not intersect the route at intersection point 2 (second type). For example, if the arrows indicate the direction of travel of the reception route, in this direction, the route first passes intersection point 1 and then intersection point 2. Thus, intersection point 2 is the furthest intersection point of the second type. The midpoint corresponding to intersection point 2 (i.e., the midpoint of the left side) is determined as the target point.

[0128] Step 508: Determine the pick-up route and the relative position of the target vehicle based on the target point.

[0129] For example, if the target point is located on the front side, the relative position is the front position; if the target point is located on the left side, the relative position is the left position; if the target point is located on the right side, the relative position is the right position.

[0130] Third, regarding step 305, a target parking direction can be selected from all candidate parking directions based on the drivable area. This target parking direction can be a front parking direction, a left parking direction, or a right parking direction. For example, if there is one candidate parking direction, that candidate parking direction is determined as the target parking direction. Alternatively, if there are at least two candidate parking directions, one candidate parking direction is selected from all candidate parking directions and used as the target parking direction.

[0131] For example, if there are at least two candidate parking directions, this embodiment designs a screening method based on drivable area. This method calculates the drivable area for each candidate parking direction and selects the target parking direction from all candidate parking directions. See [link to documentation] Figure 7A The diagram shows a schematic of the drivable area. The shaded areas represent obstacles, and the dashed and dotted-line rectangles represent the drivable areas on the front and left sides, respectively. The arrows indicate the pick-up route. Based on the drivable areas on the front and left sides, the difficulty of parking the target vehicle from each candidate parking direction can be estimated. Finally, the optimal parking direction that balances overall pick-up efficiency and local parking efficiency is selected and recorded as the target parking direction.

[0132] For example, given at least two candidate berthing directions, the target berthing direction is obtained using the following steps:

[0133] Step S11: For each candidate parking direction (taking one candidate parking direction as an example below), determine the minimum driving area and the maximum driving area for that candidate parking direction. The minimum driving area is determined based on the minimum envelope area of ​​the target vehicle, and the maximum driving area is obtained by expanding the minimum driving area.

[0134] For example, to ensure that the implementation method is applicable to different types of parking spaces (such as perpendicular, horizontal, and angled parking spaces) and to maximize the use of the actual drivable space around the vehicle (such as when obstacles are distributed along the parking space), this embodiment defines the drivable area of ​​the target vehicle as a rectangular area. To accurately calculate the drivable area, two concepts can be introduced: the minimum drivable area and the maximum drivable area. The minimum drivable area can be denoted as the deadzone, and the maximum drivable area can be denoted as the checkzone.

[0135] The deadzone is defined as the minimum range (lower bound) of the drivable area, which must be free of any obstacles. The checkzone is defined as the maximum range (upper bound) of the drivable area, and its size is determined by the sensor's perception capability. The checkzone is the initial range and needs to be iteratively adjusted subsequently.

[0136] For example, if the target vehicle's parking space is a parallel parking space, then the deadzone can be set as the minimum envelope rectangle with an outer extension distance equal to the safety distance. See [reference needed]. Figure 7B The diagram illustrates the deadzone and checkzone of a level parking space. First, determine the minimum envelope rectangle of the target vehicle. Then, expand each side of the minimum envelope rectangle outwards by a safe distance to obtain the target vehicle's deadzone. After obtaining the deadzone, expand it a certain distance in each of the four directions (i.e., expand it outwards by a certain distance) to obtain the target vehicle's checkzone. The expansion / outward expansion distance is related to the detection capability of sensors such as ultrasonic radar. For example, the expansion distance on the left and right sides can be set to 1m, and the expansion distance in front and behind can be set to 2.5m.

[0137] For example, if the target vehicle's parking space is a perpendicular parking space, the deadzone can be set as a rectangle formed by combining the minimum envelope rectangle and the front detection area, see [reference]. Figure 7C The diagram illustrates the deadzone and checkzone for a perpendicular parking space. First, the minimum envelope rectangle of the target vehicle is determined. Then, the front edge of the minimum envelope rectangle is expanded outwards by a certain distance, aligning with the front edge of the front detection area, thus obtaining the target vehicle's deadzone. After obtaining the deadzone, it can be further expanded outwards by a certain distance in all four directions to obtain the target vehicle's checkzone.

[0138] For example, if the target vehicle's parking space is an angled parking space, the deadzone can be set as a rectangle formed by combining the minimum envelope rectangle and the front detection area, see [reference]. Figure 7D The diagram illustrates the deadzone and checkzone for an inclined parking space. First, the minimum envelope rectangle of the target vehicle is determined. Then, the front edge of the minimum envelope rectangle is expanded outwards by a certain distance, aligning with the front edge of the front detection area, thus obtaining the target vehicle's deadzone. After obtaining the deadzone, it can be further expanded outwards by a certain distance in all four directions to obtain the target vehicle's checkzone.

[0139] Step S12: Determine the maximum driving area as the current driving area, i.e., checkzone is the current driving area. In subsequent processes, the current driving area needs to be iteratively adjusted.

[0140] Step S13: Determine whether there are obstacles in the current driving area.

[0141] If yes, then step S14 can be executed; otherwise, step S17 can be executed.

[0142] Step S14: Based on the distances between all obstacles in the current driving area and each edge of the minimum driving area (deadzone), determine the edge corresponding to the minimum distance as the reference edge; select the adjustment edge corresponding to the reference edge from the four edges of the current driving area, and the direction of the adjustment edge is the same as the direction of the reference edge.

[0143] For example, see Figure 7E The diagram illustrates obstacle filtering. It can filter out all obstacles within the current driving area (checkzone), with circular shaded areas representing obstacles. For each obstacle, the distance between the obstacle and each side of the deadzone can be calculated, such as the distance d between the obstacle and the front side. e (i.e., the shortest distance between the obstacle and the front side), the distance d between the obstacle and the left side. l The distance d between the obstacle and the rear side b The distance d between the obstacle and the right side r Clearly, each obstacle can correspond to four distance values, and Figure 7E The eight obstacles in the game can be matched with 32 distance values.

[0144] Then, we can iterate through all distance values, select the minimum distance from all distance values ​​(e.g., 32 distance values), and determine the edge corresponding to the minimum distance as the reference edge. For example, see... Figure 7F The diagram shown illustrates the determination of the minimum distance. Assume d... min If the minimum distance is the minimum distance among all distance values, then the edge corresponding to the minimum distance is the bottom edge of the deadzone. Therefore, the bottom edge of the deadzone is used as the reference edge.

[0145] Then, you can select the adjustment edge corresponding to the reference edge from the four edges of the current driving area checkzone. That is, when the lower edge of the deadzone is used as the reference edge, the lower edge of the checkzone is used as the adjustment edge.

[0146] Step S15: Based on the distance between the obstacle corresponding to the minimum distance and the adjustment edge, move the adjustment edge of the current driving area in the direction of the reference edge, without moving other edges of the current driving area.

[0147] For example, see Figure 7F As shown, the minimum distance corresponds to the distance d between the obstacle and the adjustment edge. move Move the adjustment edge (i.e., the lower edge) of the current driving area in the direction of the reference edge, and the moving distance of the adjustment edge is d. move In this way, the adjustment edge of the current driving area is shifted inward by a distance d. moveThis yields a new driving zone, which is the iterated checkzone. (See also...) Figure 7G The diagram shows the driving area after the movement. The arrows indicate the direction of movement of the adjusted edge, and the distance the edge moves is d. move .

[0148] By moving the adjustment edge of the current driving area, obstacles not within the checkzone can be removed. See [link / reference]. Figure 7H The diagram shows the removal of obstacles, with the obstacles indicated by gray circles being removed.

[0149] Step S16: Determine the moved driving area as the current driving area, and return to step S13.

[0150] For example, if the number of obstacles in the current driving area is not 0, the above steps are repeated to continue iterative calculation until the number of obstacles in the current driving area is 0, then the calculation is terminated and step S17 is executed.

[0151] Step S17: Determine the current driving area as the target driving area. Obviously, when there are 0 obstacles in the current driving area, the current driving area is the final drivable area. Record the final drivable area as the target driving area, and the target driving area is a drivable area without obstacles.

[0152] For example, the location of the deadzone is predetermined, and the calculation process remains unchanged; the distance from the obstacle to the deadzone only needs to be calculated once before the iteration. Furthermore, the operation of translating the checkzone edge enables batch removal of obstacle points, and the iterative method can quickly obtain the target driving area.

[0153] In one possible implementation, a dead zone is defined as an obstacle that is equidistant from both sides of the dead zone; that is, an obstacle within the overlapping area corresponding to the two sides is considered a dead zone. See also Figure 8A The diagram illustrates a problematic obstacle. The distance from this obstacle to the left side is the same as the distance from the obstacle to the bottom side. The gray rectangular area represents the overlapping area of ​​the two sides, and this obstacle can be considered a problematic obstacle. When a problematic obstacle is determined to be the closest point, which side to shrink will affect the generation of the drivable area. Improper handling of problematic obstacles may cause the drivable area boundary to jump, resulting in frequent vehicle braking.

[0154] In response to the above findings, in this embodiment, regarding step S14, when determining the edge corresponding to the minimum distance as the reference edge, if the distance between the obstacle and the first and second edges of the minimum driving area is the same, and this distance is the minimum distance, meaning the obstacle is a difficult point, a dividing line can be drawn within the overlapping area of ​​the first and second edges based on the configured reference angle. Here, the vertex of the overlapping area is the intersection of the first and second edges, the distance between the obstacle and the first edge is the distance between the obstacle and that vertex, and the distance between the obstacle and the second edge is the distance between the obstacle and that vertex. The reference angle is the angle between the first edge and the dividing line. Based on this, if the obstacle is located in the area formed by the dividing line and the first edge, then the first edge is determined as the reference edge; if the obstacle is located in the area formed by the dividing line and the second edge, then the second edge is determined as the reference edge.

[0155] For example, see Figure 8B The diagram shows a schematic of the dividing line for difficult points. This dividing line divides the overlapping area into two sub-regions. The edge corresponding to the sub-region where the obstacle falls will be shrunk. Figure 8B If an obstacle falls into the sub-region corresponding to the left side, then the left side will shrink.

[0156] For example, by adjusting the angle between the dividing line and the dead zone. The size of the angle can be controlled to adjust the generated drivable area according to different needs. This is a pre-configured reference angle, representing the angle between the first side of the deadzone and the dividing line. The first side of the deadzone can be either of the two sides and can be pre-configured. See also Figure 8B As shown, while keeping the relative positions of the obstacle and the target vehicle unchanged, adjust... The size of the control allows for priority contraction of the sides or bottom edge, thus adapting to the different space requirements of the target vehicle when parking horizontally or perpendicularly. This approach to handling challenging situations ensures more stable boundaries of the drivable area, improving parking safety and reliability.

[0157] In summary, it can be seen that, from the perspective of reference... Different timing will result in different ways of shrinking the checkzone. See also Figure 8C The image shown is a reference angle. A diagram illustrating the movement of the side (i.e., the left side) when the size is smaller. See also... Figure 8D The image shown is a reference angle. A schematic diagram showing the movement of the bottom edge (i.e., the lower side edge) when the size is large.

[0158] Step S18: After obtaining the target driving area for the candidate parking exit direction, determine the driving area score for the candidate parking exit direction based on the width and length of the target driving area. For example, the driving area score for the candidate parking exit direction can be determined based on the width and length of the target driving area, the width of the target vehicle, and the length of the target vehicle. In this way, the driving area score for each candidate parking exit direction can be obtained.

[0159] For example, the driving area score for the candidate parking direction can be determined using the following formula. Of course, the following formula is just an example, and there are no restrictions on the method of determining the driving area score, as long as the driving area score is related to the size of the target driving area and the size of the target vehicle.

[0160]

[0161] In the above formula, p e This represents the driving area score for the candidate parking direction, where D represents the width of the target driving area, L represents the length of the target driving area, and d represents the driving area score for the target parking direction. v Indicates the width of the target vehicle, l v Let w1 represent the width weight and w2 represent the length weight. Both the width and length weights can be configured empirically, as long as w1 + w2 = 1.

[0162] Step S19: For each candidate berthing direction, determine the priority score of that candidate berthing direction.

[0163] For example, if the relative position of the pick-up route and the target vehicle is forward, the priority score for the forward parking direction is higher than that for the left parking direction, and higher than that for the right parking direction. If the direction of the pick-up route (i.e., the direction of travel) is left, the priority score for the left parking direction is higher than that for the right parking direction. If the direction of the pick-up route is right, the priority score for the right parking direction is higher than that for the left parking direction. For instance, if there are three candidate parking directions, the priority score for the high-priority direction is 0.5, the priority score for the medium-priority direction is 0.3, and the priority score for the low-priority direction is 0.2. If there are two candidate parking directions, the priority score for the high-priority direction is 0.5, and the priority score for the low-priority direction is 0.2.

[0164] For example, if the pick-up route is positioned to the left of the target vehicle, the priority score for parking from the left is higher than that of parking from the front, which in turn is higher than that of parking from the right. Considering the left-side position, parking from the right is excluded; therefore, the priority score for parking from the left is higher than that of parking from the front. For instance, the priority score for parking from the left might be 0.5, and the priority score for parking from the front might be 0.2.

[0165] For example, if the pick-up route is positioned to the right relative to the target vehicle, the priority score for parking from the right is higher than that of parking from the front, which in turn is higher than that of parking from the left. Considering the right-side position, parking from the left is excluded; therefore, the priority score for parking from the right is higher than that of parking from the front. For instance, the priority score for parking from the right might be 0.5, and the priority score for parking from the front might be 0.2.

[0166] Step S20: For each candidate berthing direction, determine the target score for that candidate berthing direction based on its driving area score and priority score.

[0167] For example, the target score for the candidate berthing direction can be obtained by weighting the exercise area score and the priority score. Alternatively, the sum of the exercise area score and the priority score can be used as the target score for the candidate berthing direction. For instance, the target score for a candidate berthing direction can be determined using the following formula: P = p g +p e P represents the target score, p g p represents the priority score. e This indicates the exercise of the area score.

[0168] In summary, by combining the pick-up route score and the drivable area score, a target score P can be obtained for each candidate parking direction. The parking direction with the highest target score P is the optimal parking direction that satisfies both safety and overall pick-up efficiency as well as local parking efficiency; this direction is selected as the target parking direction for the target vehicle.

[0169] Step S21: Based on the target score of each candidate parking direction, the candidate parking direction with the largest target score is determined as the target parking direction. Thus, the target parking direction of the target vehicle is obtained, and the target vehicle can be controlled to park out of the parking space from the target parking direction without any restrictions on the parking process.

[0170] In one possible implementation, all data involved in this embodiment is acquired and used only with the knowledge and authorization of the relevant users, that is, the acquisition and use of data are authorized by the users.

[0171] As can be seen from the above technical solutions, the adaptive calculation method for the automatic pick-up and exit direction proposed in this application embodiment can obtain a feasible exit direction in the scenario of automatic valet parking in public parking lots where the parking space attributes are uncertain (or change) or the surrounding environment of the parking space is prone to change. This method makes full use of the drivable space around the parking space and improves the success rate of vehicle pick-up. In the scenario of automatic pick-up and exit in public parking lots where the global path is not fixed, the problem of detours in the exit path will not occur, thus affecting the vehicle pick-up efficiency.

[0172] This system does not rely on prior attributes of parking spaces (such as vertical or horizontal), is compatible with both private and public parking spaces, and achieves safe and efficient parking direction calculation, ensuring both safety and efficiency. Safety is ensured by primarily considering the distribution of obstacles around the parking space to select parking directions that allow safe exit. Efficiency is ensured by primarily considering the pick-up route and the size of the parking space to select parking directions with fewer gear shifts and less detours. Taking into account vehicle kinematic constraints and dynamic adjustment capabilities, the system calculates the minimum drivable areas for vertical (front) and horizontal (left and right) parking, ensuring the safety of candidate parking directions. The system calculates parking directions that balance global pick-up efficiency and local parking efficiency. By calculating the relative position of the parking space and the pick-up route, and by eliminating and scoring candidate feasible directions, it ranks candidate directions based on global pick-up efficiency. By calculating and evaluating the drivable areas of candidate feasible directions, it ranks them based on local parking efficiency. By combining the global pick-up efficiency score and the local parking efficiency score, it finally obtains the parking direction with the highest overall efficiency. This paper proposes an iterative method for rapid calculation of the drivable area of ​​a vehicle. By presetting the upper and lower bounds of the drivable area rectangle, it adapts to different types of parking spaces (such as vertical, horizontal, and inclined). The method achieves rapid convergence of the drivable area through shrinkage and iteration. The method also uses a problem segmentation processing technique to make the boundary of the drivable area more stable, thereby improving the safety and reliability of vehicle parking.

[0173] The adaptive calculation method for parking direction can be applied to intelligent driving functions such as automatic vehicle pick-up and drop-off between private and public parking spaces, and automatic valet charging at public battery swapping stations. When applied to intelligent driving functions, the vehicle does not require the driver to specify the parking direction during automatic pick-up and drop-off, nor does it need to obtain accurate parking space type information in advance, thus improving the usability and convenience of intelligent driving functions. In complex and variable public parking areas, applying this method can improve the safety and success rate of vehicle parking.

[0174] Based on the same concept as the above method, this application proposes a mooring direction determination device, see [link to relevant documentation]. Figure 9A The diagram shown is a structural schematic of the device, which may include:

[0175] The acquisition module 911 is used to acquire candidate parking directions corresponding to the target vehicle. The detection area corresponding to the candidate parking direction is free of obstacles, and the detection area is the minimum space area required for the target vehicle to exit the parking space. If there are at least two candidate parking directions, for each candidate parking direction, the minimum driving area and the maximum driving area are determined. The minimum driving area is determined based on the minimum envelope area of ​​the target vehicle, and the maximum driving area is obtained by expanding the minimum driving area.

[0176] Processing module 912 is used to move the edge of the maximum driving area in the direction of the minimum driving area if there is an obstacle in the maximum driving area, so as to obtain a target driving area without obstacles.

[0177] The determining module 913 is used to determine the driving area score of the candidate parking direction based on the width and length of the target driving area; and to select a target parking direction from all candidate parking directions based on the driving area score of each candidate parking direction, so as to control the target vehicle to park out of the parking space from the target parking direction.

[0178] For example, when the processing module 912 moves the edge of the maximum driving area towards the direction of the minimum driving area to obtain a target driving area without obstacles, it specifically performs the following steps: determining the maximum driving area as the current driving area; determining whether there are obstacles in the current driving area; if not, determining the current driving area as the target driving area; if so, determining the edge corresponding to the minimum distance from all obstacles in the current driving area to each edge of the minimum driving area as a reference edge; selecting an adjustment edge corresponding to the reference edge from the four edges of the current driving area, wherein the direction of the adjustment edge is the same as the direction of the reference edge; moving the adjustment edge of the current driving area towards the direction of the reference edge based on the distance between the obstacle corresponding to the minimum distance and the adjustment edge, determining the moved driving area as the current driving area, returning to perform the operation of determining whether there are obstacles in the current driving area, until there are no obstacles in the current driving area, and determining the current driving area as the target driving area.

[0179] For example, when the processing module 912 determines the edge corresponding to the minimum distance as the reference edge based on the distances between all obstacles in the current driving area and each edge of the minimum driving area, it specifically performs the following: If the distance between the obstacle and the first edge and the second edge of the minimum driving area is the same, and this distance is the minimum distance, then a dividing line is drawn in the overlapping area of ​​the first edge and the second edge based on the configured reference angle; wherein, the vertex of the overlapping area is the intersection of the first edge and the second edge, the distance between the obstacle and the first edge is the distance between the obstacle and the vertex, and the distance between the obstacle and the second edge is the distance between the obstacle and the vertex; the reference angle is the angle between the first edge and the dividing line; if the obstacle is located in the area formed by the dividing line and the first edge, then the first edge is determined as the reference edge; if the obstacle is located in the area formed by the dividing line and the second edge, then the second edge is determined as the reference edge.

[0180] For example, when determining the driving area score of the candidate berthing direction, the determining module 913 is specifically used to determine the driving area score of the candidate berthing direction using the following formula: Where, p e This represents the driving area score for the candidate parking direction, where D represents the width of the target driving area, L represents the length of the target driving area, and d represents the driving area score for the target parking direction. v Indicates the width of the target vehicle, l v w1 represents the width weight, and w2 represents the length weight.

[0181] For example, when the determining module 913 selects the target berthing direction from all candidate berthing directions based on the driving area score of each candidate berthing direction, it is specifically used to: determine the candidate berthing direction with the largest driving area score as the target berthing direction based on the driving area score of each candidate berthing direction; or, for each candidate berthing direction, determine the target score of the candidate berthing direction based on the driving area score and priority score of the candidate berthing direction; and determine the candidate berthing direction with the largest target score as the target berthing direction based on the target score of each candidate berthing direction; wherein, all candidate berthing directions include at least two of the following: front berthing direction, left berthing direction, and right berthing direction;

[0182] Based on the acquired pick-up route, which is the planned driving route of the target vehicle when it exits the parking space, the determining module 913 is further used to determine the priority score of the candidate parking directions; wherein, if the relative position of the pick-up route and the target vehicle is a front position, the priority score of the front parking direction is greater than the priority score of the left parking direction, and the priority score of the front parking direction is greater than the priority score of the right parking direction; if the direction of the pick-up route is to the left, the priority score of the left parking direction is greater than the priority score of the right parking direction; if the direction of the pick-up route is to the right, the priority score of the right parking direction is greater than the priority score of the left parking direction; if the relative position is to the left, the priority score of the left parking direction is greater than the priority score of the front parking direction, and the priority score of the front parking direction is greater than the priority score of the right parking direction; if the relative position is to the right, the priority score of the right parking direction is greater than the priority score of the front parking direction, and the priority score of the front parking direction is greater than the priority score of the left parking direction.

[0183] For example, when the acquisition module 911 acquires the candidate parking direction corresponding to the target vehicle, it is specifically used to: acquire the initial parking direction corresponding to the target vehicle, wherein there are no obstacles in the detection area corresponding to the initial parking direction; all initial parking directions include at least one of the front parking direction, left parking direction, and right parking direction; if there is one initial parking direction, the initial parking direction is determined as a candidate parking direction; if there are at least two initial parking directions, when the relative position of the pick-up route and the target vehicle is the front position, all initial parking directions are determined as candidate parking directions; when the relative position is the left position, if the initial parking direction includes the right parking direction, the right parking direction is eliminated, and the remaining initial parking direction is determined as a candidate parking direction; when the relative position is the right position, if the initial parking direction includes the left parking direction, the left parking direction is eliminated, and the remaining initial parking direction is determined as a candidate parking direction.

[0184] The determining module 913 is further configured to determine the relative position of the pick-up route and the target vehicle; specifically, when determining the relative position of the pick-up route and the target vehicle, the determining module 913 is configured to: determine whether there is a first type of intersection between the pick-up route and the parking space's comparison edge, wherein the comparison edge includes the front side edge, the left side edge, and the right side edge; if yes, then the first type of intersection is determined as the target point; if no, then a perpendicular line segment of a specified length is generated starting from the midpoint of each comparison edge; determine whether there is a second type of intersection between the pick-up route and the perpendicular line segment of each comparison edge; if there is no second type of intersection, then based on the midpoint of each comparison edge... The distance between the point and the pick-up route is used to determine the midpoint corresponding to the minimum distance as the target point; if there is a second type of intersection, the midpoint corresponding to that second type of intersection is determined as the target point; if there are at least two second type of intersections, the midpoint corresponding to the farthest second type of intersection along the pick-up route is determined as the target point; the relative position is determined based on the target point; wherein, if the target point is located on the front side, the relative position is the front position; if the target point is located on the left side, the relative position is the left position; if the target point is located on the right side, the relative position is the right position.

[0185] For example, the candidate parking direction is the front parking direction, the left parking direction, or the right parking direction. The detection area corresponding to the front parking direction is a rectangular front detection area, the detection area corresponding to the left parking direction is a trapezoidal left detection area, and the detection area corresponding to the right parking direction is a trapezoidal right detection area. The determining module 913 is also used to determine the left detection area or the right detection area. When determining the left detection area or the right detection area, the determining module 913 is specifically used to: determine the inner diameter and outer diameter of the fan ring based on the minimum turning radius, the width of the target vehicle, and the length of the vehicle. The inner diameter of the fan ring is the radius of the inner circle formed by the target vehicle turning based on the minimum turning radius, and the outer diameter of the fan ring is the radius of the outer circle formed by the turn. The arc angle corresponding to the vehicle length is determined based on the outer diameter of the fan ring and the vehicle length. The fan ring angle is determined based on the arc angle and the configured initial vehicle tilt angle for parking. The positions of the four vertices of the trapezoid are determined based on the inner diameter of the fan ring, the outer diameter of the fan ring, the fan ring angle, the initial corner position of the target vehicle, and the configured potential parking distance. The left detection area or the right detection area is determined based on the four vertex positions.

[0186] Based on the same application concept as the above method, this application proposes an electronic device (such as an autonomous driving device or an assisted driving device), see [link to relevant documentation]. Figure 9BAs shown, the electronic device may include a processor 921 and a machine-readable storage medium 922, the machine-readable storage medium 922 storing machine-executable instructions that can be executed by the processor 921; the processor 921 is used to execute the machine-executable instructions to implement the mooring direction determination method disclosed in the above example of this application.

[0187] Based on the same concept as the above method, this application also provides a machine-readable storage medium storing a plurality of computer instructions, which, when executed by a processor, can implement the berthing direction determination method disclosed in the above examples of this application.

[0188] The aforementioned machine-readable storage medium can be any electronic, magnetic, optical, or other physical storage device that can contain or store information, such as executable instructions, data, etc. For example, machine-readable storage media can be: RAM (Random Access Memory), volatile memory, non-volatile memory, flash memory, storage drives (such as hard disk drives), solid-state drives, any type of storage disk (such as optical discs, DVDs, etc.), or similar storage media, or combinations thereof.

[0189] Based on the same application concept as the above method, this application embodiment also provides a computer program product, which may include a computer program; wherein, when the computer program is executed by a processor, it implements the mooring direction determination method disclosed in the above examples of this application.

[0190] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, embodiments of this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0191] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for determining the berthing direction, characterized in that, The method includes: Obtain candidate parking directions corresponding to the target vehicle. The detection area corresponding to the candidate parking directions is free of obstacles. The detection area is the minimum space area required for the target vehicle to park out of the parking space. If there are at least two candidate parking directions, for each candidate parking direction, determine the minimum driving area and the maximum driving area of ​​that candidate parking direction; wherein, the minimum driving area is determined based on the minimum envelope area of ​​the target vehicle, and the maximum driving area is obtained by expanding the minimum driving area. If there are obstacles in the maximum driving area, then move the edge of the maximum driving area in the direction of the minimum driving area to obtain a target driving area without obstacles; The driving area score of the candidate parking direction is determined based on the width and length of the target driving area; the target parking direction is selected from all candidate parking directions based on the driving area score of each candidate parking direction, so as to control the target vehicle to park out of the parking space from the target parking direction; The step of moving the edge of the maximum driving area in the direction of the minimum driving area to obtain a target driving area without obstacles includes: determining the maximum driving area as the current driving area; determining whether there are obstacles in the current driving area; if not, determining the current driving area as the target driving area. If so, based on the distances between all obstacles in the current driving area and each edge of the minimum driving area, the edge corresponding to the minimum distance is determined as the reference edge; an adjustment edge corresponding to the reference edge is selected from the four edges of the current driving area, and the direction of the adjustment edge is the same as the direction of the reference edge; based on the distance between the obstacle corresponding to the minimum distance and the adjustment edge, the adjustment edge of the current driving area is moved in the direction of the reference edge, and the driving area after the movement is determined as the current driving area. The process returns to determine whether there are obstacles in the current driving area until there are no obstacles in the current driving area, and the current driving area is determined as the target driving area.

2. The method according to claim 1, characterized in that, The step of determining the edge corresponding to the minimum distance as a reference edge based on the distances between all obstacles in the current driving area and each edge of the minimum driving area specifically includes: If the distance between the obstacle and the first and second sides of the minimum driving area is the same, and this distance is the minimum distance, then a dividing line is drawn within the overlapping area of ​​the first and second sides based on a configured reference angle; wherein, the vertex of the overlapping area is the intersection of the first and second sides, the distance between the obstacle and the first side is the distance between the obstacle and that vertex, and the distance between the obstacle and the second side is the distance between the obstacle and that vertex; the reference angle is the angle between the first side and the dividing line; If the obstacle is located in the area formed by the dividing line and the first edge, then the first edge is determined as the reference edge; If the obstacle is located in the area formed by the dividing line and the second side, then the second side is determined as the reference side.

3. The method according to claim 1 or 2, characterized in that, The step of determining the driving area score for the candidate parking direction based on the width and length of the target driving area includes: The driving area score for the candidate parking direction is determined using the following formula: ； in, This indicates the score of the travel area for the candidate berthing direction. This indicates the width of the target driving area. This indicates the length of the target driving area. This indicates the width of the target vehicle. This indicates the length of the target vehicle. Indicates the weight in the width direction. This indicates the weight in the length direction.

4. The method according to claim 1, characterized in that, The selection of the target exit direction from all candidate exit directions based on the driving area score of each candidate exit direction includes: Based on the driving area score of each candidate berthing direction, the candidate berthing direction with the largest driving area score is determined as the target berthing direction; or, for each candidate berthing direction, the target score of the candidate berthing direction is determined based on the driving area score and priority score of the candidate berthing direction; based on the target score of each candidate berthing direction, the candidate berthing direction with the largest target score is determined as the target berthing direction. Among them, all candidate parking directions include at least two of the following: front parking direction, left parking direction, and right parking direction; based on the acquired pick-up route, which is the planned driving route of the target vehicle when parking out of the parking space, in the process of determining the priority score of the candidate parking directions, then: If the pick-up route is positioned in front of the target vehicle, the priority score for the front parking direction is greater than the priority score for the left parking direction, and the priority score for the front parking direction is greater than the priority score for the right parking direction. If the pick-up route is in the left direction, the priority score for the left parking direction is greater than the priority score for the right parking direction. If the pick-up route is in the right direction, the priority score for the right parking direction is greater than the priority score for the left parking direction. If the relative position is the left side, the priority score for the left parking direction is greater than the priority score for the front parking direction, and the priority score for the front parking direction is greater than the priority score for the right parking direction. If the relative position is the right side, then the priority score for the right side parking direction is greater than the priority score for the front side parking direction, and the priority score for the front side parking direction is greater than the priority score for the left side parking direction.

5. The method according to claim 1 or 4, characterized in that, The step of obtaining the candidate parking direction corresponding to the target vehicle includes: obtaining the initial parking direction corresponding to the target vehicle, wherein there are no obstacles in the detection area corresponding to the initial parking direction; wherein, all initial parking directions include at least one of the front parking direction, the left parking direction, and the right parking direction; If there is an initial berthing direction, then that initial berthing direction is determined as a candidate berthing direction; If there are at least two initial parking directions, when the relative position of the pick-up route and the target vehicle is in the front position, all initial parking directions are determined as candidate parking directions; when the relative position is on the left, if the initial parking directions include the right parking direction, the right parking direction is eliminated, and the remaining initial parking directions are determined as candidate parking directions; when the relative position is on the right, if the initial parking direction includes the left parking direction, the left parking direction is eliminated, and the remaining initial parking directions are determined as candidate parking directions. The process of determining the relative position of the pick-up route and the target vehicle includes: Determine whether there is a first type of intersection between the pick-up route and the parking space, wherein the edges to be compared include the front side, the left side, and the right side; if so, then the first type of intersection is determined as the target point. If not, then generate a perpendicular line segment of a specified length starting from the midpoint of each edge to be compared; Determine whether there is a second type of intersection between the pick-up route and the perpendicular segment of each edge to be compared; If there is no second type of intersection, the midpoint corresponding to the minimum distance between the midpoint of each edge to be compared and the pick-up route is determined as the target point; if there is a second type of intersection, the midpoint corresponding to the second type of intersection is determined as the target point; if there are at least two second type of intersections, the midpoint corresponding to the farthest second type of intersection is determined as the target point along the direction of the pick-up route. The relative position is determined based on the target point; wherein, if the target point is located on the front side, the relative position is the front position; if the target point is located on the left side, the relative position is the left position; if the target point is located on the right side, the relative position is the right position.

6. The method according to any one of claims 1-2 and 4, characterized in that, The candidate berthing direction is the front berthing direction, the left berthing direction, or the right berthing direction. The detection area corresponding to the front berthing direction is a rectangular front detection area, the detection area corresponding to the left berthing direction is a trapezoidal left detection area, and the detection area corresponding to the right berthing direction is a trapezoidal right detection area. The process of determining the left detection region or the right detection region includes: Based on the configured minimum turning radius, the target vehicle's body width, and body length, the inner diameter and outer diameter of the fan ring are determined; wherein, the inner diameter of the fan ring is the radius of the inner circle formed by the target vehicle turning based on the minimum turning radius, and the outer diameter of the fan ring is the radius of the outer circle formed by the turn. The arc angle corresponding to the vehicle length is determined based on the outer diameter of the fan ring and the vehicle length. The fan ring angle is determined based on the arc angle and the initial vehicle tilt angle expected when parking. Based on the inner diameter of the fan ring, the outer diameter of the fan ring, the angle of the fan ring, the initial corner position of the target vehicle, and the configured potential parking distance, the positions of the four vertices of the trapezoid are determined; The left detection region or the right detection region is determined based on the positions of the four vertices.

7. The method according to claim 6, characterized in that, The inner diameter of the fan ring is determined using the following formula: ; The outer diameter of the fan ring is determined using the following formula: ; The radian angle is determined using the following formula: ; The positions of the four vertices are determined using the following formula: ， ， ； in, This indicates the inner diameter of the fan ring. This represents the minimum turning radius. This indicates the length of the vehicle body. This indicates the width of the vehicle body. This indicates the outer diameter of the fan ring. This indicates the radian angle. Indicates the angle of the fan ring. Indicates the potential parking distance; in, The horizontal coordinates of the first vertex position are represented, and the vertical coordinates of the first vertex position are determined based on the vertical coordinates of the initial corner positions of the target vehicle. This represents the horizontal coordinate of the fourth vertex. The vertical coordinates representing the position of the fourth vertex; The lateral coordinates of the second vertex position are determined based on the lateral coordinates of the initial corner position and the potential berthing distance, and the longitudinal coordinates of the second vertex position are determined based on the longitudinal coordinates of the initial corner position. The horizontal coordinate of the third vertex position is determined based on the horizontal coordinate of the first vertex position, and the vertical coordinate of the third vertex position is determined based on the vertical coordinate of the fourth vertex position.

8. A device for determining the direction of mooring, characterized in that, The device includes: The acquisition module is used to acquire candidate parking directions corresponding to the target vehicle. The detection area corresponding to the candidate parking direction is free of obstacles, and the detection area is the minimum space area required for the target vehicle to exit the parking space. If there are at least two candidate parking directions, for each candidate parking direction, the minimum driving area and the maximum driving area are determined. The minimum driving area is determined based on the minimum envelope area of ​​the target vehicle, and the maximum driving area is obtained by expanding the minimum driving area. The processing module is used to move the edge of the maximum driving area in the direction of the minimum driving area if there is an obstacle in the maximum driving area, so as to obtain a target driving area without obstacles. The determination module is used to determine the driving area score of the candidate parking direction based on the width and length of the target driving area; and to select the target parking direction from all candidate parking directions based on the driving area score of each candidate parking direction, so as to control the target vehicle to park out of the parking space from the target parking direction. Specifically, when the processing module moves the edge of the maximum driving area towards the direction of the minimum driving area to obtain a target driving area without obstacles, it performs the following steps: determining the maximum driving area as the current driving area; determining whether there are obstacles in the current driving area; if not, determining the current driving area as the target driving area; if so, determining the edge corresponding to the minimum distance between all obstacles in the current driving area and each edge of the minimum driving area as a reference edge; selecting an adjustment edge corresponding to the reference edge from the four edges of the current driving area, wherein the direction of the adjustment edge is the same as the direction of the reference edge; moving the adjustment edge of the current driving area towards the direction of the reference edge based on the distance between the obstacle corresponding to the minimum distance and the adjustment edge, and determining the moved driving area as the current driving area; returning to perform the operation of determining whether there are obstacles in the current driving area, until there are no obstacles in the current driving area, and determining the current driving area as the target driving area.

9. An electronic device, characterized in that, include: A processor and a machine-readable storage medium, the machine-readable storage medium storing machine-executable instructions that can be executed by the processor; The processor is configured to execute machine-executable instructions to implement the method of any one of claims 1-7.

Images