Vehicle parking method and apparatus, and vehicle

Abstract

A vehicle parking method and apparatus, and a vehicle. The method is applied to a domain controller. The method comprises: when a current position of a vehicle is located in a parking area and the distance between the current position and a target parking space in the parking area is greater than a first preset distance, acquiring first image information captured by a first camera and second image information captured by a second camera (S301), distance ranges captured by the first camera and the second camera being different; sending the first image information and a processing request instruction to a vehicle controller (S302), the processing request instruction being used for instructing the vehicle controller to process the first image information to obtain first image feature information (S304); receiving the first image feature information sent by the vehicle controller (S306); and, on the basis of the first image feature information and the second image information, generating a first parking route (S307). According to the parking method, the problem of collisions that may occur during long-distance parking can be solved, thereby improving the safety performance of the vehicle while reducing the costs of the vehicle.

Description

Vehicle parking method, device, and vehicle

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office on November 24, 2023, with application number 202311587153.6 and application name “Parking method, device, and vehicle for vehicle”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of vehicles, and in particular to a parking method, a parking device, and a vehicle. Background Art

[0003] An intelligent driving system is a system that uses sensors, controllers, actuators, communication modules, and other equipment on a vehicle to assist the user in controlling the vehicle. Currently, cars equipped with intelligent driving systems are increasingly entering the consumer market, and their adoption is increasing.

[0004] Smart parking has become a standard feature in many intelligent driving systems. However, the domain controllers in current intelligent driving systems use low-computing system-on-chips (SOCs). During parking, the SOCs can only identify obstacles within a certain range of the vehicle. Consequently, in long-distance parking scenarios, when planning a parking route, the vehicle may not be able to effectively identify obstacles within a sufficient range, potentially leading to collisions.

[0005] Application Contents

[0006] The present application provides a parking method, device, and vehicle for a vehicle, wherein the method can improve the safety performance of the vehicle while reducing the cost of the vehicle.

[0007] In a first aspect, a vehicle parking method is provided. The method is applied to a domain controller of the vehicle. The vehicle includes: a first camera, a second camera, a vehicle controller, and the domain controller. The first camera, the second camera, and the vehicle controller are respectively connected to the domain controller. The method includes:

[0008] When the current position of the vehicle is within the parking area and the distance between the current position and the target parking space within the parking area is greater than a first preset distance, first image information captured by a first camera and second image information captured by a second camera are obtained, and the distance ranges captured by the first camera and the second camera are different; the first image information and a processing request instruction are sent to the vehicle controller, the processing request instruction is used to instruct the vehicle controller to process the first image information to obtain first image feature information; the first image feature information sent by the vehicle controller is received; and a first parking route is generated based on the first image feature information and the second image information.

[0009] Thus, in an embodiment of the present application, when the current position of the vehicle is within the parking area and the distance between the current position and the target parking space within the parking area is greater than a first preset distance, the first image information captured by the first camera and the second image information captured by the second camera are obtained, and the first image information and a processing request instruction are sent to the vehicle controller, where the processing request instruction is used to instruct the vehicle controller to process the first image information to obtain first image feature information; receive the first image feature information sent by the vehicle controller; and generate a first parking route based on the first image feature information and the second image information. Thus, in an embodiment of the present application, the domain controller can process image information within different ranges captured during long-distance parking, thereby resolving the problem of the domain controller being unable to simultaneously process multiple types of image information when using a low-computing power SOC. The parking route in the long-distance parking scenario is planned based on the received first image feature information and second image information, thereby resolving the potential collision problem during long-distance parking, and thereby improving the safety performance of the vehicle while reducing vehicle costs.

[0010] In a second aspect, a vehicle parking method is provided. The method is applied to a vehicle controller. The vehicle includes: a first camera, a second camera, a vehicle controller, and a domain controller. The first camera, the second camera, and the vehicle controller are respectively connected to the domain controller. The method includes:

[0011] Receiving first image information and a processing request instruction sent by a domain controller; wherein the first image information is captured by a first camera;

[0012] Extracting features from the first image information based on the processing request instruction to obtain first image feature information;

[0013] The first image feature information is sent to the domain controller, so that the domain controller generates a first parking route based on the first image feature information and the second image information; wherein the second image information is obtained by taking pictures by the second camera, and the distance ranges taken by the first camera and the second camera are different.

[0014] Thus, in this embodiment of the present application, when the vehicle's current position is within a parking area and the distance between the current position and the target parking space within the parking area is greater than a first preset distance, the vehicle controller can process the first image information sent by the domain controller to obtain first image feature information, and then send this first image feature information to the domain controller, so that the domain controller can generate a first parking route based on the first image feature information and the second image information. This embodiment of the present application can process the first image information captured during long-distance parking by the vehicle controller and send the obtained first image feature information back to the domain controller, thereby resolving the issue of the domain controller being unable to simultaneously process multiple types of image information when using a low-computing SOC.

[0015] In a third aspect, a vehicle parking device is provided. The device is applied to a domain controller. The vehicle includes: a first camera, a second camera, a vehicle controller, and the domain controller. The first camera, the second camera, and the vehicle controller are respectively connected to the domain controller. The device includes:

[0016] a first acquisition module configured to acquire, when the current position of the vehicle is within the parking area and the distance between the current position and a target parking space within the parking area is greater than a first preset distance, first image information captured by the first camera and second image information captured by the second camera; the first camera and the second camera capture different distance ranges;

[0017] a first sending module, configured to send first image information and a processing request instruction to the vehicle controller, wherein the processing request instruction is used to instruct the vehicle controller to process the first image information to obtain first image feature information;

[0018] A first receiving module is used to receive first image feature information sent by the vehicle controller;

[0019] The route generation module is configured to generate a first parking route based on the first image feature information and the second image information.

[0020] In a fourth aspect, a vehicle parking device is provided. The device is applied to a vehicle controller. The vehicle includes: a first camera, a second camera, a vehicle controller, and a domain controller. The first camera, the second camera, and the vehicle controller are respectively connected to the domain controller. The device includes:

[0021] A second receiving module is configured to receive first image information and a processing request instruction sent by the domain controller; wherein the first image information is captured by the first camera;

[0022] a feature extraction module, configured to extract features from the first image information based on the processing request instruction to obtain first image feature information;

[0023] The second sending module is used to send the first image feature information to the domain controller, so that the domain controller generates a first parking route based on the first image feature information and the second image information; wherein the second image information is obtained by capturing the second camera, and the first camera and the second camera have different shooting distance ranges.

[0024] According to a fifth aspect, a vehicle is provided, comprising:

[0025] a memory for storing executable program code;

[0026] A processor is used to call and run the executable program code from the memory, so that the vehicle executes any one of the methods described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] FIG1 is a schematic diagram of the connection structure of various components of a vehicle in a parking method provided by an embodiment of the present application;

[0028] FIG2 is a schematic diagram of the internal structure of a domain controller provided in an embodiment of the present application;

[0029] FIG3 is a flow chart of a parking method provided in an embodiment of the present application;

[0030] FIG4 is a schematic diagram of an application scenario of a parking method provided in an embodiment of the present application;

[0031] FIG5 is a flow chart of another parking method provided in an embodiment of the present application;

[0032] FIG6 is a schematic diagram of an in-vehicle interface of another parking method provided in an embodiment of the present application;

[0033] FIG7 is a schematic diagram of a vehicle interface displaying multiple parking spaces provided by an embodiment of the present application;

[0034] FIG8 is a schematic diagram of the connection relationship between the domain controller and other devices in the vehicle in another parking method provided by an embodiment of the present application;

[0035] FIG9A is a schematic structural diagram of a parking device provided in an embodiment of the present application;

[0036] FIG9B is a schematic structural diagram of another parking device provided in an embodiment of the present application;

[0037] FIG10 is a schematic structural diagram of a vehicle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0038] Figure 1 is an application scenario diagram of the parking method provided by this application. The multi-sensor fusion solution adopted by the intelligent driving system can improve the safety factor of the entire vehicle and ensure the driving safety of users. The number of sensors in the multi-sensor fusion solution is large. For a single camera sensor, the vehicle is usually provided with a forward-looking camera for driving (for example: 1) and a surround-view camera for parking (for example: 4 front, rear, left and right). In addition, a radar system for detecting point cloud information of obstacles in the scene can be provided on the vehicle to form a driving perception solution, so that when all camera-type sensors fail (this failure may be caused by natural reasons, such as dense fog, or human interference), a certain degree of redundancy can be provided by the radar sensor.

[0039] The vehicle controller can be connected to the onboard display, which can display images of the vehicle's surroundings captured by the surround-view camera. The vehicle controller can also send data such as the vehicle's current location to the domain controller, allowing the domain controller to plan driving routes based on the vehicle's current location. The domain controller can control multiple electronic control units (ECUs) separately and transmit data via the Ethernet protocol through the gateway. For example, data for controlling vehicle acceleration, deceleration, or turning can be transmitted through the lateral and longitudinal control interfaces.

[0040] Figure 2 is a schematic diagram of the internal structure of the domain controller used in the parking method provided by this application. The domain controller may include: a SOC, a microcontroller unit (MCU), and an Ethernet module. The SOC can be a low-computing SOC; a low-computing SOC refers to a chip with limited computing power (for example, a computing power of less than 30 trillion operations per second (TOPS)). The SOC has a video image processing function, which can time-share the images captured by the camera. In other words, it can time-share the vehicle's driving process and parking process. For example, the driving process uses the image information collected by the front-view camera, and the parking process uses the image information collected by the surround-view camera. The MCU is a controller that can be used for planning and control. It can be connected to the SOC, radar system, vehicle controller, etc., and then detect the vehicle status and user-related operations through information such as the vehicle chassis and body system, or perform path planning, complete driving assistance driving, parking assistance driving and other functions based on the image data provided by the SOC. The specific settings can be based on actual needs.

[0041] Driving assistance: This function is designed to assist users in completing lateral and longitudinal vehicle control while driving on the road, thereby reducing user operation and reducing user fatigue. This function primarily utilizes a forward-looking camera and millimeter-wave radar. The forward-looking camera is used for image-based detection of road boundaries (e.g., lane markings) and image-based detection of vehicles or pedestrians. The millimeter-wave radar emits electromagnetic waves to detect point cloud information of obstacles and calculates obstacle-related information based on this point cloud information to generate driving trajectories and provide longitudinal assistance.

[0042] Parking Assist: The automated parking feature integrates ultrasonic sensors with a surround-view camera for panoramic photography to detect obstacles near the vehicle and search for marked parking spaces. By embedding a neural network learning algorithm into the domain controller, it can identify objects that could affect the parking process, such as internal ground locks and no-parking signs. The domain controller then plans the parking trajectory, allowing the user to automatically enter the target parking space horizontally, vertically, or diagonally without having to manipulate the steering wheel, gears, or brakes. The intelligent driving system also supports voice commands, allowing the user to initiate parking with a single command.

[0043] Next, the parking method provided in the embodiment of the present application is introduced with reference to the application scenario diagram of the parking method introduced in FIG1 and the internal structure diagram of the domain controller introduced in FIG2 .

[0044] In one embodiment, as shown in FIG3 , the present application provides a flow chart of a method for parking a vehicle. The vehicle may include: a first camera, a second camera, a vehicle controller, and a domain controller, wherein the first camera, the second camera, and the vehicle controller are respectively connected to the domain controller. As shown in FIG3 , the parking method may include the following steps:

[0045] S301, when the current position of the vehicle is within the parking area and the distance between the current position and the target parking space within the parking area is greater than a first preset distance, the domain controller obtains first image information captured by the first camera and second image information captured by the second camera.

[0046] Among them, in the embodiment of the present application, the distance range of the first camera and the second camera are different.

[0047] Exemplarily, when the shooting range includes a close range and a long range, the shooting range of the first camera and the shooting range of the second camera are complementary. For example, when the shooting range of the first camera is a long range, the shooting range of the second camera is a close range.

[0048] The parking area is the area where the user's pre-configured target parking space is located, for example, a parking lot, residential complex, or school with available parking spaces. The target parking space represents a designated parking space within the parking area, for example, a user's designated parking space within a parking lot. The first preset distance represents the preset distance from the boundary of the target parking space. For example, a first preset distance of 6 meters indicates a distance of 6 meters from the boundary of the target parking space.

[0049] It can be understood that the first camera and the second camera in the embodiment of the present application can respectively capture scene image information within different distance ranges. The first camera can be used to capture scene image information within a close range around the vehicle, and the second camera can be used to capture scene image information within a relatively long distance from the vehicle (for example, about 150 meters); or, the first camera can be used to capture scene image information within a long distance range, and the second camera can be used to capture scene image information within a close range. This application does not impose any restrictions.

[0050] Specifically, the embodiment of the present application can use a surround-view camera to capture scene image information within a range of 6 to 8 meters around the vehicle, and use a forward-looking camera to capture scene image information within a range of about 150 meters in front of the vehicle.

[0051] In a specific example, when a vehicle enters the parking lot entrance and the distance between the vehicle and the target parking space is greater than 6 meters, long-range image information captured by the front-view camera and short-range image information captured by the surround-view camera can be obtained.

[0052] S302: The domain controller sends first image information and a processing request instruction to the vehicle controller. The processing request instruction is used to instruct the vehicle controller to process the first image information.

[0053] S303: The vehicle controller receives the first image information and the processing request instruction sent by the domain controller.

[0054] S304: The vehicle controller extracts features from the first image information based on the processing request instruction to obtain first image feature information.

[0055] Possibly, the vehicle controller in the embodiment of the present application may include a SOC-AI chip, which can use artificial intelligence (AI) to extract features from image information in the video captured by the camera and output information including but not limited to the object type, attributes, lane lines, etc. of the obstacle.

[0056] For example, convolutional neural networks, natural language processing, image segmentation, etc. are used to extract feature information such as pedestrians, vehicles, lane lines, or signs from image information.

[0057] See Figure 1. Possibly, the vehicle controller in the embodiment of the present application can extract feature information from the close-range image information directly received from the surround view camera.

[0058] See Figure 4. Possibly, the vehicle controller in the embodiment of the present application can also receive the long-distance image information captured by the front-view camera in a transparent manner through the domain controller, thereby extracting feature information from the long-distance image information.

[0059] In practical applications, the transparent transmission method described above is called transparent transmission technology, which means that during the data transmission process, the data or signal can maintain its original format, content, and characteristics without any form of change, processing, or interpretation. For example, in this embodiment of the application, the vehicle controller can receive data that is completely consistent with the domain controller.

[0060] It should be noted that during parking, the vehicle controller in this application only extracts features from a single type of image information, processing only long-range or short-range image information. The domain controller can then plan a parking route based on the received feature information and send relevant information for lateral vehicle control to the steering system, and relevant information for longitudinal vehicle control to the braking system and powertrain, respectively.

[0061] S305: The vehicle controller sends the first image feature information to the domain controller.

[0062] S306: The domain controller receives the first image feature information sent by the vehicle controller.

[0063] S307: The domain controller generates a first parking route based on the first image feature information and the second image information.

[0064] See Figure 2. While the vehicle is driving, if the MCU in the domain controller detects a user shifting into reverse gear, triggering a parking command, or the intelligent assistance system, based on location information, determines that the user is about to park, it triggers a signal to switch from driving to parking and transmits this signal to the SOC. Upon receiving the signal, the SOC performs image preprocessing and semantic segmentation on the received image information and sends the results to the MCU, which then generates a parking path planning route based on the results.

[0065] Specifically, the embodiment of the present application can extract the second image feature information from the second image information; based on the distance range corresponding to the first image feature information and the distance range corresponding to the second image feature information, the first image feature information and the second image feature information are spliced ​​to obtain the image feature information corresponding to the parking area; based on the image feature information corresponding to the parking area, a first parking route is generated.

[0066] It is understandable that the domain controller in the embodiment of the present application can obtain image feature information corresponding to the parking area by splicing image feature information corresponding to different distance ranges in the long-distance parking scenario, thereby performing global planning for the first parking route generated in the long-distance parking scenario. This not only improves the accuracy of the first parking route, but also avoids problems such as duplication of image features in some areas and excessive occupation of system resources when generating the first parking route based on unspliced ​​image feature information corresponding to different distance ranges.

[0067] Possibly, after receiving the switching signal, the domain controller's SOC in the embodiment of the present application can send an on command to the surround-view camera to control the camera's on-state and receive the image information captured by the surround-view camera. The SOC can also send a stop command and a first control command to the forward-view camera, causing the forward-view camera to stop sending image information to the SOC via the stop command and to send its image information to the vehicle controller via the domain controller's transparent transmission function via the first control command. In other words, the image information captured by the forward-view camera is extracted from its features by the SOC-AI chip in the vehicle controller, and the image information captured by the surround-view camera is processed by the domain controller's SOC.

[0068] Possibly, after receiving the switching signal, the domain controller's SOC in the embodiment of the present application can send a start instruction and a second control instruction to the surround-view camera, control the surround-view camera to turn on through the start instruction, and control the surround-view camera to send its captured image information to the vehicle controller through the second control instruction. The front-view camera continues to send its captured image information to the domain controller's SOC. In other words, the SOC-AI chip in the vehicle controller performs feature extraction on the image information captured by the surround-view camera, and the domain controller's SOC continues to perform feature extraction processing on the image information captured by the front-view camera.

[0069] Understandably, the forward-looking and surround-view cameras may not function properly in adverse weather conditions. However, radar systems have good penetration capabilities against rain, smoke, and dust, and can also accurately detect relevant information about moving obstacles, such as distance, relative speed, and azimuth. Therefore, the vehicle in the embodiments of this application may also be equipped with a radar system to more accurately plan the vehicle's route.

[0070] Specifically, the embodiment of the present application can determine point cloud feature information based on the point cloud information detected by the radar system through the domain controller, and determine relevant information of each obstacle in the target scene based on the point cloud feature information.

[0071] Specifically, a radar system can calculate a target's range and speed by transmitting radio waves and receiving reflected waves from the target. A radar system can include analog components such as a synthesizer, power amplifier, transmitter, low-noise amplifier, receiver, and mixer, as well as digital components such as an analog-to-digital converter and a digital signal processor. Based on the time-of-flight principle, a radar system calculates the distance to an object by the time difference between the transmitted and reflected signals. Based on the Doppler principle, when there is relative motion between a transmitted electromagnetic wave and a detected target, the frequency of the returned wave differs from the frequency of the transmitted wave. This frequency difference can be used to measure the target's relative speed relative to the radar. When the radar is operational, the synthesizer generates a linear frequency-modulated pulse, which is amplified and transmitted by the transmitter. The receiver captures the target's reflected pulse, amplifies it with low noise, and uses array signal processing to calculate the azimuth angle from the phase difference between the reflected pulses.

[0072] Although the radar system can distinguish moving obstacles from the surrounding scene based on speed, its perception of stationary obstacles is weak, and it is almost impossible to distinguish between gantries, signboards, and parked vehicles. In addition, the radar system is highly sensitive to metal, which often causes strong reflections from stationary objects such as manhole covers. Therefore, in actual applications, stationary objects are usually filtered out by the radar system, resulting in the inability to detect stationary obstacles. Therefore, the embodiment of the present application adopts a solution of fusing static image information + dynamic point cloud information to determine the obstacle information and related identification information in the target scene, so as to plan the parking route more accurately.

[0073] Specifically, the embodiment of the present application can also receive point cloud information sent by the radar system; perform feature extraction on the point cloud information to obtain point cloud feature information; and generate a first parking route based on the image feature information and point cloud feature information corresponding to the parking area.

[0074] Understandably, cameras are widely used in autonomous driving for obstacle detection and object segmentation, but they can only detect two-dimensional static information. Radar systems can detect an object's three-dimensional outline, speed, direction, and other information. Fusion of these complementary features provided by camera and radar systems can improve obstacle detection, semantic segmentation, and tracking capabilities, thereby enhancing the accuracy of parking route planning.

[0075] The radar system in the embodiment of the present application may include: a first radar and a second radar. The first radar and the second radar have different detection ranges.

[0076] Furthermore, in the embodiment of the present application, the first point cloud feature information and the second point cloud feature information can be spliced ​​based on the distance range corresponding to the first point cloud feature information and the distance range corresponding to the second point cloud feature information to obtain point cloud feature information corresponding to the parking area. A first parking route can be generated based on the image feature information corresponding to the parking area and the point cloud feature information corresponding to the parking area.

[0077] The first point cloud feature information is obtained by performing feature extraction on the first point cloud information collected by the first radar, and the second point cloud feature information is obtained by performing feature extraction on the second point cloud information collected by the second radar.

[0078] Possibly, the first radar in the embodiment of the present application may be a millimeter-wave radar (Millimeter-wave Radar), the second radar may be an ultrasonic radar (Ultrasonic Sensor System, USS), or the first radar may be an ultrasonic radar and the second radar may be a millimeter-wave radar. This application does not impose any restrictions.

[0079] Millimeter-wave radars are typically installed on the front bumper of a vehicle. They have a wavelength of millimeters, a detection range of tens to 150 meters, and can be used in inclement weather. Ultrasonic radars are typically installed at the rear of a vehicle. They have a wavelength of millimeters, a detection range of several meters to over ten meters, and can be used in inclement weather.

[0080] It can be understood that the embodiments of the present application can detect the point cloud information of long-distance obstacles through millimeter-wave radar, detect the point cloud information of short-distance obstacles through ultrasound, and extract features of the point cloud information of long-distance obstacles and the point cloud information of short-distance obstacles through MCU to obtain point cloud feature information of long-distance obstacles and point cloud feature information of short-distance obstacles, so as to identify the type of obstacle according to the point cloud feature information, and combine the recognition result with the recognition result of the image feature information to achieve the purpose of mutual integration of dynamic detection and static detection, thereby effectively identifying obstacles around the vehicle body and in distant scenes to ensure driving safety.

[0081] Specifically, the vehicle in the embodiment of the present application can use the memory parking function to enable automatic parking for parking route planning. The memory parking function refers to the pre-rough route planning of the target parking space in the user-specified area. For example, when the user drives home and needs to park the car in the garage or outdoor parking space in the community, after the user activates the memory parking function, the memory parking system will perform memory map learning, allowing the user to drive into the community at a certain speed. The co-processor of the vehicle controller and the domain controller will perform preliminary planning of the parking route, identify and record the spatial parking space, until the user drives the vehicle into the target parking space, completing the memory map learning. After that, the user can activate the memory parking function at the starting point of the memory map (i.e., the community gate) to complete the process of automatically parking the vehicle into the target parking space.

[0082] In a specific example, the user activates the memory parking function. When the vehicle enters the gate of the community, the vehicle performs the automatic parking function. The vehicle positioning system in the vehicle controller detects that the current distance between the vehicle and the target parking space is greater than 6 meters. The front-view camera continues to shoot to obtain long-distance image information, and sends the image information to the SOC-AI in the vehicle controller for feature extraction to obtain long-distance image feature information; the surround-view camera starts to shoot close-range image information and sends the image information to the SOC in the domain controller for feature extraction to obtain close-range image feature information. The MCU in the domain controller combines the received long-range image feature information, close-range image feature information, long-range point cloud feature information, and close-range point cloud feature information to identify static obstacles, dynamic obstacles, lane lines, signs and other information in the current scene, and plans the parking route based on the above recognition results.

[0083] In an embodiment of the present application, when the current position of the vehicle is within a target area and the distance between the current position and a target parking space within the target area is greater than a first preset distance, the first image information captured by the first camera and the second image information captured by the second camera can be obtained; first image feature information in the first image information can be extracted by the vehicle controller; and a first parking route can be planned by the domain controller based on the first image feature information and the second image information. Thus, in an embodiment of the present application, the vehicle controller and the domain controller can respectively process image information within different ranges captured during long-distance parking, thereby resolving the problem that a low-computing-power SOC cannot process multiple types of image information simultaneously. Furthermore, the domain controller can plan a parking route in a long-distance parking scenario based on the long-distance image feature information and the short-distance image information, thereby resolving the potential collision problem during long-distance parking and improving the vehicle's safety performance while reducing vehicle costs.

[0084] In some embodiments, FIG5 exemplarily illustrates a flow chart of a parking method provided in an embodiment of the present application. As shown in FIG5 , the parking method may include at least the following steps:

[0085] S501 , in response to a parking instruction input by a user, obtaining the current position of the vehicle through the vehicle controller.

[0086] Possibly, the parking instruction input by the user in the embodiment of the present application may be a voice instruction or a manually input instruction, and the present application does not impose any limitation thereto.

[0087] See Figure 6. The user can input a parking command by voice, such as "find a parking space" 61. When the vehicle controller receives the voice command through the vehicle display 60, it can obtain the vehicle's current location information through the Global Positioning System (GPS) in the vehicle program and send a prompt tone to the user, "Started, entering search mode" 62.

[0088] S502: Determine at least one parking space within a preset range based on the current position of the vehicle through the vehicle controller.

[0089] It can be understood that the vehicle controller in the embodiment of the present application can determine whether there is a parking space within a preset range near the vehicle based on the current position of the vehicle through the GPS positioning system. If so, the parking space that the vehicle can park in is displayed through the on-board display screen.

[0090] S503: Based on at least one parking space, a target parking space determination instruction input by a user is received through the vehicle controller.

[0091] Specifically, the user can determine the target parking space to be parked by clicking on the parking space displayed on the vehicle display screen or inputting a voice command.

[0092] See Figure 7. The vehicle display screen 70 shows the current position A of the vehicle and available parking spaces B, C, D, E, F, G, and H. The user can click on parking space H to select parking space H as the target parking space.

[0093] S504 , in response to the target parking space determination instruction, determining the target parking space through the vehicle controller.

[0094] Specifically, after receiving the target parking space confirmation instruction input by the user, the vehicle controller in the embodiment of the present application will determine the location information of the target parking space corresponding to the target parking space confirmation instruction through GPS, and send the location information corresponding to the target parking space to the domain controller.

[0095] It can be understood that the embodiment of the present application enables the user to complete the customization process of the target parking space while driving by having the user input the parking command, the vehicle controller search for nearby parking spaces, and the user select the target parking space from the nearby parking spaces. This not only improves the human-computer interaction function of the vehicle, but also avoids the problem of the user having to observe the surrounding environment and spend a lot of time looking for a parking space.

[0096] At step S505, if the vehicle's current position is within the parking area and the distance between the current position and the target parking space within the parking area is greater than a first preset distance, the domain controller obtains first image information captured by the first camera and second image information captured by the second camera. Specifically, step S505 is the same as step S301 and is not further described here.

[0097] S506 , the domain controller sends the first image information and a processing request instruction to the vehicle controller, where the processing request instruction is used to instruct the vehicle controller to process the first image information.

[0098] S507: The vehicle controller receives the first image information and the processing request instruction sent by the domain controller.

[0099] S508 : The vehicle controller extracts features from the first image information based on the processing request instruction to obtain first image feature information.

[0100] Specifically, S506-S508 are consistent with S302-S304 and will not be repeated here.

[0101] S509: The vehicle controller sends the first image feature information to the domain controller.

[0102] S510: The domain controller receives first image feature information sent by the vehicle controller.

[0103] S511: The domain controller generates a first parking route based on the first image feature information and the second image information.

[0104] Specifically, S509-S511 are consistent with S305-S307 and will not be repeated here.

[0105] Furthermore, in an embodiment of the present application, when the current position of the vehicle is within the target area and the distance between the current position and the target parking space within the target area is less than or equal to a first preset distance, second image information captured by the second camera is obtained, and a second parking route between the current position and the target parking space is planned based on the second image information by the domain controller.

[0106] The shooting range of the second camera is different from that of the first camera. For example, when the first camera is a front-view camera and the second camera is a surround-view camera, the shooting range of the second camera is smaller than that of the first camera.

[0107] It is understandable that when the vehicle drives near the target parking space and within the first preset distance of 6 meters, the vehicle generally moves at a slower speed and no longer considers vehicles traveling from a long distance. Therefore, only the surround-view camera can be used to capture close-range image information within a 360-degree range around the vehicle, so that the vehicle can determine the obstacle information, movable angle, movable distance and other related information in the four directions of front, rear, left and right based on the current position information.

[0108] Specifically, after the domain controller receives the second image information, it needs to first perform feature extraction processing on the second image through the SOC to extract feature information of obstacles around the vehicle and near the target parking space, marking line information of the target parking space, lane lines and other information.

[0109] Figure 8 shows a schematic diagram of the connection relationship between the domain controller and other devices in the vehicle. When the vehicle's current position is in a parking lot and the distance between the current position and the target parking space is greater than 6 meters, both the front-view camera and the surround-view camera can send compressed image information to the deserializer via the Low Voltage Differential Signaling (LVDS) interface. The deserializer sends the decompressed image information to the SOC via the Mobile Industry Processor Interface (MIPI). The SOC can use the transparent transmission function to send the image information captured by the front-view camera through the serializer to the vehicle controller for image feature extraction. The image information captured by the surround-view camera is sent to the vehicle controller for display to the user on the vehicle display, allowing the user to observe the surrounding scene status. After the vehicle controller extracts features from the long-range image information captured by the front-view camera, it obtains long-range image feature information and sends this information to the SOC in the domain controller. The domain controller can use the relevant image processing algorithms stored in the embedded multimedia controller (EMMC) and double data rate synchronous dynamic random access memory (DDR) to extract features from the short-range image information in the SOC to obtain short-range image feature information. The SOC can send the short-range image feature information and long-range image information to the MCU through the Ethernet module via the general purpose input / output port (GPIO) and serial peripheral interface (SPI). The Ethernet module transmits Ethernet packets to the SOC and MCU respectively through the reduced media independent interface (RMII). The MCU can also use the CAN-FD bus and fiber optic transceiver TX / RX to receive long-distance point cloud information sent by the millimeter-wave radar 5R-CAN through the CAN chip, and receive short-distance point cloud information sent by the ultrasonic radar system-transceiver driver (Ultrasonic Sensor System-Transceiver Driver, USS-TRCV) module and GPIO interface. The USS-TRCV module can be connected to the MCU through the display pixel interface (DPI) and SPI.After receiving the near-range and long-range point cloud information, the MCU performs feature extraction on the near-range and long-range point cloud information to obtain near-range and long-range point cloud feature information. It then plans a first parking route based on the near-range image feature information, long-range image feature information, near-range point cloud feature information, and long-range point cloud feature information. If the vehicle's current location is within the parking lot and the distance between the current location and the target parking space is less than 6 meters, the MCU may plan a second parking route based solely on the near-range image feature information, the near-range point cloud feature information, and the long-range point cloud feature information.

[0110] Therefore, when the vehicle enters the first preset distance range between the vehicle and the target parking space, the embodiment of the present application can only obtain image information of the surrounding environment of the target parking space captured by the second camera, that is, the surround-view camera, to determine information about nearby obstacles, and determine the second parking route based on the obstacle information, so as to achieve accurate planning of the parking route within a small range, avoid multiple parking failures caused by insufficient available parking area in the target parking space, thereby shortening parking time and improving user satisfaction.

[0111] In order to better understand the parking method of the vehicle provided in the embodiment of the present application, the embodiment of the present application also provides a parking device for the vehicle. Figure 9A is a structural schematic diagram of a parking device provided in an exemplary embodiment of the present application. The parking device can be applied to a domain controller. The vehicle includes: a first camera, a second camera, a vehicle controller, and a domain controller. The first camera, the second camera and the vehicle controller are respectively connected to the domain controller. Further, the parking method of any of the above embodiments of the present application can be executed. As shown in Figure 9, the parking device of the vehicle may include:

[0112] A first acquisition module 91 is configured to acquire, when the current position of the vehicle is within the parking area and the distance between the current position and a target parking space within the parking area is greater than a first preset distance, first image information captured by the first camera and second image information captured by the second camera; the first camera and the second camera have different ranges of distance.

[0113] A first sending module 92 is configured to send the first image information and a processing request instruction to the vehicle controller, wherein the processing request instruction is configured to instruct the vehicle controller to process the first image information to obtain first image feature information;

[0114] A first receiving module 93 is configured to receive first image feature information sent by the vehicle controller;

[0115] The route generating module 94 is configured to generate a first parking route based on the first image feature information and the second image information.

[0116] Thus, in an embodiment of the present application, when the current position of the vehicle is within the parking area and the distance between the current position and the target parking space within the parking area is greater than a first preset distance, the first image information captured by the first camera and the second image information captured by the second camera are obtained, and the first image information and a processing request instruction are sent to the vehicle controller, where the processing request instruction is used to instruct the vehicle controller to process the first image information to obtain first image feature information; receive the first image feature information sent by the vehicle controller; and generate a first parking route based on the first image feature information and the second image information. Thus, in an embodiment of the present application, the domain controller can process image information within different ranges captured during long-distance parking, thereby resolving the problem of the domain controller being unable to simultaneously process multiple types of image information when using a low-computing power SOC. The parking route in the long-distance parking scenario is planned based on the received first image feature information and second image information, thereby resolving the potential collision problem during long-distance parking, and thereby improving the safety performance of the vehicle while reducing vehicle costs.

[0117] In some embodiments, the route generation module 94 includes:

[0118] an extraction unit, configured to extract second image feature information from the second image information;

[0119] a first splicing unit configured to splice the first image feature information and the second image feature information based on a distance range corresponding to the first image feature information and a distance range corresponding to the second image feature information to obtain image feature information corresponding to the parking area;

[0120] The first parking route generating unit is configured to generate a first parking route based on image feature information corresponding to the parking area.

[0121] In some embodiments, the vehicle further comprises: a radar system;

[0122] Before the route generation module 94, the apparatus further includes:

[0123] Point cloud information receiving module, used to receive point cloud information sent by the radar system;

[0124] The first determination module is used to extract features from the point cloud information to obtain point cloud feature information;

[0125] The route generation module 94 is specifically configured to:

[0126] Based on the image feature information and point cloud feature information corresponding to the parking area, a first parking route is generated. In some embodiments,

[0127] The radar system includes: a first radar and a second radar; the first radar and the second radar have different detection ranges; and a first determination module includes:

[0128] a second stitching unit configured to stitch the first point cloud feature information and the second point cloud feature information based on a distance range corresponding to the first point cloud feature information and a distance range corresponding to the second point cloud feature information to obtain point cloud feature information corresponding to the parking area; wherein the first point cloud feature information is obtained by performing feature extraction on the first point cloud information collected by the first radar, and the second point cloud feature information is obtained by performing feature extraction on the second point cloud information collected by the second radar;

[0129] The second parking route generating unit is configured to generate a first parking route based on the image feature information corresponding to the parking area and the point cloud feature information corresponding to the parking area.

[0130] In some embodiments, the parking area is an area where the target parking space pre-configured by the user is located. In some embodiments, before the first acquisition module 91, the device further includes:

[0131] a first acquiring unit, configured to acquire a current position of the vehicle in response to a parking instruction input by a user;

[0132] a parking space determining unit, configured to determine at least one parking space within a preset range based on the current position of the vehicle;

[0133] a target parking space input unit, configured to receive a target parking space determination instruction input by a user based on at least one parking space;

[0134] The target parking space determining unit is configured to determine the target parking space in response to the target parking space determining instruction.

[0135] To better understand the vehicle parking method provided in the embodiments of the present application, the embodiments of the present application also provide a vehicle parking device. Figure 9B is a schematic structural diagram of a parking device provided in an exemplary embodiment of the present application. The parking device can be applied to a vehicle controller. Furthermore, any of the parking methods described above can be executed. As shown in Figure 9B, the vehicle parking device may include:

[0136] The second receiving module 95 is configured to receive the first image information and the processing request instruction sent by the domain controller; wherein the first image information is captured by the first camera;

[0137] A feature extraction module 96 is configured to extract features from the first image information based on the processing request instruction to obtain first image feature information;

[0138] The second sending module 97 is used to send the first image feature information to the domain controller, so that the domain controller generates a first parking route based on the first image feature information and the second image information; wherein the second image information is obtained by taking pictures by the second camera, and the distance ranges taken by the first camera and the second camera are different.

[0139] In addition, the device provided in the embodiments of the present application can specifically be a chip, component or module, and the chip may include a connected processor and memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute a parking method provided in the above embodiment.

[0140] To better understand the parking method provided in the embodiments of the present application, the embodiments of the present application also provide a vehicle. FIG10 exemplarily illustrates a schematic structural diagram of a vehicle provided in the embodiments of the present application. As shown in FIG10 , vehicle 10 may include: at least one processor 11, at least one network interface 14, a user interface 13, a memory 15, and at least one communication bus 12.

[0141] The communication bus 12 is used to enable communication between these components. The user interface 13 may include a display and a camera. Optionally, the user interface 13 may also include a standard wired interface or a wireless interface. The memory 15 stores executable program code, and the processor 11 is used to call and execute the executable program code to perform a parking method.

[0142] This embodiment further provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer executes the above-mentioned related method steps to implement a parking method provided in the above embodiment.

[0143] This embodiment further provides a computer program product. When the computer program product is run on a computer, it enables the computer to execute the above-mentioned related steps to implement a parking method provided in the above embodiment.

[0144] Among them, this embodiment provides a computer-readable storage medium, a computer program product or a chip for executing the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method provided above and will not be repeated here.

Claims

1. A method for parking a vehicle, characterized in that: A domain controller applied to the vehicle, the vehicle comprising: a first camera, a second camera, a vehicle controller and the domain controller, the first camera, the second camera and the vehicle controller are respectively connected to the domain controller, the method comprising: When the current position of the vehicle is within the parking area, and the distance between the current position and the target parking space within the parking area is greater than a first preset distance, obtaining first image information captured by the first camera and second image information captured by the second camera, wherein the distance ranges captured by the first camera and the second camera are different; Sending the first image information and a processing request instruction to the vehicle controller, wherein the processing request instruction is used to instruct the vehicle controller to process the first image information to obtain first image feature information; Receiving first image feature information sent by the vehicle controller; A first parking route is generated based on the first image feature information and the second image information.

2. The method according to claim 1, characterized in that The generating a first parking route based on the first image feature information and the second image information includes: extracting second image feature information from the second image information; Based on the distance range corresponding to the first image feature information and the distance range corresponding to the second image feature information, the first image feature information and the second image feature information are spliced ​​to obtain image feature information corresponding to the parking area; The first parking route is generated based on the image feature information corresponding to the parking area.

3. The method according to claim 2, characterized in that The vehicle further includes: a radar system; before obtaining a first parking route based on the first image feature information and the second image information, the method further includes: Receiving point cloud information sent by the radar system; Performing feature extraction on the point cloud information to obtain point cloud feature information; The generating the first parking route based on the image feature information corresponding to the parking area includes: The first parking route is generated based on the image feature information corresponding to the parking area and the point cloud feature information.

4. The method according to claim 3, characterized in that The radar system includes: a first radar and a second radar; the first radar and the second radar have different detection ranges; and the generating of the first parking route based on the image feature information corresponding to the parking area and the point cloud feature information includes: Based on the distance range corresponding to the first point cloud feature information and the distance range corresponding to the second point cloud feature information, the first point cloud feature information and the second point cloud feature information are spliced ​​to obtain point cloud feature information corresponding to the parking area; wherein the first point cloud feature information is obtained by performing feature extraction on the first point cloud information collected by the first radar, and the second point cloud feature information is obtained by performing feature extraction on the second point cloud information collected by the second radar; The first parking route is generated based on the image feature information corresponding to the parking area and the point cloud feature information corresponding to the parking area.

5. The method according to claim 1, characterized in that The parking area is the area where the target parking space is located, which is pre-configured by the user.

6. The method according to claim 1, characterized in that When the current position of the vehicle is within the parking area and the distance between the current position and the target parking space within the parking area is greater than a first preset distance, before acquiring the first image information captured by the first camera and the second image information captured by the second camera, the method further includes: In response to a parking instruction input by a user, obtaining a current position of the vehicle; Based on the current position of the vehicle, determining at least one parking space within a preset range; Based on the at least one parking space, receiving a target parking space determination instruction input by the user; In response to the target parking space determination instruction, the target parking space is determined.

7. A method for parking a vehicle, characterized in that: Applied to a vehicle controller, the vehicle includes: a first camera, a second camera, the vehicle controller, and a domain controller, the first camera, the second camera and the vehicle controller are respectively connected to the domain controller, and the method includes: Receiving first image information and a processing request instruction sent by the domain controller; wherein the first image information is obtained by taking pictures by the first camera; Extracting features from the first image information based on the processing request instruction to obtain first image feature information; The first image feature information is sent to the domain controller so that the domain controller generates a first parking route based on the first image feature information and the second image information; wherein the second image information is obtained by taking pictures by the second camera, and the distance ranges taken by the first camera and the second camera are different.

8. A parking device for a vehicle, characterized in that: Applied to a domain controller, the vehicle includes: a first camera, a second camera, a vehicle controller, and the domain controller, the first camera, the second camera and the vehicle controller are respectively connected to the domain controller, and the device includes: A first acquisition module is used to acquire first image information captured by the first camera and second image information captured by the second camera when the current position of the vehicle is within a parking area and the distance between the current position and a target parking space within the parking area is greater than a first preset distance; the distance ranges captured by the first camera and the second camera are different; a first sending module, configured to send the first image information and a processing request instruction to the vehicle controller, wherein the processing request instruction is used to instruct the vehicle controller to process the first image information to obtain first image feature information; A first receiving module, used for receiving first image feature information sent by the vehicle controller; A route generating module is used to generate a first parking route based on the first image feature information and the second image information.

9. A parking device for a vehicle, characterized in that: Applied to a vehicle controller, the vehicle includes: a first camera, a second camera, the vehicle controller, and a domain controller, the first camera, the second camera and the vehicle controller are respectively connected to the domain controller, and the device includes: A second receiving module is used to receive the first image information and the processing request instruction sent by the domain controller; wherein the first image information is obtained by taking the first camera; a feature extraction module, configured to extract features from the first image information based on the processing request instruction to obtain first image feature information; The second sending module is used to send the first image feature information to the domain controller, so that the domain controller generates a first parking route based on the first image feature information and the second image information; wherein the second image information is obtained by taking pictures by the second camera, and the distance ranges taken by the first camera and the second camera are different.

10. A vehicle, characterized in that: The vehicle comprises: a first camera, a second camera, a vehicle controller, and a domain controller; the first camera, the second camera and the vehicle controller are respectively connected to the domain controller; The first camera is used to capture first image information, and the second camera is used to capture second image information; the first camera and the second camera have different shooting distance ranges; The domain controller is used to execute the parking method according to any one of claims 1 to 6; The vehicle controller is used to execute the parking method as claimed in claim 7.

Images