Parking method and device

Abstract

The embodiment of the present application provides a parking method and device, which relates to the field of data processing technology. The method includes: based on sensor data and a pre-built parking map of a parking space to be parked, generating a first driving path to a gear shift location, and controlling the vehicle to drive based on the first driving path; if the gear shift condition is met, sensing obstacles around the vehicle based on sensor data, and sensing the parking space to be parked; based on the sensed obstacles and the sensed parking space, generating a second driving path, controlling the vehicle to switch gears and park in the parking space to be parked based on the second driving path; during the vehicle driving based on the second driving path, if the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the sensed parking space is greater than a preset posture threshold, updating the untraveled section in the second driving path based on the sensed obstacles and the sensed parking space. The solution provided by the embodiment of the present application can realize automatic parking.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to a parking method and device. Background Art

[0002] As people's living standards improve, cars have become one of the main means of transportation in people's daily lives. When car owners start the automatic parking function, they often need to manually select the parking space they want to park. In view of this situation, it is necessary to provide an automatic parking solution to free the car owner's hands and bring convenience to the car owner's parking. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a parking method and device to control a vehicle to automatically park into a parking space. The specific technical solution is as follows:

[0004] In a first aspect, an embodiment of the present application provides a parking method, the method comprising:

[0005] generating a first driving path to a shifting location based on sensor data collected by sensors carried by the vehicle and a pre-built parking space map of a parking space to be parked, and controlling the vehicle to travel based on the first driving path;

[0006] If the gear shifting condition is met, obstacles around the vehicle are sensed based on the sensor data, and the parking space to be parked is sensed, a second driving path is generated for the vehicle to be parked in the parking space to be parked based on the sensed obstacles and the sensed parking space, and the vehicle is controlled to switch gears and park in the parking space to be parked based on the second driving path;

[0007] During the driving of the vehicle based on the second driving path, if the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the sensed parking space is greater than a preset posture threshold, the untraveled section in the second driving path is updated based on the sensed obstacles and the sensed parking space.

[0008] In a second aspect, an embodiment of the present application provides a parking device, the device comprising:

[0009] a first driving path generating module, configured to generate a first driving path to a gear shifting location based on sensor data collected by sensors carried by the vehicle and a pre-built parking space map of a parking space to be parked, and control the vehicle to travel based on the first driving path;

[0010] a second driving path generating module, configured to sense obstacles around the vehicle and the parking space to be parked based on the sensor data if a shifting condition is met, generate a second driving path for the vehicle to be parked in the parking space to be parked based on the sensed obstacles and the sensed parking space, and control the vehicle to switch gears and park in the parking space to be parked based on the second driving path;

[0011] A path updating module is used to update an untraveled section in the second driving path based on a perceived obstacle and a perceived parking space if a difference between a vehicle terminal posture indicated by the second driving path and a vehicle terminal posture indicated by a perceived parking space is greater than a preset posture threshold value during the driving process of the vehicle based on the second driving path.

[0012] In a third aspect, an embodiment of the present application provides a vehicle-mounted control device, including:

[0013] Memory, used to store computer programs;

[0014] The controller is used to implement the method described in the first aspect when executing the program stored in the memory.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a controller, the method described in the first aspect is implemented.

[0016] In a fifth aspect, an embodiment of the present application provides a vehicle, wherein a computer program is stored in the vehicle, and when the computer program is executed by a controller, the method described in the first aspect is implemented.

[0017] In a sixth aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the method described in the first aspect.

[0018] As can be seen from the above, when the solution provided in the embodiment of the present application is applied for parking, a first driving path to the gear shift location is generated based on sensor data and a pre-built parking space map of the parking space to be parked, and the vehicle is controlled to travel based on the first driving path. When the gear shift conditions are met, a second driving path for parking the vehicle in the parking space can be generated based on the perceived obstacles and the perceived parking space, and the vehicle is controlled to successfully park in the parking space based on the second driving path, thereby realizing automatic parking and bringing convenience to the car owner.

[0019] Of course, implementing any product or method of the present application does not necessarily require achieving all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0021] Figure 1 A schematic diagram of a first parking method provided in an embodiment of the present application;

[0022] Figure 2 A schematic diagram of the first path provided in the embodiment of the present application;

[0023] Figure 3 A schematic diagram of a second path provided in an embodiment of the present application;

[0024] Figure 4 A schematic diagram of a third path provided in an embodiment of the present application;

[0025] Figure 5 A schematic diagram of a boundary line provided in an embodiment of the present application;

[0026] Figure 6 A schematic diagram of a collision detection process provided in an embodiment of the present application;

[0027] Figure 7 A schematic diagram of a second parking method provided in an embodiment of the present application;

[0028] Figure 8 A schematic diagram of a parking process provided in an embodiment of the present application;

[0029] Fig. 9 A schematic diagram of the structure of a parking device provided in an embodiment of the present application;

[0030] Fig.10 A schematic diagram of the structure of a vehicle-mounted control device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field based on the present application belong to the scope of protection of the present application.

[0032] First, the execution subject of the solution provided in the embodiment of the present application is described.

[0033] The executor of the solution provided in the embodiment of the present application is: a vehicle-mounted control device with functions such as data processing and storage.

[0034] The parking solution provided in the embodiment of the present application is described in detail below.

[0035] See also Figure 1 , is a flow chart of a first parking method provided in an embodiment of the present application, the method comprising the following steps S101 to S103.

[0036] Step S101: Based on sensor data collected by sensors mounted on the vehicle and a pre-built parking space map of a parking space to be parked, a first driving path to a shifting location is generated, and the vehicle is controlled to travel based on the first driving path.

[0037] The embodiment of the present application does not limit the type and quantity of the above-mentioned sensors. It is only necessary to ensure that the sensors can collect sensor data that can be used to determine the vehicle's position, posture and other information.

[0038] For example, the above-mentioned sensors may include visual sensors, and may also include posture sensors such as radar sensors, inertial sensors, and positioning sensors. Among them, the above-mentioned visual sensor may include a monocular camera, the above-mentioned inertial sensor may be an IMU (Inertial Measurement Unit), an acceleration sensor, a gyroscope, etc., and the above-mentioned positioning sensor may be a GPS (Global Positioning System) sensor, a Beidou satellite navigation system sensor, etc.

[0039] Depending on the type of sensor, the sensor data collected by the sensor is also different.

[0040] For example, the sensor data collected by the visual sensor can be an image, the sensor data collected by the inertial sensor can be posture data such as speed, acceleration, and angular velocity, and the sensor data collected by the positioning sensor can be positioning data such as longitude and latitude coordinates.

[0041] Based on one or more of the above sensor data, the position of the vehicle can be determined, which is not limited in the embodiments of the present application. For example, based on the image collected by the visual sensor, the position of the vehicle can be determined using the monocular visual positioning principle; based on the positioning information collected by the positioning sensor, the position of the vehicle can be determined, etc.

[0042] When car owners park their cars at places they frequently visit, such as near their homes or companies, they tend to park their cars in fixed parking spaces. Compared with the environment around random parking spaces, the environment around fixed parking spaces is relatively fixed. If the car can identify fixed parking spaces by itself, it will be more convenient for car owners to avoid manually searching for parking spaces and then manually selecting parking spaces.

[0043] The pre-built parking space map is described below.

[0044] The pre-built map may include a complete parking path for the vehicle to enter the parking space, environmental information around the path, and parking space location information. The complete parking path is a path constructed based on data collected by sensors during the process of the vehicle owner driving the vehicle into the parking space, that is, the complete parking path is the path that the vehicle owner controls to enter the parking space.

[0045] In one case, the pre-constructed map may be a parking space map constructed by issuing a map construction prompt based on historical parking information of the parking space and after confirmation by the vehicle owner.

[0046] The pre-built parking space map of the parking space to be parked includes the fixed parking space of the vehicle, and the fixed parking space of the vehicle includes:

[0047] The parking space management system synchronizes the paid parking space to the vehicle; and / or,

[0048] a pre-stored parking space received by an interactive system in an onboard system of the vehicle; and / or,

[0049] Frequently used parking spaces are automatically stored by the vehicle based on historical parking information.

[0050] The above-mentioned historical parking information may include but is not limited to the historical parking times, historical parking duration, parking frequency, etc. of the vehicle in the above-mentioned parking space.

[0051] Therefore, based on the above historical parking information, it can be determined whether the parking space is a fixed parking space.

[0052] For example, if the above-mentioned historical parking times are greater than the preset times, the parking space is determined to be a fixed parking space; if the above-mentioned historical parking duration is greater than the preset duration, the parking space is determined to be a fixed parking space; if the above-mentioned parking frequency is greater than the preset frequency, the parking space is determined to be a fixed parking space; the above-mentioned historical parking times, historical parking duration and parking frequency may also be weightedly calculated according to preset weights, and if the calculation result is greater than a preset threshold, the parking space is determined to be a fixed parking space, etc.

[0053] It can be seen that in this way, whether a parking space is a fixed parking space can be determined more comprehensively and reasonably based on a variety of historical parking information.

[0054] For another example, the interactive system in the vehicle's onboard system may include, but is not limited to, a display screen. After the vehicle is parked, the display screen prompts the owner whether to add the parking space on the left side of the door as a fixed parking space. After the owner agrees, the onboard system constructs a pre-built parking space map of the newly added parking space on the left side of the door. The next time the vehicle enters the gate of the community, it can be prompted whether the vehicle starts the automatic parking function, and based on the pre-built parking space map of the parking space on the left side of the door, a first driving path to the gear shift location is generated, and the vehicle is controlled to drive based on the first driving path; when the gear shift condition is met (the vehicle drives to the door), the obstacles around the vehicle are sensed based on the sensor data, and the parking space on the left side of the door is sensed, and the vehicle is controlled to switch gears and park in the parking space to be parked based on the second driving path.

[0055] The starting time of the buffered sensor data is described below.

[0056] Among them, the sensor data can be cached after the vehicle is detected to have arrived at the specified position. At this time, the starting time of the cached sensor data is the time when the vehicle is detected to have arrived at the specified position. The above-mentioned specified position can be a position representing the area where the vehicle enters the parking space, such as the location of the parking lot gate, etc.; the data can also be cached when a cache start instruction is received from the owner. At this time, the starting time of the cached sensor data is the time when the above-mentioned instruction is received.

[0057] Specifically, based on the cached sensor data, a parking map can be constructed using SLAM algorithms such as Gmapping, Hector, Karto, and Cartographer, but this is not limited in the embodiments of the present application.

[0058] It can be seen that in this way, when the parking space is determined to be a fixed parking space for the vehicle based on the historical parking information of the parking space, the car owner can be actively reminded whether to build a parking space map, and the map can be built after confirmation, without the need for the car owner to find the parking space by himself, thereby improving the user experience; in addition, after confirmation, the map can be built directly based on the cached sensor data. This process is imperceptible to the user and is more convenient for the car owner, thereby improving the practicality of the solution.

[0059] In another case, the pre-constructed map may be a parking space map of the parking space to be parked that is constructed by the vehicle based on perception data after active indication by the user.

[0060] For example, when a vehicle is parked in a private parking space for the first time, the owner can issue a parking map construction instruction through the user interface, so that the on-board control device can construct a parking map of the private parking space based on the cached sensor data to obtain a pre-constructed parking map.

[0061] The above-mentioned shifting locations are described below.

[0062] In the scenario of parking a vehicle in a parking space, the vehicle generally needs to advance to a suitable location, switch to reverse gear at the location, back up, and gradually park in the parking space. The location where the vehicle switches to reverse gear may be referred to as a shift location.

[0063] In this step, the generated first driving path is a path that enables the vehicle to reach the above-mentioned gear shifting location.

[0064] The specific method of generating the first driving route to the gear shifting location is described below.

[0065] In one implementation, a complete driving path for the vehicle to park in the parking space can be generated based on sensor data collected by sensors carried by the vehicle and a pre-built parking space map of the parking space to be parked, and the path before the gear shift location recorded in the complete driving path is intercepted as the first driving path to the gear shift location.

[0066] It can be seen that the complete driving path of the vehicle parking in the parking space will include a location representing the transition from forward to reverse. Therefore, the first location representing the transition from forward to reverse can be determined from the complete driving path, and the determined location can be called a gear shift location.

[0067] In this way, the path before the gear shifting location recorded in the complete driving path can be intercepted and used as the first driving path to the gear shifting location.

[0068] Specifically, the current position of the vehicle and the positions of obstacles around the current position can be determined based on sensor data collected by sensors carried by the vehicle, and then the obtained current position, obstacle position and parking space position recorded in the pre-built parking map are used as inputs to the path planning algorithm to obtain a complete driving path for the vehicle to park in the parking space.

[0069] The above-mentioned path planning algorithm can be an A* (A star) algorithm, a hybrid A* (A star) algorithm, a Dijkstra (Dijkstra) algorithm, a JSP (Jump Point Search) algorithm, a greedy algorithm, or the like.

[0070] In this way, a complete driving path for parking the vehicle in the parking space is first generated, and then the path before the gear shift location recorded in the complete driving path is intercepted, so that the first driving path to the gear shift location can be easily obtained without relying on the prediction of the gear shift location, thereby improving the generation efficiency of the first driving path.

[0071] Combine the following Figure 2 The first driving path is described more intuitively.

[0072] See also Figure 2, which is a schematic diagram of the first path provided in an embodiment of the present application.

[0073] Figure 2 The parking spaces in the map refer to the parking spaces recorded in the pre-built parking space map.

[0074] Depend on Figure 2 It can be seen that the first driving path is a path generated based on the map parking space, which enables the vehicle to reach the shifting location. Therefore, after the vehicle reaches the shifting location based on the first driving path, subsequent path planning can be performed and the vehicle can be parked in the parking space based on the new path obtained by planning.

[0075] In another implementation, a gear shifting location that is convenient for parking in the parking space may be predicted based on a pre-built parking space map of the parking space to be parked, and then a path that allows the vehicle to reach the predicted gear shifting location may be generated as the first driving path.

[0076] Among them, the gear shifting location obtained by the above prediction can be a location determined based on the parking area recorded in the pre-constructed parking map, for example, it can be the side front of the parking area, etc., and the embodiment of the present application is not limited to this.

[0077] The generation method of the above-mentioned first driving path is the same as that of the previous implementation mode, the only difference is that the parking space position recorded in the pre-built map is replaced by the position of the predicted gear shift location and the path interception step is cancelled, which will not be repeated here.

[0078] In another embodiment, the avoidance priority representation value of the perceived obstacle can be obtained, and the perceived obstacle can be classified based on the obtained avoidance priority representation value, and the parking area can be constructed based on the high-priority obstacles obtained by classification, and the boundary line of the parking area can be obtained, thereby generating a first driving path that does not collide with the boundary line and the low-priority obstacles. The specific implementation method is detailed in the subsequent embodiments and is not described in detail here.

[0079] The first driving path is a path generated during the parking process for driving the vehicle to a gear shifting location. The process of generating the first driving path may be referred to as a gear shifting planning stage.

[0080] After the first driving path is generated, the vehicle can be controlled to travel based on the first driving path.

[0081] It should be noted that when controlling the vehicle to travel based on the first driving path, obstacles around the vehicle can be detected based on sensor data acquired in real time, so that the vehicle can be controlled to perform path fine-tuning on the basis of traveling along the first driving path, so that the vehicle does not collide with the detected obstacles.

[0082] Step S102: If the gear shifting condition is met, obstacles around the vehicle and a parking space to be parked are sensed based on sensor data, a second driving path for the vehicle to be parked in the parking space to be parked is generated based on the sensed obstacles and the sensed parking space, and the vehicle is controlled to switch gears and park in the parking space to be parked based on the second driving path.

[0083] The above obstacles can be any obstacles such as pillars, vehicles, roadblocks, pedestrians, cones, etc.

[0084] Specifically, based on the sensor data collected in real time, the position, size and other information of obstacles around the vehicle can be perceived in real time, and based on the sensor data collected in real time, the position, size and other information of the parking space to be parked can be perceived in real time.

[0085] In this way, as the vehicle moves, the collected sensor data is dynamically updated, the perceived obstacles are dynamically updated, and the perceived parking spaces are also dynamically updated.

[0086] For example, when the gear shifting condition is met, it can be considered that the vehicle is suitable for switching to the reverse gear for reversing. Therefore, a second driving path for parking the vehicle in the parking space can be generated, and the vehicle is controlled to switch to the reverse gear and park in the parking space based on the second driving path.

[0087] The above-mentioned shifting conditions are explained below.

[0088] The above-mentioned shifting condition may include at least one of the following conditions:

[0089] 1. The vehicle drives to the cruise destination recorded in the pre-built map.

[0090] The above-mentioned cruising end point refers to the first location in the parking path recorded in the pre-constructed map that represents the vehicle's transition from a forward state to a reverse state, and can also be called a gear shift location in the parking path recorded in the pre-constructed map.

[0091] The parking path recorded in the pre-built map is the path that the owner uses to control the vehicle to park in the parking space. When the vehicle reaches the above-mentioned cruising end point, it also reaches the gear shifting point on the above-mentioned path.

[0092] The gear shifting point of the above-mentioned path is generally the preferred gear shifting point determined by the vehicle owner. Therefore, when the vehicle reaches the above-mentioned cruise end point, it is determined that the gear shifting conditions are met and subsequent path planning is carried out. This can avoid the situation where the vehicle misses the preferred gear shifting point due to continuing to move forward, which is conducive to improving parking efficiency.

[0093] 2. The vehicle travels to a gear shifting location based on the first driving path.

[0094] The above-mentioned shift location is a preferred shift location determined in the process of generating the first driving path. When the vehicle travels to the above-mentioned shift location, it can be determined that the vehicle meets the shift conditions so as to perform subsequent path planning, which is conducive to improving parking efficiency.

[0095] 3. The vehicle stops while traveling along the first driving path.

[0096] For example, a route congestion may occur while the vehicle is traveling along the first travel route.

[0097] As can be seen from the foregoing, the vehicle can be controlled to perform path fine-tuning while traveling along the first driving path, so that the vehicle does not collide with the detected obstacle. On this basis, if the vehicle stops, it means that the vehicle has a serious path congestion while traveling along the first driving path, and cannot continue to travel based on path fine-tuning. Therefore, at this time, determining that the vehicle meets the gear shifting conditions and then performing subsequent path planning can avoid the vehicle from being stagnant for a long time, which is conducive to improving parking efficiency.

[0098] Specifically, the second driving route may be generated in the following manner.

[0099] In one implementation, the current position of the vehicle and the positions of obstacles around the current position can be determined based on sensor data collected by sensors carried by the vehicle, and then the obtained current position, the perceived obstacle position, and the perceived parking space position are used as inputs to a path planning algorithm to obtain a second driving path that allows the vehicle to park in the parking space.

[0100] The above path planning algorithm may be the same as the algorithm introduced in the aforementioned step S101, and will not be described in detail here.

[0101] Combine the following Figure 3 The second driving route is described more intuitively.

[0102] See also Figure 3 , which is a schematic diagram of the second path provided in an embodiment of the present application.

[0103] Figure 3 The black dot indicates the position of the vehicle when the gear shifting condition is met. The map parking space refers to the parking space recorded in the pre-built parking space map. The perceived parking space is the perceived parking space mentioned above.

[0104] Depend on Figure 3 It can be seen that the second driving path is: a path generated based on the sensed parking space, from the position where the vehicle is located when the gear shifting condition is met to the position where the sensed parking space is located.

[0105] In another implementation, the avoidance priority representation value of the perceived obstacle can be obtained, and the perceived obstacle can be classified based on the obtained avoidance priority representation value, and the parking area can be constructed based on the high-priority obstacles obtained by classification, and the boundary line of the parking area can be obtained, thereby generating a first driving path that does not collide with the boundary line and the low-priority obstacles. The specific implementation method is detailed in the subsequent embodiments and is not described in detail here.

[0106] The second driving path is a path generated during the parking process for parking the vehicle into the parking space. The generation process of the second driving path can be called a static planning stage.

[0107] Similar to step S101, in the process of controlling the vehicle to park in the parking space based on the second driving path, obstacles around the vehicle can be detected based on sensor data acquired in real time, so that the vehicle can be controlled to perform path fine-tuning on the basis of driving along the second driving path, so that the vehicle does not collide with the detected obstacles.

[0108] Step S103: When the vehicle is traveling based on the second driving path, if the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the sensed parking space is greater than a preset posture threshold, the untraveled section in the second driving path is updated based on the sensed obstacles and the sensed parking space.

[0109] The above-mentioned terminal position of the vehicle is: the position of the vehicle when the vehicle is parked in the parking space.

[0110] After the second driving path is determined, the posture of the vehicle when it is parked in the parking space based on the second driving path can also be determined, and the posture can be called the vehicle terminal posture indicated by the second driving path.

[0111] For example, the vehicle posture when the parking space is at the end point of the second driving path may be determined as the vehicle end point posture indicated by the second driving path.

[0112] Based on the perceived position of the parking space, the posture of the vehicle when it is parked in the parking space may also be determined, and the posture may be referred to as the vehicle terminal posture indicated by the perceived parking space.

[0113] For example, the vehicle posture when the parking space is at the center point of the perceived area where the parking space is located may be determined as the vehicle terminal posture indicated by the perceived map.

[0114] The second driving path is a path generated based on the parking space sensed last time. Since the position of the vehicle changes continuously while the vehicle is driving based on the second driving path, the parking space sensed based on the sensor data may also change dynamically due to different sensor data collected in real time. That is, the latest sensed parking space may be different from the last sensed parking space, and thus, there may be a difference between the vehicle terminal position indicated by the second driving path and the vehicle terminal position indicated by the parking space sensed in real time.

[0115] Furthermore, since the vehicle is getting closer and closer to the parking space while traveling based on the second driving path, it can be considered that the parking space dynamically sensed based on the real-time collected sensor data is becoming more and more accurate, that is, the dynamically sensed parking space is closer to the real parking space.

[0116] In summary, when the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the perceived parking space is greater than the preset posture threshold, it can be considered that the deviation between the perceived parking space based on which the original second driving path is generated and the dynamically perceived parking space is large. Therefore, in order to ensure accurate parking, the untraveled section in the second driving path can be updated in time.

[0117] Specifically, the current position of the vehicle, the position of the obstacle perceived in real time, and the position of the perceived parking space can be used as inputs of the path planning algorithm to obtain a driving path that allows the vehicle to park from the current position into the parking space, that is, the untraveled section in the second driving path is updated.

[0118] Combine the following Figure 4 A more intuitive description is provided for the updated second driving route.

[0119] See also Figure 4 , which is a schematic diagram of the third path provided in an embodiment of the present application.

[0120] Figure 4 The black dot in the middle indicates the current position of the vehicle. The map parking space refers to the parking space recorded in the pre-built parking space map. The perceived parking space is the perceived parking space mentioned above. The updated perceived parking space indicates the dynamically perceived parking space.

[0121] Depend on Figure 4 It can be seen that the second driving path is: a path generated based on the updated perceived parking space and from the current position of the vehicle to the position of the updated perceived parking space.

[0122] It can be seen that when the vehicle is traveling based on the updated second driving path, the dynamically sensed parking space can continuously detect whether the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the sensed parking space is greater than the preset posture threshold, thereby continuously and dynamically updating the untraveled sub-sections in the second driving path until the vehicle is parked in the parking space. The above process can be called the dynamic planning stage.

[0123] As can be seen from the above, when the solution provided in the embodiment of the present application is used for parking, a first driving path to the gear shift location is generated based on sensor data and a pre-built parking space map of the parking space to be parked, and the vehicle is controlled to travel based on the first driving path. When the gear shift conditions are met, a second driving path for parking the vehicle in the parking space can be generated based on the perceived obstacles and parking space, and the vehicle can be controlled to successfully park in the parking space based on the second driving path, thereby realizing automatic parking and bringing convenience to the car owner.

[0124] Among them, the first driving path is generated based on sensor data and a pre-constructed parking map of the parking space to be parked, that is, the gear shifting location recorded in the first driving path is determined based on the above sensor data and the pre-constructed parking map, so that a more optimal gear shifting location suitable for gear switching can be determined, thereby improving parking efficiency.

[0125] In addition, when the vehicle is traveling based on the second driving path, the untraveled section in the second driving path can also be updated based on the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the perceived parking space. Since the distance of the vehicle to the parking space is getting closer and closer during the driving process based on the second driving path, it can be considered that the parking space dynamically perceived based on the real-time collected sensor data is becoming more and more accurate, that is, it is considered that the dynamically perceived parking space is closer to the real parking space. Therefore, when the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the perceived parking space is greater than the preset posture threshold, it can be considered that the deviation between the perceived parking space based on which the original second driving path is generated and the dynamically perceived parking space is large. Therefore, when the untraveled section in the second driving path is updated at this time, the parking path can be adjusted in real time according to the dynamically perceived parking space, which reduces the number of parking rubbing and improves the accuracy of parking.

[0126] In one embodiment of the present application, before the above step S101, the following steps A and B may also be included.

[0127] Step A: After detecting that the vehicle is in the area described by the pre-built parking map, the automatic parking function is triggered.

[0128] Specifically, the vehicle position can be determined based on the real-time collected sensor data. If the vehicle position is within the area described by the pre-built parking map, the automatic parking function readiness prompt can be triggered.

[0129] The automatic parking function readiness prompt may be in the form of a pop-up window prompt on the user interface, or may be in the form of displaying the automatic parking function readiness prompt information on the user interface.

[0130] Step B: In response to the automatic parking function activation instruction, execute step S101.

[0131] Specifically, if any of the following conditions is detected, it can be considered that the automatic parking function activation instruction from the owner has been received:

[0132] The car owner touches the icon to activate the automatic parking function on the user interface, recognizes positive actions from images collected by the visual sensor installed in the car, and recognizes positive voices from voices picked up by the voice reception module installed in the car.

[0133] The above-mentioned actions representing affirmation may be nodding, a gesture indicating “OK”, etc.; the above-mentioned voice representing affirmation may be “OK”, “OK”, etc.

[0134] In one case, the above conditions can be pre-customized by the vehicle owner.

[0135] It can be seen that after detecting that the vehicle is in the area described by the pre-built parking space map, there is no need to wait until the human eye sees the parking space to be parked or the on-board sensor detects the parking space to be parked. The automatic parking function can be actively triggered to be ready. The parking process is executed after receiving the automatic parking function start indication. In other words, it does not rely on the owner to actively initiate the parking function start indication. Even if the parking space to be parked is blocked and cannot be perceived by the human eye or the on-board sensor, as long as the vehicle detects that the parking requirements are met, the parking function can be pushed to the owner by itself. For example, at the entrance of the parking lot, the owner is prompted to turn on the parking function, making the entire parking process more convenient and user-friendly, and improving the user experience during the parking process.

[0136] Another method for generating the first driving path and the second driving path mentioned above is described below through steps C to F. Since the first driving path and the second driving path are generated in similar ways, the term "driving path" is used below to refer to the first driving path and the second driving path.

[0137] Step C: Based on the sensor data, perceive obstacles around the vehicle and obtain avoidance priority representation values ​​of the perceived obstacles.

[0138] The embodiments of the present application do not limit the method of sensing obstacles around the vehicle based on sensors, which will not be elaborated here.

[0139] The following describes a method for obtaining the avoidance priority representation value of the perceived obstacle.

[0140] In one implementation, the obstacle category of the perceived obstacle, the second distance between each side of the obstacle and the parking space, and the confidence correction coefficient of the second distance corresponding to each side can be obtained; based on the obtained obstacle category, confidence correction coefficient, and second distance, the avoidance priority representation value of the perceived obstacle is obtained.

[0141] According to different paths to be generated, the parking spaces may be parking spaces in a pre-built parking space map or parking spaces perceived. For example, when generating a first driving path, the parking spaces may be parking spaces in a pre-built parking space map, while when generating a second driving path, the parking spaces may be parking spaces perceived.

[0142] The embodiment of the present application does not limit the second distance mentioned above, and is explained below by way of examples.

[0143] In one case, the second distance may be the distance between each side of the three-dimensional contour of the obstacle and the parking space.

[0144] For example, it may be the distance between the center point of each side of the three-dimensional contour and the center point of the parking side of the parking space.

[0145] In another case, the second distance may be the distance between each side of the projection area of ​​the obstacle on the ground and the parking space.

[0146] For example, it may be the distance between the center point of each side of the projection area and the center point of the parking side of the parking space.

[0147] In one case, the avoidance priority representation value of each perceived obstacle can be obtained according to the following expression:

[0148]

[0149] Among them, W i represents the avoidance priority representation value of obstacle i, c i represents the priority coefficient of obstacle category i, V represents the location of the parking space, E i,j represents the position of the jth edge of obstacle i, dis(E i,j ,V) represents the second distance between the jth edge of obstacle i and the parking space, δ i,j represents the confidence correction coefficient of the second distance, α represents the preset scaling factor, and W max Indicates the preset maximum avoidance priority representation value.

[0150] Based on the above expression, it can be seen that when the second distances between each side of the obstacle and the parking space are not 0, it means that there is still a certain distance between the obstacle and the above parking space. At this time, the avoidance priority representation value can be determined based on the priority coefficient of the obstacle category, the second distances between each side of the obstacle and the parking space, etc.; when the second distance between one side of the obstacle and the parking space is 0, it means that one side of the obstacle is already close to the above parking space, so that the risk of collision with the above obstacle when the vehicle is parking in the parking space is high, and therefore the avoidance priority representation value of the above obstacle can be determined as the maximum avoidance priority representation value.

[0151] It can be seen that after obtaining the obstacle category of the perceived obstacle, the second distances between each side of the obstacle and the parking space, and the confidence correction coefficient of the second distance corresponding to each side, the avoidance priority representation value of the obstacle can be quickly and accurately obtained based on the above expression.

[0152] In another embodiment, based on the correspondence between the preset obstacle category, correction coefficient, second distance and the priority score, the priority scores corresponding to the obstacle category, correction coefficient and second distance of the obstacle can be determined, and then the priority scores corresponding to the above obstacle category, correction coefficient and second distance can be weighted calculated according to the corresponding preset weights, and the calculated result can be used as the obstacle avoidance priority representation value.

[0153] It can be seen that the obstacle avoidance priority representation value is determined based on the obstacle category of the perceived obstacle, the second distance between each side of the obstacle and the parking space, and the confidence correction coefficient of the second distance corresponding to each side. In this way, the determined avoidance priority representation value is related to the obstacle category and the second distance between each side of the obstacle and the parking space, thereby obtaining a more comprehensive and accurate avoidance priority representation value.

[0154] Step D: Classify the perceived obstacles to obtain high-priority obstacles whose avoidance priority representation values ​​are greater than a preset representation value threshold and low-priority obstacles whose avoidance priority representation values ​​are not greater than the representation value threshold.

[0155] It can be seen that the avoidance priority characterization value represents the degree of influence of an obstacle on the vehicle parking in a parking space. Therefore, an obstacle whose avoidance priority characterization value is greater than a preset characterization value threshold can be determined as a high-priority obstacle, and an obstacle whose avoidance priority characterization value is less than the preset characterization value threshold can be determined as a low-priority obstacle.

[0156] In one case, the category of the perceived obstacle i can be determined based on the following expression:

[0157]

[0158] Among them, type i When it is 1, it means obstacle i is a high priority obstacle, type i When it is 0, it means that obstacle i is a low priority obstacle, W standard Represents the above characterization value threshold.

[0159] Step E: Based on the high-priority obstacles, a parking area is constructed and a boundary line of the parking area is obtained.

[0160] Specifically, since the high-priority obstacle has been determined at this time, the parking area can be constructed based on the sides of the high-priority obstacle, and the boundary line of the parking area can be obtained.

[0161] For example, based on the edge of each high-priority obstacle that is closest to the parking space, the parking area can be determined, so that the boundary line of the parking area can be obtained.

[0162] The following is based on Figure 5 The above boundary lines are explained more intuitively.

[0163] See also Figure 5 , which is a schematic diagram of a boundary line provided in an embodiment of the present application.

[0164] Figure 5 In the figure, non-motor vehicle 1, non-motor vehicle 2, motor vehicle 1, motor vehicle 2, and cone 2 represent high-priority obstacles; pedestrian 1, motor vehicle 3, non-motor vehicle 3, cone 1, and cone 3 represent low-priority obstacles; and the black line represents the boundary line of the parking area.

[0165] Depend on Figure 5 It can be seen that the parking area can be determined based on the edge of each high-priority obstacle that is closest to the parking space, so that the boundary line of the parking area can be obtained.

[0166] In one embodiment of the present application, when constructing a parking area based on high-priority obstacles, the parking area may be constructed based on the perception results of the high-priority obstacles corresponding to various types of sensors.

[0167] The perception result corresponding to each type of sensor is: the result of perceiving the obstacle using the data collected by the sensor of this type.

[0168] In one embodiment, the sensors may include visual cameras and ultrasonic radars.

[0169] Correspondingly, the perception results of the visual camera on the high-priority obstacles may include the obstacle classification results for the images captured by the visual camera, and the obstacle detection results for the above images; the perception results of the ultrasonic radar on the high-priority obstacles may include the detection results of the radar.

[0170] In this case, the confidence priority of the perception results corresponding to each type of sensor can be comprehensively considered to construct a parking area.

[0171] For example, in order of confidence priority from high to low, the ultrasonic radar's perception results of high-priority obstacles, the obstacle classification results of the visual camera, and the obstacle detection results of the visual camera are integrated to obtain a comprehensive perception result of the high-priority obstacles, and then a parking area is constructed based on the comprehensive perception result of the high-priority obstacles.

[0172] It can be seen that the parking area is constructed based on the perception results of high-priority obstacles corresponding to various types of sensors. Since high-priority obstacles are obstacles that have a greater impact on vehicles parking into parking spaces, comprehensively considering the perception results of high-priority obstacles corresponding to various types of sensors can improve the accuracy of perceiving high-priority obstacles, thereby improving the accuracy of the parking area constructed based on the perception results of high-priority obstacles corresponding to various types of sensors.

[0173] Step F: Generate a driving path that does not collide with boundary lines and low-priority obstacles.

[0174] In one embodiment of the present application, a driving path that does not collide with boundary lines and low-priority obstacles can be generated based on the perception results of high-priority obstacles corresponding to sensors of various types and the perception results of low-priority obstacles corresponding to sensors of preset types.

[0175] In one case, the preset type of sensor may be a visual camera. Thus, the perception result of the low-priority obstacle corresponding to the preset type of sensor may be an obstacle classification result for an image captured by the visual camera, and an obstacle detection result for the image.

[0176] It can be seen that the perception results of low-priority obstacles corresponding to the preset type of sensors are considered when generating the above driving path. Since low-priority obstacles are obstacles that have a smaller impact on the vehicle when parking in the parking space, the perception accuracy requirement for low-priority obstacles is relatively low. Therefore, only obtaining the perception results of low-priority obstacles corresponding to the preset type of sensors can improve the efficiency of perceiving high-priority obstacles, thereby improving the efficiency of generating the above driving path based on the perception results of low-priority obstacles.

[0177] The following describes a method for generating a driving path that does not collide with boundary lines and low-priority obstacles.

[0178] In one implementation, a candidate path can be generated based on boundary lines and low-priority obstacles, and then the boundary lines and low-priority obstacles are projected onto a grid map to obtain a grid map that records obstacle information, thereby determining a collision path in the candidate path based on the collision situation between the candidate path and the obstacle information in the grid map.

[0179] The above candidate path generation method is similar to the path generation method introduced in the aforementioned step S101, the only difference is that the boundary line and low-priority obstacles are used as inputs of the path planning algorithm, which will not be repeated here.

[0180] After the boundary lines and low-priority obstacles are projected onto the grid map, the boundary lines and low-priority obstacles are recorded in the grid map, so that the collision path in the candidate path can be determined based on the collision situation between the candidate path and the obstacle information in the grid map.

[0181] The following is an explanation of the method for determining the above-mentioned collision situation.

[0182] In one case, it can be determined whether the candidate path collides with the grid where the low-priority obstacle is located in the grid map. If so, the candidate path is directly determined to be the collision path. If not, when the candidate path collides with the grid where the boundary line is located in the grid map and the vector line of the boundary line, the candidate path is determined to be the collision path.

[0183] Combine the following Figure 6 , which provides a more intuitive explanation of the above collision detection process.

[0184] See also Figure 6 , is a schematic diagram of a collision detection process provided in an embodiment of the present application, Figure 6 The process shown includes the following steps S601-S608.

[0185] Step S601: Classify the perceived obstacles to obtain high-priority obstacles and low-priority obstacles.

[0186] Step S602: constructing a parking area based on high-priority obstacles and obtaining a boundary line of the parking area.

[0187] Step S603: Projecting the boundary lines and low-priority obstacles onto the grid map.

[0188] Step S604: Determine whether the candidate path collides with the grid where the low-priority obstacle is located. If yes, execute step S608; if no, execute step S605.

[0189] If it is determined that the candidate path collides with the grid where the low priority obstacle is located, step S608 is executed to directly determine that the candidate path is the collision path; otherwise, step S605 is executed to determine whether the candidate path collides with the grid where the boundary line is located.

[0190] Step S605: Determine whether the candidate path collides with the grid where the boundary line is located. If yes, execute step S606; if no, execute step S607.

[0191] If it is determined that the candidate path collides with the grid where the boundary line is located, step S606 can be executed to continue to determine whether the candidate path collides with the boundary line vector line; otherwise, it is determined that the candidate path is not a collision path.

[0192] Step S606: Determine whether the candidate path collides with the boundary vector line, if yes, execute step S608, if no, execute step S607.

[0193] If it is determined whether the candidate path collides with the boundary line vector line, it means that the candidate path collides with both the grid where the boundary line is located and the boundary line vector line. Therefore, step S608 can be executed to determine that the candidate path is a collision path; otherwise, step S607 is executed to determine that the candidate path is not a collision path.

[0194] Step S607: Determine that the candidate path is not a collision path.

[0195] Step S608: Determine that the candidate path is a collision path.

[0196] It can be seen that, on the one hand, since low-priority obstacles have a smaller impact on vehicles parking in parking spaces, they are generally farther away from parking spaces, and the accuracy requirements for the perception results of low-priority obstacles are also lower. Therefore, if the candidate path collides with the grid where the low-priority obstacle is located in the grid map, it can be directly considered that the candidate path collides with the low-priority obstacle, that is, the candidate path is a collision path, which improves the efficiency of determining the candidate path.

[0197] On the other hand, since high-priority obstacles are obstacles that have a greater impact on vehicles parking in parking spaces, they are generally closer to parking spaces, and the accuracy requirements for the perception results of high-priority obstacles are also higher, so the accuracy of the boundary line of the parking area determined based on high-priority obstacles is higher. Based on this, when the candidate path collides with the grid where the boundary line is located in the grid map, an accurate comparison between the candidate path and the vector line of the boundary line can also be performed. If the candidate path collides with the grid where the boundary line is located in the grid map and the vector line of the boundary line, it is determined that the candidate path has collided with the boundary line, that is, the candidate path is determined to be a collision path, which improves the accuracy of determining the candidate path.

[0198] It can be seen that after projecting the boundary lines and low-priority obstacles to the grid map, based on the grids where the boundary lines and low-priority obstacles are located in the grid map, the collision situation between the candidate path and the boundary lines and low-priority obstacles in the grid map can be easily determined, so that the collision path in the candidate path can be quickly determined based on the collision situation.

[0199] In another case, if it is determined that the candidate path collides with any grid of the grid where the low priority obstacle is located or the grid where the high priority obstacle is located in the grid map, the candidate path can be determined to be a collision path.

[0200] In another implementation, the boundary line, the position of the low-priority obstacle, and the candidate path may all be unified into the same two-dimensional coordinate system, and then based on the intersection of the line segments of the candidate path in the two-dimensional coordinate system and the boundary line in the two-dimensional coordinate system, as well as the positional relationship between the line segments corresponding to the candidate path and the points of the low-priority obstacle in the two-dimensional coordinate system, it is determined whether the candidate path is a collision path.

[0201] Specifically, if the line segment corresponding to the candidate path intersects with the line segment corresponding to the boundary line, or the point corresponding to the low-priority obstacle is in the line segment corresponding to the candidate path, it can be determined whether the candidate path is a collision path.

[0202] As can be seen from the above, when the solution provided in the embodiment of the present application is used to generate a driving path, first, the avoidance priority representation value of the perceived obstacle is obtained, and the perceived obstacle is classified based on the avoidance priority representation value to obtain high-priority obstacles and low-priority obstacles. Based on the high-priority obstacles, a parking area is constructed, and the boundary line of the parking area is obtained, so that a driving path that does not collide with the boundary line and the low-priority obstacles can be generated, which can reduce the occurrence of collisions between the vehicle and obstacles during the driving process based on the generated driving path, thereby improving the safety of the parking process.

[0203] In addition, the parking area is constructed based on high-priority obstacles. The boundary line of the parking area is a continuous line segment, while the positions of the high-priority obstacles are discrete points. Therefore, ensuring that the generated path does not collide with the above boundary line can further reduce the probability of the generated path colliding with high-priority obstacles compared to ensuring that the generated path does not collide with each high-priority obstacle, thereby further improving the rationality and safety of the parking process.

[0204] See also Figure 7 , is a flow chart of a second parking method provided in an embodiment of the present application, the method includes the following steps S701-S706.

[0205] Step S701: Control the vehicle to cruise based on sensor data and driving path information recorded in a pre-built map.

[0206] Specifically, the vehicle position can be determined based on sensor data, and then based on the deviation between the vehicle position and the path represented by the driving path information recorded in the pre-built map, the vehicle is controlled to cruise based on the path represented by the driving path information.

[0207] Step S702: During cruising, a first distance between the vehicle and a parking space to be parked is obtained based on sensor data.

[0208] Specifically, the first distance between the vehicle and the parking space may be obtained in the following manner.

[0209] In one case, the distance between the rear wheel axle of the vehicle and the parking boundary line of the parking space can be obtained based on the sensor data as the first distance between the vehicle and the parking space.

[0210] The above-mentioned parking boundary line refers to the boundary line among the four boundary lines of the parking space that the vehicle passes through when entering the parking space.

[0211] In this way, the distance between the rear axle of the vehicle and the parking boundary line of the parking space can be used as the first distance. Since the vehicle generally parks into the parking space by reversing, the rear wheels of the vehicle will first touch the parking boundary line of the parking space. Therefore, the distance between the rear axle of the vehicle and the parking boundary line of the parking space can more accurately represent the distance between the vehicle and the parking space, thereby improving the accuracy of subsequently generating the first driving path based on the first distance.

[0212] Step S703: Determine whether the first distance is less than a preset distance threshold, if yes, execute step S704.

[0213] The above distance threshold can be set by the staff according to the actual scenario or experience.

[0214] Step S704: Based on the sensor data collected by the sensors carried by the vehicle and the pre-built parking space map of the parking space to be parked, a first driving path to the gear shifting location is generated, and the vehicle is controlled to travel based on the first driving path.

[0215] This step is similar to the previous Figure 1 Step S101 in the illustrated embodiment is the same and will not be described again here.

[0216] In one case, if the first driving path generation fails, it can also be determined whether the vehicle has traveled to the cruising destination recorded in the pre-built map. If not, the vehicle is controlled to continue cruising based on the sensor data and the driving path information recorded in the pre-built parking map.

[0217] In the event that the generation of the first driving path fails, if the vehicle has not yet reached the cruise destination recorded in the pre-built map, the vehicle can be controlled to continue cruising. This is conducive to determining a new opportunity to meet the requirement that the first distance is less than the preset distance threshold during the subsequent cruising process, thereby regenerating the first driving path and improving the completeness and rationality of the solution.

[0218] Step S705: If the gear shifting condition is met, obstacles around the vehicle and the parking space to be parked are sensed based on the sensor data, a second driving path for the vehicle to be parked in the parking space to be parked is generated based on the sensed obstacles and the sensed parking space, and the vehicle is controlled to switch gears and park in the parking space to be parked based on the second driving path.

[0219] Step S706: When the vehicle is traveling based on the second driving path, if the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the perceived parking space is greater than a preset posture threshold, the untraveled section in the second driving path is updated based on the perceived obstacles and the perceived parking space.

[0220] The above steps S705-S706 are the same as the above Figure 1 In the illustrated embodiment, steps S102 to S103 are the same and will not be described in detail here.

[0221] As can be seen from the above, when the solution provided by the embodiment of the present application is used for parking, the vehicle can be controlled to cruise based on the sensor data and the driving path information recorded in the pre-built map, and during the cruising process, the first distance between the vehicle and the parking space to be parked is obtained, so that when the first distance is less than the preset distance threshold, the first driving path to the shifting location can be generated, that is, the first driving path can be generated when the distance between the vehicle and the parking space is small. Since the sensor mounted on the vehicle has a more accurate perception result of the information around the parking space when the distance between the vehicle and the parking space is small, generating the first driving path at this time can improve the accuracy of the generated first driving path.

[0222] In addition, this can generate a first driving path to a shift location before the vehicle reaches the cruising end point, which is helpful for determining a shift location earlier than the cruising end point, achieving early reversing, and thus improving parking efficiency.

[0223] In one embodiment of the present application, the above Figure 7 The shifting condition in the illustrated embodiment includes at least one of the following conditions:

[0224] The vehicle drives to a cruising destination recorded in the pre-built map, the vehicle drives to a gear shifting location based on the first driving path, and the vehicle stops while driving based on the first driving path.

[0225] The above-mentioned various shifting conditions have been Figure 1 This is described in step S102 of the illustrated embodiment and will not be described again here.

[0226] It should be noted that the different implementation methods of each step and the implementation methods of adding steps in the automatic parking method embodiment introduced above can be referenced and combined with each other to obtain a new embodiment, which will not be repeated here.

[0227] The following is an overall description of a parking process provided in an embodiment of the present application.

[0228] See also Figure 8 , which is a schematic diagram of a parking process provided in an embodiment of the present application.

[0229] Depend on Figure 8 It can be seen that after certain parking conditions are met, the parking process can be executed. The above parking process can be divided into three stages: gear shift planning stage, static planning stage and dynamic planning stage, which are explained below respectively.

[0230] Gear shift planning phase:

[0231] First, it can be determined whether the distance between the vehicle and the parking space is less than the distance threshold. If not, the vehicle can be controlled to continue cruising and determine whether the vehicle has reached the cruising end. If so, it enters the static planning stage. Otherwise, it returns to determine whether the distance between the vehicle and the parking space is less than the distance threshold. If yes, it can enter the gear shift planning stage, generate the first planned path, and determine whether the generation is successful. If not, it returns to determine whether the vehicle has reached the cruising end. If so, it determines whether the gear shift conditions are met. If yes, it enters the static planning stage.

[0232] Static planning stage:

[0233] Generate a second planned path and determine whether it is generated successfully. If so, control the vehicle to travel based on the second planned path, and during driving, determine whether the posture difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the perceived parking space is greater than the posture threshold. If so, enter the dynamic planning stage. If not, determine whether to park in the parking space; if not, that is, the path generation fails, then end the process.

[0234] Dynamic planning stage:

[0235] Update the untraveled sub-section in the second planned path, and determine whether the update is successful. If the update is successful, control the vehicle to travel based on the updated second planned path, and during the driving process, detect in real time whether the posture difference between the vehicle terminal posture indicated by the updated second driving path and the vehicle terminal posture indicated by the perceived parking space is greater than the posture threshold. If so, return to the step of updating the untraveled sub-section in the second planned path, otherwise determine whether to park in the parking space, and if not, continue to determine whether the posture difference is greater than the posture threshold, that is, continue to detect whether the posture difference is greater than the posture threshold before parking in the parking space; if the update fails, control the vehicle to travel to the parking point, and return to the step of generating the second planned path in the static planning stage.

[0236] Corresponding to the above parking method, an embodiment of the present application also provides a parking device.

[0237] See also Fig. 9 , is a schematic diagram of the structure of a parking device provided in an embodiment of the present application, the device includes the following modules:

[0238] A first driving path generating module 901, configured to generate a first driving path to a gear shifting location based on sensor data collected by sensors carried by the vehicle and a pre-built parking space map of a parking space to be parked, and control the vehicle to travel based on the first driving path;

[0239] A second driving path generating module 902 is configured to sense obstacles around the vehicle and the parking space to be parked based on the sensor data if a gear shifting condition is met, generate a second driving path for the vehicle to be parked in the parking space to be parked based on the sensed obstacles and the sensed parking space, control the vehicle to switch gears and park in the parking space to be parked based on the second driving path;

[0240] The path updating module 903 is used to update the untraveled section in the second driving path based on the perceived obstacles and the perceived parking space if the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the perceived parking space is greater than a preset posture threshold during the driving process of the vehicle based on the second driving path.

[0241] As can be seen from the above, when the solution provided in the embodiment of the present application is applied for parking, a first driving path to the gear shift location is generated based on sensor data and a pre-built parking space map of the parking space to be parked, and the vehicle is controlled to travel based on the first driving path. When the gear shift conditions are met, a second driving path for parking the vehicle in the parking space can be generated based on the perceived obstacles and the perceived parking space, and the vehicle is controlled to successfully park in the parking space based on the second driving path, thereby realizing automatic parking and bringing convenience to the car owner.

[0242] Among them, the first driving path is generated based on sensor data and a pre-constructed parking map of the parking space to be parked, that is, the gear shifting location recorded in the first driving path is determined based on the above sensor data and the pre-constructed parking map, so that a more optimal gear shifting location suitable for gear switching can be determined, thereby improving parking efficiency.

[0243] In addition, when the vehicle is traveling based on the second driving path, the untraveled section in the second driving path can also be updated based on the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the perceived parking space. Since the distance of the vehicle to the parking space is getting closer and closer during the driving process based on the second driving path, it can be considered that the parking space dynamically perceived based on the real-time collected sensor data is becoming more and more accurate, that is, it is considered that the position of the dynamically perceived parking space is closer to the position of the real parking space. Therefore, when the difference between the vehicle terminal posture indicated by the second driving path and the vehicle terminal posture indicated by the perceived parking space is greater than the preset posture threshold, it can be considered that the deviation between the perceived parking space based on which the original second driving path is generated and the dynamically perceived parking space is large. Therefore, when the untraveled section in the second driving path is updated at this time, the parking path can be adjusted in real time according to the real-time perceived parking space, which reduces the number of parking rubbing and improves the accuracy of parking.

[0244] In one embodiment of the present application, before the first driving path generating module 901, it also includes:

[0245] A first cruise module, used to control the vehicle to cruise based on the sensor data and the driving path information recorded in the pre-built map;

[0246] A first distance obtaining module, used for obtaining a first distance between the vehicle and the parking space to be parked based on the sensor data during cruising;

[0247] The distance judgment module is used to trigger the first driving path generation module 901 if the first distance is less than a preset distance threshold.

[0248] As can be seen from the above, when the solution provided by the embodiment of the present application is used for parking, the vehicle can be controlled to cruise based on the sensor data and the driving path information recorded in the pre-built map, and during the cruising process, the first distance between the vehicle and the parking space to be parked is obtained, so that when the first distance is less than the preset distance threshold, the first driving path to the shifting location can be generated, that is, the first driving path can be generated when the distance between the vehicle and the parking space is small. Since the sensor mounted on the vehicle has a more accurate perception result of the information around the parking space when the distance between the vehicle and the parking space is small, generating the first driving path at this time can improve the accuracy of the generated first driving path.

[0249] In addition, this can generate a first driving path to a shift location before the vehicle reaches the cruising end point, which is helpful for determining a shift location earlier than the cruising end point, achieving early reversing, and thus improving parking efficiency.

[0250] In one embodiment of the present application, the first distance obtaining module is specifically used to obtain, during cruising, the distance between the rear wheel axle of the vehicle and the parking boundary line of the parking space to be parked, based on the sensor data, as the first distance between the vehicle and the parking space to be parked.

[0251] In this way, the distance between the rear axle of the vehicle and the parking boundary line of the parking space can be used as the first distance. Since the vehicle generally parks into the parking space by reversing, the rear wheels of the vehicle will first touch the parking boundary line of the parking space. Therefore, the distance between the rear axle of the vehicle and the parking boundary line of the parking space can more accurately represent the distance between the vehicle and the parking space, thereby improving the accuracy of subsequently generating the first driving path based on the first distance.

[0252] In one embodiment of the present application, if the first driving path generation fails, the device further includes:

[0253] An endpoint judgment module, used to judge whether the vehicle has traveled to the cruise endpoint recorded in the pre-built map, and if not, trigger the second cruise module;

[0254] The second cruise module is used to control the vehicle to continue cruising based on the sensor data and the driving path information recorded in the pre-built parking map.

[0255] In the event that the generation of the first driving path fails, if the vehicle has not yet reached the cruise destination recorded in the pre-built map, the vehicle can be controlled to continue cruising. This is conducive to determining a new opportunity to meet the requirement that the first distance is less than the preset distance threshold during the subsequent cruising process, thereby regenerating the first driving path and improving the completeness and rationality of the solution.

[0256] In one embodiment of the present application, the gear shifting condition includes at least one of the following conditions:

[0257] The vehicle drives to a cruising destination recorded in the pre-built map; the vehicle drives to the gear shifting location based on the first driving path; and the vehicle stops while driving based on the first driving path.

[0258] The gear shifting point of the above-mentioned path is generally the preferred gear shifting point determined by the vehicle owner. Therefore, when the vehicle reaches the above-mentioned cruise end point, it is determined that the gear shifting conditions are met and subsequent path planning is carried out. This can avoid the situation where the vehicle misses the preferred gear shifting point due to continuing to move forward, which is conducive to improving parking efficiency.

[0259] The above-mentioned shift location is a preferred shift location determined in the process of generating the first driving path. When the vehicle travels to the above-mentioned shift location, it can be determined that the vehicle meets the shift conditions so as to perform subsequent path planning, which is conducive to improving parking efficiency.

[0260] If the vehicle stops, it means that a serious path congestion occurs while the vehicle is driving along the first driving path, and the vehicle cannot continue driving based on path fine-tuning. Therefore, determining that the vehicle meets the gear shifting conditions at this time and performing subsequent path planning can avoid the vehicle from being stagnant for a long time, which is conducive to improving parking efficiency.

[0261] In one embodiment of the present application, the first driving path generation module 901 is specifically used to generate a complete driving path for the vehicle to park in the parking space to be parked based on sensor data collected by sensors carried by the vehicle and a pre-built parking space map of the parking space to be parked; and intercept the path before the gear shift location recorded in the complete driving path as the first driving path to the gear shift location.

[0262] In this way, a complete driving path for parking the vehicle in the parking space is first generated, and then the path before the gear shift location recorded in the complete driving path is intercepted, so that the first driving path to the gear shift location can be easily obtained without relying on the prediction of the gear shift location, thereby improving the generation efficiency of the first driving path.

[0263] In one embodiment of the present application, a driving path is generated according to the following modules: wherein the driving path includes: the first driving path and / or the second driving path:

[0264] An obstacle sensing module, configured to sense obstacles around the vehicle based on the sensor data and obtain an avoidance priority representation value of the sensed obstacles;

[0265] An obstacle classification module is used to classify the perceived obstacles to obtain high-priority obstacles whose avoidance priority characterization values ​​are greater than a preset characterization value threshold and low-priority obstacles whose avoidance priority characterization values ​​are not greater than the characterization value threshold;

[0266] A boundary line obtaining module, used to construct a parking area based on high priority obstacles and obtain a boundary line of the parking area;

[0267] The third driving path generation module is used to generate a driving path that does not collide with the boundary line and low-priority obstacles.

[0268] As can be seen from the above, when the solution provided in the embodiment of the present application is used to generate a driving path, first, the avoidance priority representation value of the perceived obstacle is obtained, and the perceived obstacle is classified based on the avoidance priority representation value to obtain high-priority obstacles and low-priority obstacles. Based on the high-priority obstacles, a parking area is constructed, and the boundary line of the parking area is obtained, so that a driving path that does not collide with the boundary line and the low-priority obstacles can be generated, which can reduce the occurrence of collisions between the vehicle and obstacles during the driving process based on the generated driving path, thereby improving the safety of the parking process.

[0269] In addition, the parking area is constructed based on high-priority obstacles. The boundary line of the parking area is a continuous line segment, while the positions of the high-priority obstacles are discrete points. Therefore, ensuring that the generated path does not collide with the above boundary line can further reduce the probability of the generated path colliding with high-priority obstacles compared to ensuring that the generated path does not collide with each high-priority obstacle, thereby further improving the rationality and safety of the parking process.

[0270] In one embodiment of the present application, the third driving path generation module includes:

[0271] A candidate path generation submodule, for generating a candidate path based on the boundary line and low priority obstacles;

[0272] A projection submodule, used for projecting the boundary line and low-priority obstacles onto a grid map to obtain a grid map recording obstacle information;

[0273] A collision path determination submodule, used to determine a collision path in a candidate path based on a collision condition between the candidate path and the obstacle information in the grid map;

[0274] The driving path selection submodule is used to select a driving path from candidate paths other than the collision path.

[0275] It can be seen that after projecting the boundary lines and low-priority obstacles to the grid map, based on the grids where the boundary lines and low-priority obstacles are located in the grid map, the collision situation between the candidate path and the boundary lines and low-priority obstacles in the grid map can be easily determined, so that the collision path in the candidate path can be quickly determined based on the collision situation.

[0276] In one embodiment of the present application, the collision path determination submodule is specifically used to determine whether the candidate path collides with the grid where the low-priority obstacle is located in the grid map; if so, directly determine the candidate path as the collision path; if not, when the candidate path collides with both the grid where the boundary line is located in the grid map and the vector line of the boundary line, determine the candidate path as the collision path.

[0277] It can be seen that, on the one hand, since low-priority obstacles have a smaller impact on vehicles parking in parking spaces, they are generally farther away from parking spaces, and the accuracy requirements for the perception results of low-priority obstacles are also lower. Therefore, if the candidate path collides with the grid where the low-priority obstacle is located in the grid map, it can be directly considered that the candidate path collides with the low-priority obstacle, that is, the candidate path is a collision path, which improves the efficiency of determining the candidate path.

[0278] On the other hand, since high-priority obstacles are obstacles that have a greater impact on vehicles parking in parking spaces, they are generally closer to parking spaces, and the accuracy requirements for the perception results of high-priority obstacles are also higher, so the accuracy of the boundary line of the parking area determined based on high-priority obstacles is higher. Based on this, when the candidate path collides with the grid where the boundary line is located in the grid map, an accurate comparison between the candidate path and the vector line of the boundary line can also be performed. If the candidate path collides with the grid where the boundary line is located in the grid map and the vector line of the boundary line, it is determined that the candidate path has collided with the boundary line, that is, the candidate path is determined to be a collision path, which improves the accuracy of determining the candidate path.

[0279] In one embodiment of the present application, the obstacle perception module includes:

[0280] an information acquisition submodule, for sensing obstacles around the vehicle based on the sensor data, and obtaining obstacle categories of the sensed obstacles, second distances between each side of the obstacle and the parking space, and confidence correction coefficients of the second distances corresponding to each side, wherein the parking space includes: a parking space in the pre-built parking space map and / or a sensed parking space;

[0281] The characterization value acquisition submodule is used to obtain the avoidance priority characterization value of the perceived obstacle based on the obtained obstacle category, the second distance and the confidence correction coefficient.

[0282] It can be seen that the obstacle avoidance priority representation value is determined based on the obstacle category of the perceived obstacle, the second distance between each side of the obstacle and the parking space, and the confidence correction coefficient of the second distance corresponding to each side. In this way, the determined avoidance priority representation value is related to the obstacle category and the second distance between each side of the obstacle and the parking space, thereby obtaining a more comprehensive and accurate avoidance priority representation value.

[0283] In one embodiment of the present application, the characterization value obtaining submodule is specifically used to obtain the avoidance priority characterization value of each perceived obstacle according to the following expression:

[0284]

[0285] Among them, W i represents the avoidance priority representation value of obstacle i, c i represents the priority coefficient of the obstacle category of the obstacle i, V represents the position of the parking space, E i,j represents the position of the jth edge of the obstacle i, dis(E i,j , V) represents the second distance between the jth edge of the obstacle i and the parking space, δ i,j represents the confidence correction coefficient of the second distance, α represents the preset scaling factor, and W max Indicates the preset maximum avoidance priority representation value.

[0286] It can be seen that after obtaining the obstacle category of the perceived obstacle, the second distances between each side of the obstacle and the parking space, and the confidence correction coefficient of the second distance corresponding to each side, the avoidance priority representation value of the obstacle can be quickly and accurately obtained based on the above expression.

[0287] In one embodiment of the present application, the sensor includes multiple types of sensors;

[0288] The boundary line obtaining module is specifically used to construct a parking area based on the perception results of high-priority obstacles corresponding to each type of sensor, wherein the perception result corresponding to each type of sensor is: the result of using the data collected by the sensor of this type to perceive the obstacle, and obtain the boundary line of the parking area;

[0289] The third driving path generation module is specifically used to generate a driving path that does not collide with the boundary line and the low-priority obstacles based on the perception results of the high-priority obstacles corresponding to each type of sensor and the perception results of the low-priority obstacles corresponding to the preset type of sensor.

[0290] It can be seen that the parking area is constructed based on the perception results of high-priority obstacles corresponding to various types of sensors. Since high-priority obstacles are obstacles that have a greater impact on vehicles parking into parking spaces, comprehensively considering the perception results of high-priority obstacles corresponding to various types of sensors can improve the accuracy of perceiving high-priority obstacles, thereby improving the accuracy of the parking area constructed based on the perception results of high-priority obstacles corresponding to various types of sensors.

[0291] In addition, the perception results of low-priority obstacles corresponding to the preset type of sensors are considered when generating the above driving path. Since low-priority obstacles are obstacles that have a smaller impact on the vehicle when parking in the parking space, the perception accuracy requirement for low-priority obstacles is relatively low. Therefore, only obtaining the perception results of low-priority obstacles corresponding to the preset type of sensors can improve the efficiency of perceiving high-priority obstacles, thereby improving the efficiency of generating the above driving path based on the perception results of low-priority obstacles.

[0292] In one embodiment of the present application, before the first driving path generating module 901, it also includes:

[0293] A parking function prompt module, configured to trigger an automatic parking function ready prompt after detecting that the vehicle is in the area described by the pre-built parking map;

[0294] The indication response module is used to trigger the first driving path generation module 901 in response to the automatic parking function activation indication.

[0295] It can be seen that after detecting that the vehicle is in the area described by the pre-built parking map, the automatic parking function readiness prompt can be actively triggered, so that the parking process is executed after receiving the automatic parking function activation instruction. That is, it does not rely on the owner to actively initiate the parking function activation instruction, but pushes the parking function to the owner automatically after detecting that the parking requirements are met, so that the owner only needs a simple confirmation to start parking, making the entire parking process more convenient and user-friendly, and improving the user experience during the parking process.

[0296] In one embodiment of the present application, a pre-built parking space map is constructed in the following manner:

[0297] After detecting that the vehicle is parked in the parking space to be parked, determining whether the parking space to be parked is a fixed parking space for the vehicle based on historical parking information of the vehicle in the parking space to be parked; if yes, triggering a map building prompt for the parking space to be parked; and in response to the map building instruction, building a parking space map of the parking space to be parked based on cached sensor data to obtain the pre-built parking space map.

[0298] It can be seen that in this way, when the parking space is determined to be a fixed parking space for the vehicle based on the historical parking information of the parking space, the car owner can be actively reminded whether to build a parking space map, and the map can be built after confirmation, without the need for the car owner to perform additional operations, thereby improving the user experience; in addition, after confirmation, the map can be built directly based on the cached sensor data. This process is imperceptible to the user and is more convenient for the car owner, thereby improving the practicality of the solution.

[0299] In one embodiment of the present application, the pre-built parking space map of the parking space to be parked includes the fixed parking space of the vehicle, and the fixed parking space of the vehicle includes:

[0300] The parking space management system synchronizes the paid parking space to the vehicle; and / or,

[0301] a pre-stored parking space received by an interactive system in an onboard system of the vehicle; and / or,

[0302] Frequently used parking spaces are automatically stored by the vehicle based on historical parking information.

[0303] In the solutions provided in the embodiments of the present application, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are all carried out under the premise of the user's knowledge and authorization, and comply with the relevant laws and regulations and do not violate public order and good morals.

[0304] The present application also provides a vehicle-mounted control device, such as Fig.10 As shown, including:

[0305] Memory 1001, used for storing computer programs;

[0306] The controller 1002 is used to implement the above parking method when executing the program stored in the memory 1001.

[0307] Furthermore, the above-mentioned vehicle-mounted control device may further include a communication bus and / or a communication interface, and the controller 1002, the communication interface, and the memory 1001 communicate with each other via the communication bus.

[0308] The communication bus mentioned in the above-mentioned vehicle-mounted control device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0309] The communication interface is used for communication between the above-mentioned vehicle-mounted control device and other devices.

[0310] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned controller.

[0311] The above-mentioned controller can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0312] In another embodiment provided in the present application, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a controller, the above parking method is implemented.

[0313] In another embodiment provided in the present application, a vehicle is further provided, wherein a computer program is stored in the vehicle, and the computer program implements the above parking method when executed by a controller.

[0314] In another embodiment provided by the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute any parking method in the above embodiments.

[0315] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a solid-state hard disk (SSD), etc.

[0316] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0317] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, vehicle-mounted control device, and storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0318] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the protection scope of the present application.

Claims

1. A parking method, characterized in that: The method comprises: generating a first driving path to a shifting location based on sensor data collected by sensors carried by the vehicle and a pre-built parking space map of a parking space to be parked, and controlling the vehicle to travel based on the first driving path; If the gear shifting condition is met, obstacles around the vehicle are sensed based on the sensor data, and the parking space to be parked is sensed, a second driving path is generated for the vehicle to be parked in the parking space to be parked based on the sensed obstacles and the sensed parking space, and the vehicle is controlled to switch gears and park in the parking space to be parked based on the second driving path; During the driving of the vehicle based on the second driving path, if a difference between a vehicle terminal position indicated by the second driving path and a vehicle terminal position indicated by a sensed parking space is greater than a preset position threshold, updating an untraveled section in the second driving path based on a sensed obstacle and a sensed parking space; A driving path is generated in the following manner: wherein the driving path includes: the first driving path and / or the second driving path: based on the sensor data, obstacles around the vehicle are sensed, and avoidance priority characterization values ​​of the sensed obstacles are obtained; the sensed obstacles are classified to obtain high-priority obstacles whose avoidance priority characterization values ​​are greater than a preset characterization value threshold and low-priority obstacles whose avoidance priority characterization values ​​are not greater than the characterization value threshold; based on the high-priority obstacles, a parking area is constructed, and a boundary line of the parking area is obtained; and a driving path that does not collide with the boundary line and low-priority obstacles is generated.

2. The method according to claim 1, characterized in that Before generating the first driving path to the gear shift location based on the sensor data collected by the sensors mounted on the vehicle and the pre-constructed parking map of the parking space to be parked, the method further includes: controlling the vehicle to cruise based on the sensor data and the driving path information recorded in the pre-constructed parking map; during the cruising process, obtaining a first distance between the vehicle and the parking space to be parked based on the sensor data; if the first distance is less than a preset distance threshold, executing the step of generating the first driving path to the gear shift location based on the sensor data collected by the sensors mounted on the vehicle and the pre-constructed parking map of the parking space to be parked.

3. The method according to claim 2, characterized in that The obtaining, based on the sensor data, a first distance between the vehicle and the parking space to be parked, comprises: obtaining, based on the sensor data, a distance between a rear wheel axle of the vehicle and a parking boundary line of the parking space to be parked as the first distance between the vehicle and the parking space to be parked; and / or If the first driving path generation fails, the method further includes: determining whether the vehicle has traveled to a cruising destination recorded in the pre-constructed parking map; if not, controlling the vehicle to continue cruising based on the sensor data and the driving path information recorded in the pre-constructed parking map; and / or The shifting condition includes at least one of the following conditions: the vehicle travels to the cruising end point recorded in the pre-built parking map; the vehicle travels to the shifting location based on the first driving path; the vehicle stops while traveling based on the first driving path.

4. The method according to claim 1, characterized in that: The generating of a driving path that does not collide with the boundary line and the low-priority obstacle includes: generating a candidate path based on the boundary line and the low-priority obstacle; projecting the boundary line and the low-priority obstacle onto a grid map to obtain a grid map recording obstacle information; determining a collision path in the candidate path based on a collision condition between the candidate path and the obstacle information in the grid map; and selecting a driving path from the candidate paths other than the collision path. and / or The obtaining of the avoidance priority representation value of the perceived obstacle includes: obtaining the obstacle category of the perceived obstacle, the second distance between each side of the obstacle and the parking space, and the confidence correction coefficient of the second distance corresponding to each side, wherein the parking space includes: a parking space in the pre-built parking space map and / or a perceived parking space; and obtaining the avoidance priority representation value of the perceived obstacle based on the obtained obstacle category, the second distance, and the confidence correction coefficient.

5. The method according to claim 4, characterized in that The method of determining a collision path in a candidate path based on a collision situation between the candidate path and the obstacle information in the grid map includes: determining whether the candidate path collides with a grid where a low-priority obstacle is located in the grid map; if yes, directly determining the candidate path as a collision path; if not, determining the candidate path as a collision path when the candidate path collides with both the grid where the boundary line is located in the grid map and the vector line of the boundary line.

6. The method according to claim 1, characterized in that The sensor includes multiple types of sensors; The method of constructing a parking area based on high-priority obstacles includes: Based on the perception results of high-priority obstacles corresponding to each type of sensor, a parking area is constructed, wherein the perception result corresponding to each type of sensor is: the result of perception of the obstacle using the data collected by the sensor of that type; The generating of a driving path that does not collide with the boundary line and low-priority obstacles includes: Based on the perception results of the high-priority obstacles corresponding to the sensors of each type and the perception results of the low-priority obstacles corresponding to the sensors of the preset type, a driving path is generated that does not collide with the boundary line and the low-priority obstacles.

7. The method according to any one of claims 1 to 6, characterized in that Before generating the first driving path to the gear shift location based on the sensor data collected by the sensors on the vehicle and the pre-built parking map of the parking space to be parked, the method further includes: after detecting that the vehicle is in the area described by the pre-built parking map, triggering an automatic parking function ready prompt; in response to the automatic parking function start indication, executing the step of generating the first driving path to the gear shift location based on the sensor data collected by the sensors on the vehicle and the pre-built parking map of the parking space to be parked; and / or The pre-constructed parking space map is constructed in the following manner: after detecting that the vehicle is parked in the parking space to be parked, determining whether the parking space to be parked is a fixed parking space for the vehicle based on historical parking information of the vehicle in the parking space to be parked; if yes, triggering a map construction prompt for the parking space to be parked; in response to the map construction instruction, constructing a parking space map of the parking space to be parked based on cached sensor data to obtain the pre-constructed parking space map.

8. The method according to any one of claims 1 to 6, characterized in that The pre-built parking space map of the parking space to be parked includes the fixed parking space of the vehicle, and the fixed parking space of the vehicle includes: The parking space management system synchronizes the paid parking space to the vehicle; and / or, a pre-stored parking space received by an interactive system in an onboard system of the vehicle; and / or, Frequently used parking spaces are automatically stored by the vehicle based on historical parking information.

9. A vehicle-mounted control device, characterized in that: include: Memory, used to store computer programs; A controller, used to implement any method described in claims 1-8 when executing a program stored in a memory.

10. A vehicle, characterized in that: A computer program is stored in the vehicle, and when the computer program is executed by the controller, the method described in any one of claims 1 to 8 is implemented.

Images