Parking methods, parking systems, and electronic devices with autonomous exploration modes

Abstract

This disclosure provides a parking method with an autonomous exploration mode, comprising: activating an onboard parking module; a terminal parking module generating a parking exploration command in response to user control and sending it to the onboard parking module; the onboard parking module executing a parking control process in autonomous exploration mode based on the parking exploration command; or, the onboard parking module sending historical trajectory information of the vehicle associated with its current location and including at least one historical parking target location to the terminal parking module; and the onboard parking module executing a parking control process based on the historical parking target location selection command received by the terminal parking module; the parking exploration command includes an exploration range command, an exploration distance command, and a parking location exploration command; the onboard parking module exploring parking locations based on the exploration range and exploration distance limited by the exploration range and exploration distance commands. This disclosure also provides a parking system, an electronic device, a readable storage medium, and a computer program product.

Description

Technical Field

[0001] This disclosure relates to the field of automated parking technology, and more particularly to parking methods, parking systems and electronic devices with autonomous exploration modes. Background Technology

[0002] Parking is generally a necessary stage in any driving activity, but it is also one of the more complex and challenging. In many public places, designated parking areas often differ from people's final destinations. For example, park parking lots are usually designed in open areas tens to hundreds of meters away from the park gate for reasons such as traffic convenience and fire prevention, which creates an additional walking burden for drivers. Therefore, automated valet parking has always been a research hotspot in the field of autonomous driving technology.

[0003] Currently, several parking methods or technologies applied to the field of autonomous driving have been developed or researched. These can be broadly categorized into four types: first, parking assistance modes where the vehicle parks itself near the target space; second, remote parking modes where the owner controls the vehicle remotely via a mobile device or remote key; third, memory-based autonomous parking modes where the driver manually trains and memorizes the vehicle's route; and fourth, valet-based autonomous parking modes that interact with the parking lot and perform global planning based on a high-definition map of the target parking area.

[0004] However, existing parking methods either require users to be in the car or assist them, preventing them from getting out of the car in their ideal position and placing an extra burden on them, or require users to manually train them in advance to memorize the parking information, thus having significant limitations in unfamiliar environments. Alternatively, parking lots may be equipped with intelligent parking systems that push high-definition maps, but the infrastructure costs for this type of parking method are very high, and it is currently only used in public parking lots of a few large shopping malls, and it is difficult to promote.

[0005] To understand the current state of technological development, this disclosure has searched, compared, and analyzed existing patents and papers, and selected the following existing technical solutions:

[0006] Technical Solution 1: Patent document CN113753030A discloses "A Memory Parking System, Method, Terminal, and Storage Medium." This system generates a local map and reference route from the drop-off point to the parking space by memorizing the parking path pre-trained by the user, enabling the vehicle to autonomously drive to and park. The user activates the memory mode at the starting point of the parking route. The system memorizes the surrounding environment data, location data, and path trajectory of the vehicle from startup until the vehicle stops in the target parking space. The memory system then constructs a local map from the memorized data along the route. When the memory system recognizes the vicinity of the starting position again, the vehicle can mimic the previously learned parking route to complete autonomous parking. This method successfully achieves separation of the user and the vehicle, allowing the user to get off at an ideal drop-off point without having to follow the car to the parking area, greatly liberating the user. However, this method has certain limitations because the system's map requires the user to actively learn it through pre-driving. Therefore, this method can only be effectively applied to parking lots commonly used by users, such as those in companies and residential communities. However, the system cannot be effectively used when users arrive in unfamiliar areas they have not yet learned to drive, such as when traveling for business, visiting relatives or friends, or when users choose a new drop-off point. Furthermore, this method has high requirements for environmental conditions, and in some cases may even require users to repeatedly drive and learn, placing an additional burden on them.

[0007] Technical Solution 2: Patent document CN113085900A discloses "A Method for Summoning a Vehicle to a User's Location." This system allows the vehicle to receive commands from the user and autonomously drive to the user's location based on map planning. The user issues a summoning command to the vehicle via a mobile app. The vehicle plans a route on the map and drives towards the user. During the journey, the user can monitor the vehicle's surroundings in real time and continuously control the vehicle. When the vehicle determines it has reached the user's vicinity, it attempts to identify the user. If identification is successful, the vehicle drives to the user and stops. This method also achieves separation of the user and vehicle and eliminates the need for the user to manually train the vehicle's route. However, this method also has significant limitations. The summoning route planning relies on the onboard map, and route errors are easily caused when the map is inaccurate. Furthermore, this method cannot be used for uncertain destinations, thus greatly limiting its usability.

[0008] Technical Solution 3: Patent document CN113313629B discloses an "Automatic Intersection Recognition Method, System, and Model Saving Method / System." This system collects 3D point cloud data around an unmanned vehicle to construct a 2D grid map, estimates the intersection center based on the grid map, and then segments the road to obtain and analyze each fork in the road. This method has significant limitations because it uses straight lines to simulate the road edges. While this works well for simple, square intersections, it easily leads to serious misjudgments for complex, winding intersections within an area. Therefore, further optimization is needed. Summary of the Invention

[0009] To address at least one of the aforementioned technical problems, this disclosure provides a parking method, parking system, and electronic device with an autonomous exploration mode.

[0010] According to one aspect of this disclosure, a parking method with an autonomous exploration mode is provided, comprising:

[0011] S1100: The vehicle parking module receives a start control signal generated and sent by the terminal parking module, and starts the vehicle parking module based on the start control signal;

[0012] S1400: The terminal parking module generates a parking exploration command in response to the user's operation (touch operation or voice control) and sends it to the vehicle parking module. The vehicle parking module executes the parking control process of autonomous exploration mode based on the parking exploration command; or, the vehicle parking module sends the vehicle's historical trajectory information, which is associated with the current location of the vehicle and includes at least one historical parking target location, to the terminal parking module. Based on the historical parking target location selection command received by the terminal parking module, the vehicle parking module executes the parking control process.

[0013] The parking exploration command includes an exploration range command, an exploration distance command, and a parking location exploration command. The vehicle parking module explores parking locations based on the exploration range and exploration distance limited by the exploration range command and the exploration distance command.

[0014] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, the parking exploration command is generated through the following steps:

[0015] A four-quadrant interface is generated, wherein each quadrant region of the four-quadrant interface can be selected in response to user operation to generate an exploration range instruction, wherein the exploration range instruction is an exploration range instruction for one or more quadrant regions.

[0016] An exploration distance adjustment interface is generated, which can respond to user operations to adjust the exploration distance, thereby generating an exploration distance command;

[0017] A parking location adjustment interface is generated, which can respond to user operations to adjust the parking location, thereby generating a parking location exploration command.

[0018] According to at least one embodiment of the parking method with an autonomous exploration mode, between step S1100 and step S1400, the method further includes:

[0019] S1200: The terminal parking module outputs a first voice signal for interaction;

[0020] S1300: The terminal parking module determines whether it has received a first characteristic voice signal in response to the first voice signal.

[0021] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, step S1200 further includes:

[0022] The vehicle parking module obtains the current location of the vehicle and determines whether there is historical trajectory information of the vehicle associated with the current location, including at least one historical parking target location.

[0023] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, the on-board parking module obtains the current position of the vehicle and determines whether there is historical trajectory information of the vehicle associated with the current position and including at least one historical parking target position. If the determination result is: there is no historical trajectory information of the vehicle associated with the current position and including at least one historical parking target position, then step S1400 includes:

[0024] S1402a: The terminal parking module outputs a control interface, and based on the control interface, the terminal parking module receives user commands.

[0025] S1404a: The terminal parking module generates a parking exploration command, including an exploration range command, an exploration distance command, and a parking location exploration command, in response to the received user's operation, for the user to confirm.

[0026] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, step S1400 further includes:

[0027] S1406a: The terminal parking module receives the confirmation command and sends the parking exploration command to the vehicle parking module to execute the parking control process of the autonomous exploration mode.

[0028] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, step S1400 further includes:

[0029] S1406b: If the terminal parking module does not receive a confirmation command or receives a denial command, it will output the control interface again to receive the user's control to execute step S1404a again.

[0030] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, the on-board parking module obtains the current position of the vehicle and determines whether there is historical trajectory information of the vehicle associated with the current position and including at least one historical parking target position. If the determination result is: there is historical trajectory information of the vehicle associated with the current position and including at least one historical parking target position, then step S1400 includes:

[0031] S1402c: The vehicle parking module outputs the vehicle's historical trajectory information, which is associated with the vehicle's current location and includes at least one historical parking target location, to the terminal parking module for selection.

[0032] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, step S1400 further includes:

[0033] S1404c: When the terminal parking module receives a historical parking target location selection instruction, the vehicle parking module executes a parking control process based on the historical parking target location selection instruction.

[0034] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, step S1400 further includes:

[0035] S1404d: If the terminal parking module does not receive a historical parking target location selection instruction, the terminal parking module outputs a control interface, which is used to receive user commands.

[0036] S1406d: The terminal parking module generates a parking exploration command, including an exploration range command, an exploration distance command, and a parking location exploration command, in response to the user's operation received from the control interface, for the user to confirm.

[0037] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, step S1400 further includes:

[0038] S1408d: The terminal parking module receives the confirmation command and sends the parking exploration command to the vehicle parking module to execute the parking control process of autonomous exploration mode.

[0039] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, step S1400 further includes:

[0040] S1408e: If the terminal parking module does not receive a confirmation command or receives a denial command, it will output the control interface again to execute step S1406d again.

[0041] A parking method with an autonomous exploration mode according to at least one embodiment of the present disclosure executes a parking control process in the autonomous exploration mode based on the parking exploration command, including:

[0042] The on-board parking module determines whether the vehicle meets the starting conditions. Once the starting conditions are met, the on-board parking module generates a control signal to control the vehicle to start.

[0043] The vehicle parking module obtains the feasible domain of the current road based on the exploration range and exploration distance limits set by the exploration range command and the exploration distance command, performs local path planning in real time according to road conditions, and generates control signals to control the vehicle to drive along the feasible domain.

[0044] Once the vehicle parking module detects a parking location, it executes the parking process based on that location.

[0045] A parking method with an autonomous exploration mode according to at least one embodiment of the present disclosure further includes:

[0046] Once the vehicle parking module obtains the branch road information, it performs local path planning based on the branch road information to update the feasible region.

[0047] A parking method with an autonomous exploration mode according to at least one embodiment of the present disclosure, which performs local path planning based on branch information to update the feasible region, includes:

[0048] Construct a two-dimensional grid map for the intersection;

[0049] Based on each edge line of the intersection, all rasters covering that edge line are assigned an initial value. Then, rasters adjacent to all assigned rasters within the feasible region of the 2D raster map are assigned values. Rasteres sharing an edge are assigned a first value, and rasters sharing a corner are assigned a second value. If a raster shares an edge or a corner with multiple rasters, the value with the smallest result after adding the first value to the edge-sharing raster or the second value to the corner-sharing raster is selected for raster assignment. This process continues until all rasters within the feasible region are assigned values, thus completing the raster assignment based on the edge lines of each road at the intersection.

[0050] The raster assignment results based on each edge line are overlaid with raster map values ​​to obtain an initial overlaid assignment map.

[0051] The center point of the grid with the smallest superposition value in the initial superposition assignment map is taken as the center point of the intersection. If there are multiple grids with the smallest superposition values, their geometric center is taken as the center point of the intersection.

[0052] Based on the center point of the intersection, obtain the intersection's traffic area and each sub-path;

[0053] The feasible region is updated based on the intersection access region and each sub-path.

[0054] A parking method with an autonomous exploration mode according to at least one embodiment of the present disclosure executes a parking control process based on historical parking target location selection instructions, including:

[0055] S1441: Compare the current position of the vehicle with at least one historical trajectory of the vehicle that includes the selected historical parking target position, and obtain the optimal historical trajectory based on the number of times the vehicle's historical trajectory has been driven and the distance between the current position of the vehicle and the historical trajectory of the vehicle.

[0056] S1442: Determine whether the minimum distance between the optimal historical trajectory and the current position of the vehicle exceeds the preset distance. If it exceeds, proceed to step S1443; if it does not exceed, proceed to step S1444.

[0057] S1443: The on-board parking module controls the vehicle to move forward along the current feasible domain and continuously marks the historical trajectory point closest to the vehicle's current position to find the optimal historical trajectory. Whenever the vehicle encounters a fork in the road, the on-board parking module controls the vehicle to turn into the fork in the road that the line connecting the marked point points points in the direction of the fork, until the minimum distance between the vehicle's current position and the optimal historical trajectory is less than or equal to the preset distance, and then proceeds to step S1444.

[0058] S1444: The on-board parking module controls the vehicle to reposition itself in order to move the vehicle to the optimal historical trajectory and align it with the historical pose.

[0059] S1445: Plan the optimal global route from the optimal historical trajectory, and control the vehicle to travel to the target location based on the optimal global route.

[0060] According to at least one embodiment of the parking method with an autonomous exploration mode of the present disclosure, step S1400 further includes the following parking processing procedure:

[0061] S1431: The vehicle parking module locates the parking spot;

[0062] S1432: If the parking location is a roadside parking area, proceed to step S1433; if the parking location is a parking lot, proceed to step S1434; if the parking location is an open space / plaza, proceed to step S1435.

[0063] S1433: Identify roadside parking areas, and the on-board parking module generates control signals to control the vehicle to enter the roadside parking area;

[0064] S1434: Identify the parking lot, identify the parking area, identify the parking space number, and generate a control signal to control the vehicle to drive into the parking space.

[0065] S1435: Identify open spaces / plazas, and the on-board parking module generates control signals to control the vehicle to enter the open spaces / plazas;

[0066] S1436: When steps S1433 / S1434 / S1435 are executed successfully, the vehicle parking module generates a control signal to control the vehicle to stop and generates parking success information to send to the user terminal device; when steps S1433 / S1434 / S1435 fail to execute, the vehicle parking module generates parking failure information and sends it to the mobile device.

[0067] According to another aspect of this disclosure, a parking system with an autonomous exploration mode is provided, comprising:

[0068] A terminal parking module, which generates and sends out a start control signal;

[0069] The vehicle parking module receives the start control signal and starts;

[0070] The terminal parking module includes:

[0071] A parking exploration command generation module, which generates exploration range commands, exploration distance commands, and parking location exploration commands in response to user operations;

[0072] The vehicle parking module includes:

[0073] The historical trajectory output module is used to output the historical trajectory information of the vehicle that is associated with the current position of the vehicle and includes at least one historical parking target position.

[0074] The parking processing module executes a parking control process based on the parking exploration command generated by the terminal parking module or the selected historical parking target location and its corresponding historical trajectory information of the vehicle.

[0075] The parking processing module of the vehicle parking module explores parking locations based on the exploration range and exploration distance limits set by the exploration range command and the exploration distance command.

[0076] According to at least one embodiment of the parking system with an autonomous exploration mode of the present disclosure, the terminal parking module further includes:

[0077] A voice signal output module, wherein the voice signal output module outputs a first voice signal for interaction;

[0078] A voice signal receiving module, wherein the voice signal receiving module is used to receive the user's voice signal;

[0079] The first characteristic speech signal acquisition module determines whether a first characteristic speech signal in response to the first speech signal has been received.

[0080] According to at least one embodiment of the parking system with an autonomous exploration mode, the terminal parking module further includes: a touch command receiving module, which is at least used to receive historical parking target location selection commands.

[0081] According to another aspect of this disclosure, an electronic device is provided, comprising: a memory storing execution instructions; and a processor executing the execution instructions stored in the memory, causing the processor to perform a parking method according to any embodiment of this disclosure.

[0082] According to another aspect of this disclosure, a readable storage medium is provided, wherein executable instructions are stored therein, which, when executed by a processor, are used to implement a parking method according to any embodiment of this disclosure.

[0083] According to another aspect of this disclosure, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement a parking method according to any embodiment of this disclosure. Attached Figure Description

[0084] The accompanying drawings illustrate exemplary embodiments of the present disclosure and, together with the description thereof, serve to explain the principles of the present disclosure. These drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification.

[0085] Figure 1 This is a flowchart illustrating a parking method with an autonomous exploration mode, according to one embodiment of this disclosure.

[0086] Figure 2 A schematic diagram of a four-quadrant interface according to one embodiment of the present disclosure is shown.

[0087] Figure 3 A schematic diagram of an exploration distance adjustment interface according to one embodiment of this disclosure is shown.

[0088] Figure 4 A schematic diagram of a parking location exploration interface according to one embodiment of this disclosure is shown.

[0089] Figure 5 This is a publicly available feedback chart.

[0090] Figure 6 This is a flowchart illustrating a parking method with an autonomous exploration mode according to one embodiment of the present disclosure.

[0091] Figure 7 This is a flowchart illustrating another embodiment of the parking method with an autonomous exploration mode.

[0092] Figure 8 This is a flowchart illustrating another embodiment of the parking method with an autonomous exploration mode.

[0093] Figure 9 This is a flowchart illustrating a parking method with an autonomous exploration mode, which is yet another embodiment of this disclosure.

[0094] Figure 10 This is a schematic diagram of a parking control process with an autonomous exploration mode, according to one embodiment of this disclosure.

[0095] Figure 11 This is a route map of one embodiment of the present disclosure, showing the vehicle traveling along the right edge of the road within its field of vision when not encountering an intersection.

[0096] Figure 12 This is a schematic diagram illustrating the obstacle avoidance and detour of the vehicle when encountering an obstacle, according to one embodiment of this disclosure.

[0097] Figure 13 This is a schematic diagram of obtaining a new curved path that satisfies the minimum turning radius of the vehicle during a cornering process according to one embodiment of this disclosure.

[0098] Figure 14 A schematic diagram of a fork in the road is shown, representing one embodiment of this disclosure.

[0099] Figure 15 A schematic diagram illustrating the isolation zone determination of one embodiment of this disclosure is shown.

[0100] Figure 16 A schematic flowchart of a fork-in-the-road processing method according to one embodiment of the present disclosure is shown.

[0101] Figure 17 A two-dimensional grid map constructed for a fork in the road is shown as one embodiment of this disclosure.

[0102] Figure 18 and Figure 19An exemplary diagram illustrates grid assignment based on an edge line (the road edge line to the right front of the vehicle).

[0103] Figures 20 to 22 The grid assignment results for the right-side road edge line, left-side road edge line, and left-front road edge line of this vehicle are shown as an example.

[0104] Figure 23 An exemplary diagram of the fork intersection assignment of this disclosure is shown.

[0105] Figure 24 and Figure 25 This is a branching analysis diagram of one embodiment of the present disclosure, which exemplarily illustrates the traffic area and sub-paths.

[0106] Figure 26 This is a partial planning path diagram of one embodiment of the present disclosure.

[0107] Figure 27 This illustration shows a left turn at a fork in the road according to one embodiment of the present disclosure.

[0108] Figure 28 An example image of a parking lot sign is shown.

[0109] Figure 29 An example diagram of a branch road section is shown.

[0110] Figure 30 An example diagram of the processing of field of view occlusion is shown.

[0111] Figure 31 The image shows a segmentation map of the plate pixels in the field of view.

[0112] Figure 32 A diagram illustrating the field of view analysis is shown.

[0113] Figure 33 A diagram illustrating the driving process is shown.

[0114] Figure 34 This is a schematic flowchart of a parking processing method according to one embodiment of the present disclosure.

[0115] Figure 35 This is an example image of a parking lot sign.

[0116] Figure 36 This is an example diagram of an open parking lot.

[0117] Figure 37 This is an example image of parking area signs (area numbers) in a parking lot.

[0118] Figure 38 This is an example diagram for finding the optimal historical trajectory.

[0119] Figure 39 It is a repositioning trajectory point matching map.

[0120] Figure 40 and Figure 41 The schematic block diagrams of the terminal parking module and the vehicle-mounted parking module are shown respectively. Detailed Implementation

[0121] The present disclosure will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the disclosure. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present disclosure are shown in the accompanying drawings.

[0122] It should be noted that, where there is no conflict, the embodiments and features described in this disclosure can be combined with each other. The technical solutions of this disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0123] Unless otherwise stated, the exemplary implementations / embodiments shown are to be understood as providing exemplary features of various details that provide ways in which the technical concepts of this disclosure can be implemented in practice. Therefore, unless otherwise stated, the features of various implementations / embodiments may be additionally combined, separated, interchanged and / or rearranged without departing from the technical concepts of this disclosure.

[0124] The use of crosshairs and / or shading in the accompanying drawings is generally used to clarify the boundaries between adjacent components. Thus, unless otherwise stated, the presence or absence of crosshairs or shading does not convey or indicate any preference or requirement for the specific material, material properties, dimensions, proportions, commonalities between the illustrated components, or any other characteristics, properties, etc., of the components. Furthermore, in the accompanying drawings, the dimensions and relative dimensions of components may be exaggerated for clarity and / or descriptive purposes. When exemplary embodiments can be implemented differently, a specific process sequence may be performed in a different order than that described. For example, two consecutively described processes may be performed substantially simultaneously or in the reverse order of their description. Furthermore, the same reference numerals denote the same components.

[0125] When a component is referred to as being "on" or "above" another component, "connected to," or "joined to" another component, the component may be directly on, directly connected to, or directly joined to the other component, or there may be intermediate components. However, when a component is referred to as being "directly on" another component, "directly connected to," or "directly joined to" another component, there are no intermediate components. Therefore, the term "connection" can refer to a physical connection, an electrical connection, etc., and may or may not have intermediate components.

[0126] The terminology used herein is for the purpose of describing particular embodiments and is not intended to be limiting. As used herein, unless the context clearly indicates otherwise, the singular forms “a” and “the” are intended to include the plural forms as well. Furthermore, when the terms “comprising” and / or “including” and variations thereof are used in this specification, it indicates the presence of the stated features, integrals, steps, operations, parts, components, and / or groups thereof, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, parts, components, and / or groups thereof. It should also be noted that, as used herein, the terms “substantially,” “about,” and other similar terms are used as approximate terms rather than as terms of degree, thus explaining the inherent biases in measurements, calculated values, and / or provided values ​​that would be recognized by one of ordinary skill in the art.

[0127] The following text combines Figures 1 to 41 The parking method and parking system disclosed herein are described in detail.

[0128] Figure 1 This is a flowchart illustrating a parking method with an autonomous exploration mode, according to one embodiment of this disclosure.

[0129] refer to Figure 1 The parking method S1000 with autonomous exploration mode disclosed herein includes the following steps:

[0130] S1100: The vehicle parking module receives the start control signal generated and sent by the terminal parking module, and starts the vehicle parking module based on the start control signal;

[0131] S1400: The terminal parking module generates a parking exploration command in response to user operations (including but not limited to touch operation and voice control) and sends it to the vehicle parking module. The vehicle parking module executes the parking control process of autonomous exploration mode based on the parking exploration command; or, the vehicle parking module sends the vehicle's historical trajectory information associated with the current location of the vehicle and including at least one historical parking target location to the terminal parking module. Based on the historical parking target location selection command received by the terminal parking module, the vehicle parking module executes the parking control process.

[0132] The parking exploration commands include exploration range commands, exploration distance commands, and parking location exploration commands. The on-board parking module explores parking locations based on the exploration range and exploration distance limits set by the exploration range and exploration distance commands.

[0133] The start control signal described in this disclosure is preferably a start control signal sent by the user through a terminal device such as a mobile phone, and the vehicle parking module starts based on the start control signal.

[0134] The terminal device includes the terminal parking module (parking APP, mini program, etc.) described in this disclosure. The start control signal can be generated based on the terminal parking module and sent to the vehicle parking module via the terminal device.

[0135] In some embodiments of this disclosure, a start control signal can be generated and sent at the same time as the terminal parking module is started. In other embodiments of this disclosure, a start control signal is generated and sent based on the user's operation on the terminal parking module (e.g., operation controls displayed on a touch terminal device).

[0136] The parking method S1000 with autonomous exploration mode disclosed herein can realize the generation of parking exploration instructions by the terminal parking module, and the execution of parking control process by the vehicle parking module based on the parking exploration instructions.

[0137] According to a preferred embodiment of this disclosure, the parking exploration command of this disclosure is generated through the following steps:

[0138] A four-quadrant interface is generated, wherein each quadrant region of the four-quadrant interface can be selected in response to user operation to generate an exploration range instruction, wherein the exploration range instruction is an exploration range instruction for one or more quadrant regions.

[0139] An exploration distance adjustment interface is generated, which can respond to user operations to adjust the exploration distance, thereby generating an exploration distance command;

[0140] A parking location adjustment interface is generated, which can respond to user operations to adjust the parking location, thereby generating a parking location exploration command.

[0141] In some embodiments of this disclosure, if the four-quadrant interface does not receive user input, the generated exploration range instruction is an exploration range instruction covering the entire four-quadrant region, i.e., the default exploration range instruction. In this disclosure, the default exploration range instruction can be adjusted.

[0142] Figure 2 A schematic diagram of a four-quadrant interface according to one embodiment of the present disclosure is shown. Figure 2 In the diagram, the positive Y-axis corresponds to the direction the vehicle is facing. Users can select one or more of the four quadrants as their intended exploration area. If the user chooses to skip, the exploration area instruction will be the default value.

[0143] According to a preferred embodiment of this disclosure, the distance adjustment interface is generated based on a four-quadrant interface.

[0144] Figure 3 A schematic diagram of an exploration distance adjustment interface according to one embodiment of this disclosure is shown. Figure 3 In this context, preferably, the exploration distance is represented by the radius. Figure 3 The maximum exploration distance (1000 meters, adjustable) and the exploration distance adjusted in this instance (500 meters) are shown. In some embodiments of this disclosure, the exploration distance can be adjusted by dragging the arrow.

[0145] In some embodiments of this disclosure, the parking location may be a roadside parking area, a plaza parking area, an open space parking area, a parking lot, etc.

[0146] Figure 4 A schematic diagram of a parking location exploration interface according to one embodiment of this disclosure is shown. Figure 4 The document illustrates two parking location adjustment controls: a parking area search control and a parking lot search control. In some embodiments of this disclosure, only one of the different parking location adjustment controls can be determined, while in other embodiments, two or more parking location adjustment controls can be determined simultaneously.

[0147] like Figure 4 As shown, users can choose to explore an open area where parking is possible (including roadsides and plazas) or a parking lot. They can also choose both; in this case, the parking system will attempt to find an open parking lot if no parking lot is found, and then park in the parking lot when one is found. After selecting to explore a parking lot, users can further specify the parking area and space number to ensure precise vehicle location.

[0148] After the vehicle parking module of this disclosure generates a parking exploration command and the user confirms it, the vehicle parking module will send the generated parking exploration command back to the user and request the user to confirm or modify it. A specific feedback diagram is shown below. Figure 5 As shown.

[0149] Example of instruction feedback:

[0150] Users select the first and second quadrants as the exploration area (direction), and 500 meters as the exploration distance, exploring both open parking areas and parking lots. Feedback images are shown below. Figure 5 As shown.

[0151] Figure 6 This is a flowchart illustrating a parking method with an autonomous exploration mode according to one embodiment of the present disclosure.

[0152] refer to Figure 6 The parking method S1000 with autonomous exploration mode in this embodiment includes the following steps:

[0153] S1100: The vehicle parking module receives the start control signal generated and sent by the terminal parking module, and starts the vehicle parking module based on the start control signal;

[0154] S1200: The vehicle parking module obtains the current position of the vehicle and determines whether there is historical trajectory information of the vehicle associated with the current position of the vehicle and including at least one historical parking target position;

[0155] The judgment result is: there is no historical trajectory information of this vehicle that is associated with the current location of this vehicle and includes at least one historical parking target location;

[0156] S1402a: The terminal parking module outputs a control interface. Based on the control interface, the terminal parking module receives user control (touch operation or voice control).

[0157] S1404a: The terminal parking module generates a parking exploration command, including exploration range command, exploration distance command, and parking location exploration command, in response to the received user control, for user confirmation;

[0158] S1406a: The terminal parking module receives the confirmation command and sends the parking exploration command to the vehicle parking module to execute the parking control process of the autonomous exploration mode; S1406b: If the terminal parking module does not receive the confirmation command or receives the denial command, it outputs the control interface again to receive the user's control to execute step S1404a again.

[0159] Among them, the historical trajectory information of the vehicle associated with the current location of the vehicle can be the historical trajectory information of the vehicle that is less than a preset distance from the current location of the vehicle, such as within a circle with a radius of 1000 meters centered on the current location of the vehicle. Any definition of the historical trajectory information of the vehicle associated with the current location of the vehicle by those skilled in the art under the guidance of the technical solution of this disclosure shall fall within the protection scope of this disclosure.

[0160] The current location information of this vehicle described in this disclosure can be obtained based on satellite positioning data (such as GPS positioning system or Beidou positioning system).

[0161] The vehicle's historical trajectory information can be obtained or updated by continuously accumulating the vehicle's map-based navigation journey trajectory, and the navigation destination can be used as the historical parking target location.

[0162] Figure 7 This is a flowchart illustrating another embodiment of the parking method with an autonomous exploration mode.

[0163] refer to Figure 7 The parking method S1000 with autonomous exploration mode in this embodiment includes the following steps:

[0164] S1100: The vehicle parking module receives the start control signal generated and sent by the terminal parking module, and starts the vehicle parking module based on the start control signal;

[0165] S1200: The vehicle parking module obtains the current position of the vehicle and determines whether there is historical trajectory information of the vehicle associated with the current position of the vehicle and including at least one historical parking target position;

[0166] The judgment result is: There is historical trajectory information of this vehicle that is associated with the current location of this vehicle and includes at least one historical parking target location;

[0167] S1402c: The on-board parking module outputs the vehicle's historical trajectory information, which is associated with the vehicle's current location and includes at least one historical parking target location, to the terminal parking module for selection;

[0168] S1404c: The terminal parking module receives the historical parking target location selection instruction, and the vehicle parking module executes the parking control process based on the historical parking target location selection instruction.

[0169] Figure 8 This is a flowchart illustrating another embodiment of the parking method with an autonomous exploration mode.

[0170] refer to Figure 8 The parking method S1000 with autonomous exploration mode in this embodiment includes the following steps:

[0171] S1100: The vehicle parking module receives the start control signal generated and sent by the terminal parking module, and starts the vehicle parking module based on the start control signal;

[0172] S1200: The vehicle parking module obtains the current position of the vehicle and determines whether there is historical trajectory information of the vehicle associated with the current position of the vehicle and including at least one historical parking target position;

[0173] The judgment result is: There is historical trajectory information of this vehicle that is associated with the current location of this vehicle and includes at least one historical parking target location;

[0174] S1402c: The on-board parking module outputs the vehicle's historical trajectory information, which is associated with the vehicle's current location and includes at least one historical parking target location, to the terminal parking module for selection;

[0175] S1404d: If the terminal parking module does not receive a historical parking target location selection instruction (including receiving an operation instruction to abandon the historical parking target location), the terminal parking module outputs a control interface, which is used to receive user commands.

[0176] S1406d: The terminal parking module generates a parking exploration command, including the exploration range command, the exploration distance command, and the parking location exploration command, in response to the user's operation received from the control interface, for the user to confirm.

[0177] S1408d: The terminal parking module receives the confirmation command and sends the parking exploration command to the vehicle parking module to execute the parking control process of the autonomous exploration mode; S1408e: If the terminal parking module does not receive the confirmation command or receives the denial command, it outputs the control interface again to execute step S1406d again.

[0178] Figure 9 This is a flowchart illustrating a parking method with an autonomous exploration mode, which is yet another embodiment of this disclosure.

[0179] refer to Figure 9 In some embodiments of this disclosure, the parking method S1000 with autonomous exploration mode described above further includes, between step S1100 and step S1400:

[0180] S1200: The terminal parking module outputs the first voice signal for interaction;

[0181] S1300: The terminal parking module determines whether it has received a first characteristic voice signal in response to the first voice signal.

[0182] If the terminal parking module receives a first characteristic voice signal (representing the use of the parking system) in response to the first voice signal, then proceed to step S1400.

[0183] Preferably, in step S1200 described in this disclosure, the vehicle parking module performs the following steps: obtaining the current location of the vehicle and determining whether there is historical trajectory information of the vehicle associated with the current location of the vehicle and including at least one historical parking target location; the terminal parking module outputs a first voice signal for interaction (e.g., asking the user whether to use the parking system).

[0184] The terminal parking module described in this disclosure preferably includes a parking exploration command generation module, which generates exploration range commands, exploration distance commands, and parking location exploration commands in response to user operations.

[0185] For the parking method S1000 of this disclosure, after the terminal parking module generates a parking exploration command and sends it to the vehicle terminal module, preferably, the parking control flow S1410 with autonomous exploration mode described in this disclosure is executed based on the following control method:

[0186] S1412: The on-board parking module determines whether the vehicle meets the starting conditions. If the starting conditions are met, the on-board parking module generates a control signal to control the vehicle to start.

[0187] S1414: The on-board parking module obtains the feasible area of ​​the current road, performs local path planning in real time according to the road conditions, and generates control signals to control the vehicle to drive along the feasible area.

[0188] S1416: The on-board parking module obtains the branch road information and performs local path planning based on the branch road information to update the feasible region;

[0189] S1418: When the vehicle parking module discovers a parking location, it executes the parking process based on the discovered parking location.

[0190] Figure 10 This is a schematic diagram of a parking control process with an autonomous exploration mode, according to one embodiment of this disclosure.

[0191] The parking control process with autonomous exploration mode of this disclosure is illustrated by an example:

[0192] In step S1412, after confirming the program instructions, the vehicle parking module begins to detect whether the vehicle meets the starting conditions. Once the starting conditions are met, the vehicle parking module generates a control signal to control the vehicle to start.

[0193] For example, the following judgment process can be used to detect whether the vehicle meets the starting conditions:

[0194] First, the driver departure judgment is initiated by installing a pressure sensor and multiple sensing probes on the seat of the car chassis. When the overall pressure signal value is lower than the preset pressure value, the photoelectric signal judgment is initiated by installing a photoelectric sensor on the seat back to determine whether the photoelectric signal value is higher than the preset value. If the photoelectric signal value is higher than the preset value, it is determined that the driver has left the seat.

[0195] Then, the vehicle sensors determine whether the doors are closed. If the doors remain open for an extended period, the vehicle parking module sends a warning signal to the user terminal device. If the door closure time exceeds a certain period (default is 10 seconds, which the user can adjust to, for example, 10 to 60 seconds), the starting conditions are deemed met, and a control signal is generated to start the vehicle.

[0196] In step S1414, the vehicle parking module obtains the feasible region of the current road and generates a control signal to control the vehicle to travel along the feasible region.

[0197] After the vehicle starts, it uses cameras to create a SLAM map of the surrounding environment. Specifically, it can use a combination of visual and IMU methods, such as VINS or ORBSLAM-V3, or a pure visual method, such as ORBSLAM-V2, to reconstruct the scale of the pure visual map based on the vehicle's odometer reading.

[0198] Based on camera recognition of road edges and lane lines, any existing method for identifying lane lines and road edges can be used. For example, one or more fisheye or pinhole cameras can be used to identify lane lines and road edges using commonly used converted top-view images.

[0199] Specifically, based on the intrinsic and extrinsic parameter matrices of a fisheye / pinhole camera, an IPM (Integrated Perspective) is used to obtain a top view. On this top view, lane lines and road edge pixels are obtained using traditional edge detection or deep learning-based methods. Then, through clustering and fitting, vectorized lane lines and road edges are obtained. The top view and the real physical 2D space of vehicle movement have a consistent scale relationship, and the pixel coordinates of the lane lines and road edges specifically represent their positions relative to the vehicle. Alternatively, a BEV (Battery Electric Vehicle) viewpoint feature extraction and recognition technique similar to Tesla's can be used. First, features are extracted from the original camera image. Then, BEV viewpoint features are obtained through cross-attention or neural attention mechanisms, followed by lane line and road edge recognition. The BEV viewpoint recognition results have a consistent scale relationship with the real physical 3D space of vehicle movement.

[0200] After identifying the road edge, the onboard parking module controls the vehicle to explore along the right edge of the road at a speed not exceeding a preset speed, such as 15 km / h. During the journey, it maintains a preset distance (e.g., 50 cm) from the edge or edge obstacles to facilitate vehicle steering and aligns the vehicle's direction of travel horizontally with the road edge. Simultaneously, it marks the primary and secondary feasible zones of the current road: the primary feasible zone is the right lane of a multi-lane road excluding the 50 cm edge, and the secondary feasible zone is the area around both road edges excluding the primary feasible zone.

[0201] S1416: The on-board parking module obtains the branch road information and performs local path planning based on the branch road information to update the feasible region.

[0202] When the vehicle is not encountering an intersection, the onboard parking module performs real-time local path planning, simulating a route map of the vehicle traveling along the right edge of the road within its field of vision. After identifying an intersection, it analyzes the intersection's conditions (e.g., whether it is passable) and performs local path planning.

[0203] Figure 11 This is a route map of one embodiment of the present disclosure, showing the vehicle traveling along the right edge of the road within its field of vision when not encountering an intersection.

[0204] Among them, the thick black solid line represents the actual road edge line, the black dashed line represents the dividing line between the first and second level feasible regions, the blue-gray area represents the second level feasible region, the white area represents the first level feasible region, and the red arrow line represents the preset path trajectory.

[0205] In step S1414 described above, if an obstacle is detected, preferably, the following local path planning process is also included:

[0206] Determine the type of obstacle;

[0207] For static obstacles, a detour path is planned based on the size of the static obstacle, and the on-board parking module generates control signals to control the vehicle to detour around the static obstacle along the detour path;

[0208] If a dynamic obstacle is encountered, the vehicle parking module will remain stationary at a preset distance from the obstacle for a first preset time length. If the dynamic obstacle moves, the vehicle parking module will generate a control signal to allow the vehicle to continue driving. If the dynamic obstacle does not move, the vehicle parking module will generate a control signal to allow the vehicle to issue a warning signal (light or sound). If the dynamic obstacle remains stationary for a second preset time length, it will be converted into a static obstacle.

[0209] In step S1414, when the vehicle's predetermined route is blocked due to an obstacle, it is first determined whether the obstacle is a static obstacle or a dynamic obstacle.

[0210] According to a preferred embodiment of this disclosure, if the obstacle is determined to be a static obstacle, an obstacle avoidance measure of detour is selected. A curved path that meets the maximum turning angle limit of the vehicle is planned on the side of the obstacle. When planning the local path, the curve with the maximum turning radius in the first-level feasible region is preferentially selected as the alternative route. If there is no route that meets the minimum turning radius in the first-level feasible region, the curve with the smallest turning radius in the feasible region (including the first-level and second-level feasible regions) is selected as the alternative route.

[0211] If the obstacle ahead is determined to be a dynamic obstacle of the type of motor vehicle, non-motor vehicle, pedestrian, or animal, the vehicle will stop within a preset distance, such as one meter, for a first preset time length (e.g., 20 seconds). After that, if the dynamic obstacle does not move, the on-board parking module generates a control signal to control the vehicle to give a light signal to indicate the dynamic obstacle. It then determines whether the dynamic obstacle has moved (which can be determined by the on-board radar signal). If the movement does not affect the vehicle's driving route, the vehicle continues to drive. If the dynamic obstacle remains stationary for a second preset time length (e.g., 2 seconds), it is converted into a static obstacle, and the process for handling static obstacles described above is executed to avoid the obstacle.

[0212] Figure 12 This is a schematic diagram illustrating the obstacle avoidance and detour of the vehicle when encountering an obstacle, according to one embodiment of this disclosure.

[0213] In some embodiments of this disclosure, when a planned driving path curve with a turning angle exceeding the vehicle's maximum turning angle appears within the vehicle's preset route, the normal vector directions of the preceding and following road segments of that curve (the route that does not satisfy vehicle dynamics) are obtained to plan a new curve path that satisfies the vehicle's minimum turning radius. This new curve path replaces the aforementioned route that does not satisfy vehicle dynamics. Figure 13 .

[0214] Figure 13 This is a schematic diagram of obtaining a new curved path that satisfies the minimum turning radius of the vehicle during a cornering process according to one embodiment of this disclosure.

[0215] The red dashed line represents the original preset route, while the red solid line represents the new curved path obtained after considering the maximum turning angle of the car, where r is the minimum turning radius.

[0216] For the parking method S1000 of the above embodiments, step S1416 may further include a junction identification process, preferably including the following junction determination method:

[0217] There are more than two road edge lines that do not intersect in front of this vehicle;

[0218] Extend the endpoints of each road edge line along the tangent direction of the endpoints. If the extended lines of two endpoints intersect within a circle with these two endpoints as its diameter, then the two road edge lines are considered as one road edge line. Based on this determination, if there are no more than two road edge lines in front of the vehicle, then it is not considered a fork in the road (see reference). Figure 14 ).

[0219] Figure 14 A schematic diagram illustrating intersection identification according to one embodiment of this disclosure is shown. In this disclosure, intersections are preferably identified based on whether the road edge line is broken (i.e., the continuity of the road edge line).

[0220] Figure 14 In the diagram, the solid black line represents the road edge line perceived by the car. The reason for this road edge break may be due to the presence of an obstacle, or the location of a doorway or entrance to a space, causing the road edge line perceived by the car to break. In this case, it is not considered a fork in the road.

[0221] In some embodiments of this disclosure, to avoid the impact of median strips on intersection identification, it is desirable to temporarily ignore the presence of median strips when identifying intersections. Preferably, the intersection identification method further includes:

[0222] Within the vehicle's field of vision, if the distance between the two endpoints of an edge line is less than a preset distance, or if the edge line is a closed curve and the area divided by the edge line is less than a preset area, then the edge line is determined to be a median strip, and this edge line is ignored when identifying intersections.

[0223] Figure 15 A schematic diagram illustrating the determination of median strips according to one embodiment of this disclosure is shown. The gray area describes the region divided by each boundary line (road edge line), while the blue curve describes a median strip on the current road. This median strip is characterized by a small endpoint distance (less than a preset distance) and a small area of ​​the divided region (less than a preset area). Such median strips are ignored when determining intersections.

[0224] This disclosure also provides a method for handling intersections. Figure 16 A schematic flowchart of a fork-in-the-road processing method S1420 according to one embodiment of the present disclosure is shown.

[0225] refer to Figure 16 This document illustrates a method for processing intersections according to one embodiment of the present disclosure, using an example to explain the method for processing intersections according to the present disclosure.

[0226] S1421: After the vehicle parking module determines that an intersection is ahead, it analyzes the intersection situation and constructs a two-dimensional grid map based on the intersection recognition results. The space is divided into grids, with a preset grid side length of 0.2 meters (adjustable) to reduce the amount of computation while ensuring the algorithm's performance.

[0227] Figure 17 A two-dimensional grid map constructed for a fork in the road is shown as one embodiment of this disclosure.

[0228] S1422: After establishing the two-dimensional grid map, based on each edge line of the intersection, assign a value of 0 to all grids covering that edge line. Then, assign values ​​to the grids adjacent to all assigned grids within the feasible region of the two-dimensional grid map. For grids sharing an edge, add a first value, such as 1; for grids sharing a corner, add a second value, such as 1.5. If a grid shares an edge or a corner with multiple grids, select the value with the smallest result after adding the first value to the grids sharing the edge or the second value to the grids sharing the corner, and assign the grid value until all grids within the feasible region are assigned values.

[0229] Figure 18 and Figure 19 An exemplary diagram illustrates grid assignment based on an edge line (the road edge line to the right front of the vehicle).

[0230] Complete the grid assignment for each road edge line (right road edge line of this vehicle, right front road edge line, left road edge line, left front road edge line) based on the intersection.

[0231] Figures 20 to 22 The grid assignment results for the right-side road edge line, left-side road edge line, and left-front road edge line of this vehicle are shown as an example.

[0232] S1423: Perform raster map-based assignment overlay on the raster assignment results based on each edge line to obtain an initial overlay assignment map.

[0233] Ignore or remove grid assignments greater than a preset threshold (e.g., 14) in the initial overlay assignment map to obtain the intersection assignment map.

[0234] Figure 23 An exemplary diagram of the fork intersection assignment of this disclosure is shown.

[0235] S1424: Take the center point of the grid with the smallest superimposed value in the intersection assignment map as the intersection center point and obtain its coordinates. If there are multiple grids with the smallest superimposed value, then their geometric center is taken as the center point of the intersection.

[0236] Preferably, the center point of the intersection is obtained using the following method.

[0237] Get the center point coordinates of the grid with the smallest superimposed value. … Then their geometric center coordinates are:

[0238] ;

[0239] S1425: After obtaining the center point of the intersection, obtain the inflection points on the edge lines of each road based on the center point of the intersection.

[0240] The vehicle parking module finds the edge point on each road edge that is closest to the center point of the intersection. As an inflection point on the edge line.

[0241] For example, this disclosure uses the following method to obtain the inflection point.

[0242] Let the expression for the edge line parameters be: Finding through numerical solutions , making the function Minimum, then .

[0243] After obtaining the inflection points of each road edge line, compare the distances from the inflection points on each edge line to the center point of the intersection, and take the largest distance as the radius to draw a circle. The intersection area of ​​this circle with the entire feasible region is taken as the travel region. The non-intersecting regions divided by this circle are taken as branches, that is, the sub-paths of the intersection.

[0244] Figure 24 and Figure 25 This is a branching analysis diagram of one embodiment of the present disclosure, which exemplarily illustrates the traffic area and sub-paths.

[0245] After obtaining the traffic area and sub-paths, using the center point of the intersection as the center and the current position of the vehicle as the starting point, search for intersections (i.e., sub-paths) clockwise or counterclockwise. The first intersection passed clockwise is designated as the left-turn lane, and the first intersection passed counterclockwise is designated as the right-turn lane. At the same time, the sub-path with the deflection angle closest to 180 degrees is designated as the straight-ahead lane.

[0246] Based on the intersection processing method S1420 disclosed herein, the intersection traffic domain and each sub-path are obtained.

[0247] Furthermore, the parking method S1000 disclosed herein also includes the following steps:

[0248] S1426: When the vehicle parking module performs a turn at a fork in the road, it takes the vehicle's position before the traffic area as the starting position and the predicted position of the vehicle at the fork (i.e., sub-path) after the traffic area as the target position. It performs path planning through the traffic area and initially plans an arc path as a local planning path to control the vehicle's movement.

[0249] Figure 26 This is a partial planning path diagram of one embodiment of the present disclosure. Figure 26 In the diagram, the solid red line represents the planned path for a left turn at a fork in the road, and the dashed red line represents the planned path for a right turn at a fork in the road.

[0250] Considering that cars have a minimum turning radius, when the initial local planning path does not satisfy vehicle dynamics, it is necessary to advance the vehicle's turning start position and delay the vehicle's turning end position.

[0251] Figure 27 This illustration shows a left turn at a fork in the road according to one embodiment of the present disclosure, where the starting position of the turn is advanced and the ending position of the turn is delayed. Figure 27 In the image, the black vehicle represents the initial and final positions of the vehicle's turn, as planned during the preprocessing stage.

[0252] When the distance r between the intersection of the normals of the black vehicle's starting and ending turning positions and the starting and ending turning positions is less than the vehicle's minimum turning radius R, it is necessary to control the vehicle (red) to start turning earlier and end turning later. The local planned path is shown as the red solid line.

[0253] The parking method S1000 in some embodiments of this disclosure also includes a fork-way turning processing method, which includes the following processing steps:

[0254] The first step involves the vehicle using its vision system (image acquisition device, etc.) to attempt to identify whether there are parking signs at the intersection. If a parking sign is found, the vehicle will prioritize turning in the direction indicated by the parking sign. Deep learning methods can be used to identify parking signs and turn markings; for example, refer to the paper "Signboard Detection and Text Recognition Using Artificial Neural Networks".

[0255] Figure 28 An example image of a parking lot sign is shown.

[0256] If the parking lot sign cannot be identified, the second step is executed: the on-board parking module determines whether the intersection has been recorded. If the intersection has been recorded, the vehicle will be preferentially turned into an unrecorded side road.

[0257] If the intersection and all branch paths have been recorded, proceed to step six. If the intersection has not been recorded, or if there are more than one branch path (excluding the currently traveled road) that has not been recorded, then proceed to step three, which determines whether the parking instruction includes parking in an open parking area. If it does, proceed to step four; otherwise, proceed to step five.

[0258] The fourth step is to try to determine if there are multiple parked vehicles on either side of the road ahead. If so, turn towards the side with parked vehicles. If multiple roads have parked vehicles, turn towards the side with the most parked vehicles. If no parked vehicles are found on any side roads, prioritize turning towards the side farther from the boundary of the maximum exploration range. If the distances to the boundary are the same or approximately the same on several sides, turn to the right.

[0259] The fifth step is to prioritize turning towards the side with the wider road. If there is no significantly wider road (e.g., more than 25% wider than other roads), then turn towards the side farther from the boundary of the maximum exploration range (in the tangent direction of the edge of the field of view). If the distances to the boundary are the same or approximately the same on all sides, prioritize turning towards the side with higher "field of view" (field of view: fewer buildings on both sides of the fork in the road within the visible range). If the field of view is the same or approximately the same on both sides, prioritize turning towards the right.

[0260] According to a preferred embodiment of this disclosure, the visibility of two forks in the road is compared using the following method:

[0261] First, the current view image is divided into multiple sections using the branch road boundary curves. The number of sections equals the number of branches (e.g., a T-junction has three sections, and a crossroads has four sections). Figure 29 As shown, Figure 29 An example diagram of a branch road section is shown.

[0262] In some embodiments of this disclosure, when a branch road is partially invisible due to obstruction of view by a building, making it impossible to divide the road into sections, the branch road is extended to complete the section by extending the visible portion.

[0263] The preferred method used in this disclosure is as follows:

[0264] Connect the two endpoints of the fault, or extend along the tangent direction at the endpoints of the fault.

[0265] Figure 30 An example diagram of the processing of field of view occlusion is shown.

[0266] In some embodiments of this disclosure, the vehicle's field of view image is preferably segmented into buildings and other objects using semantic segmentation.

[0267] Figure 31 The image shows a segmentation map of the plate pixels in the field of view.

[0268] Figure 31 In the image, the black area at the top represents buildings, while the pixels in other areas represent other semantic objects. Figure 31 Given the original fisheye image, pixels semantically categorized as buildings are projected onto the ground based on the camera's intrinsic parameter matrix and its extrinsic parameters relative to the vehicle, thus constructing a two-dimensional raster map.

[0269] The space is divided using a grid, with a preset grid side length of 0.2 meters (adjustable). This reduces computational load while maintaining algorithm performance, using the same method described above. Then, a branch path is selected, and the marker numbers are incremented sequentially outwards, such as... Figure 32 As shown, Figure 32A diagram illustrating the field of view analysis is shown.

[0270] Figure 32 Each number in the grid corresponds to a weight for that cell, specifically the reciprocal of that number. For example, the cell with the number 3 has a weight of 1 / 3. Therefore, the field of view of this branch path is: Where D stands for domain, representing the set of grid cells labeled with numbers, and g stands for grid, representing the grid cells. This refers to an unoccupied grid, which represents pixels within a grid that contain no buildings. That is, weight, which represents the weight of the grid.

[0271] The sixth step is to determine whether the vehicle's trajectory has formed a closed loop. A closed loop means that the vehicle's trajectory has a closed loop, and there are no intersections with unmarked forks on the trajectory curve except for the starting point. If the trajectory has a closed loop, the parking is considered a failure. If a closed loop has not been formed, the nearest intersection containing unmarked forks is set as a temporary target, and the vehicle is controlled to move towards that intersection.

[0272] After recognizing an intersection ahead, if the intersection is not recorded, the vehicle will take a picture of the intersection and record it, then upload it to the terminal parking module (app). After completing the turn, if there are any unrecorded exits or forks in the road, the vehicle will record those forks.

[0273] During vehicle operation, the parking system (including the onboard parking module and the terminal parking module) records the vehicle's route, location, and distance in real time, and displays this information as a map on the terminal parking module for users to view at any time. Figure 33 As shown, Figure 33 A diagram illustrating the driving process is shown.

[0274] Figure 33 In the diagram, the black dot in the center at the bottom represents the vehicle's starting position when the parking system is activated. The green dots closer to the vehicle represent the intersections recorded by the vehicle. Users can click on the green dots to view photos of the intersections already visited. If the user observes an error in the vehicle's movement through the photos, they can pause the vehicle at any time. The vehicle will pull over, activate its hazard lights, and take 360-degree surround-view photos, sending the photos to the terminal device (user). At this point, the user can recall the vehicle. Figure 33 The blue numbers represent the distance the vehicle has traveled (500m), while the red numbers represent the distance the vehicle is from the starting point (300m). Figure 33 The red 300m represents the distance from the origin to the vehicle, and the blue 500m represents the distance the vehicle has traveled.

[0275] When the vehicle reaches the circular / fan-shaped boundary, the system analyzes whether the vehicle's trajectory has formed a closed shape. If the trajectory forms a closed shape, the parking is considered a failure. If a closed shape has not been formed, the nearest intersection containing unmarked side roads is set as a temporary target, and the vehicle is controlled to turn around and head towards that intersection.

[0276] The parking control process with autonomous exploration mode disclosed herein also includes a parking processing method. The on-board parking module identifies the parking lot / parking area, and after successful identification, it further identifies the target parking space.

[0277] Figure 34 This is a schematic flowchart of a parking processing method according to one embodiment of the present disclosure.

[0278] refer to Figure 34 The parking processing method S1430 of this embodiment includes the following steps.

[0279] S1431: The vehicle parking module detects parking locations (parking lots, roadside parking areas, open parking areas, etc.).

[0280] S1432: If the parking location is a roadside parking area, proceed to step S1433; if the parking location is a parking lot, proceed to step S1434; if the parking location is an open space / plaza parking area, proceed to step S1435.

[0281] S1433: Identify roadside parking areas, and the on-board parking module generates control signals to control the vehicle to enter the roadside parking area;

[0282] S1434: Identify the parking lot, identify the parking area, identify the parking space number, and generate a control signal to control the vehicle to drive into the parking space.

[0283] S1435: Identify open spaces / plazas, and the on-board parking module generates control signals to control the vehicle to enter the open spaces / plazas;

[0284] S1436: When steps S1433 / S1434 / S1435 are executed successfully, the vehicle parking module generates a control signal to control the vehicle to stop and generates parking success information to send to the user terminal device (e.g., mobile phone); when steps S1433 / S1434 / S1435 fail to execute, the vehicle parking module generates parking failure information and sends it to the user terminal device (e.g., mobile phone).

[0285] During the execution of steps S1433 / S1434 / S1435, if the road in front of the vehicle is blocked, the on-board parking module generates a control signal to control the vehicle to stop.

[0286] In the parking processing method disclosed herein, whether parking is successful or fails, a top-view image of the vehicle can be captured, recorded, and sent to the user terminal device.

[0287] The parking processing method of this disclosure is described in detail below with specific examples. In step S1433, the on-board parking module analyzes whether there are any parked vehicles on both sides of the road ahead during the vehicle's driving process. If there are no parked vehicles, the vehicle continues to drive. If there are parked vehicles, the module determines whether there is a no-parking sign on the current road. If there is a no-parking sign, the vehicle continues to drive. If there is no no-parking sign, the on-board parking module generates a control signal to make the vehicle attempt to identify whether there are parking space lines on the left and right sides of the road. The parking space line identification method can use a fisheye camera to generate a top view, and on the top view, a deep neural network model is used to detect and classify parking spaces.

[0288] Once parking space lines are detected, for example via ultrasonic radar, the system will double-check whether the parking space is vacant. If parking space lines are only marked on one side, the nearest vacant parking space on that side will be selected for parking. If parking space lines are marked on both sides, the nearest vacant parking space on the right side will be selected for parking.

[0289] For specific parking techniques, please refer to the document "Development of an automatic parking system for vehicle".

[0290] If there are no parking lines on either side of the road, the camera detects whether there are parked vehicles on either side. If so, ultrasonic ranging is used to find an empty area exceeding 1 meter (adjustable) in length of the vehicle body in front of the side with parked vehicles (default is the right side). If an empty area is detected, the vehicle is controlled to drive into the empty area, parking is successful, and the onboard parking module generates parking success information in step S1436. If no empty area is detected or there are no vehicles on either side of the road, parking fails, and the onboard parking module generates parking failure information in step S1436.

[0291] In step S1434, the vehicle parking module generates a control signal to control the vehicle to continue driving forward along the current road feasible region. During the journey, it attempts to identify whether there is a parking lot ahead. Specifically, the identification method is to train the system to recognize parking signs, such as "P" or "parking lot", through deep learning. Figure 35 This is an example image of a parking lot sign. If a parking lot sign is detected, the road intersection ahead is identified as a parking lot entrance. If no parking lot sign is detected, the vehicle continues forward along the current road, while simultaneously determining whether there is a sudden change in the width of the feasible area ahead. The feasible area width is defined as the length of the feasible area perpendicular to the current driving direction.

[0292] If the vehicle detects a sudden change in the width of the feasible area ahead, and it is not at a fork in the road, and the width change is more than a preset multiple of the current feasible area width of the vehicle, such as more than three times, then it is preliminarily determined that it has arrived at an open parking lot.

[0293] After initially determining the parking lot as an open area, the vehicle further observes whether there are stationary vehicles in the area where the road width changes. If there are more than a preset number of stationary vehicles in the area ahead, such as more than two, that are not on the current road extension line, then it is confirmed as an open parking lot. If there are no such stationary vehicles, then the vehicle captures an image of the area ahead and sends it to the user's terminal device to ask the user whether the parking lot location is correct. Figure 36 This is an example diagram of an open parking lot.

[0294] If the user confirms the parking lot, proceed to the next step; if the user denies the parking lot, continue exploring.

[0295] If a parking lot cannot be identified even at the end of the road or at a fork in the road, the on-board parking module generates parking failure information in step S1436 and sends it to the user terminal device.

[0296] Once the parking lot is successfully identified, it is determined whether the parking instructions in the route navigation program specify the parking area and location.

[0297] If no specific instructions are given, refer to step S1433. First, try to identify the parking space lines and find the nearest empty space to park based on the parking space lines. If there are no parking space lines, find a sufficiently empty area. The specific search method can also be to use ultrasonic radar ranging to find whether there is an empty distance of at least one and a half car lengths between two vehicles parked side by side. If there is, park the vehicle in the empty area.

[0298] If the user provides specific instructions regarding the parking area and location, then the following steps will be executed:

[0299] The vehicle parking module first determines whether the user has given a zone number instruction. If the user has given a specific zone number instruction, it uses deep learning to find and identify English letters or numbers on nearby signs, as well as potential left turn signs, right turn signs, and straight signs. For specific identification methods, please refer to the literature "Signboard Detection and Text Recognition Using Artificial Neural Networks". Figure 37 This is an example image of parking area signs (area numbers) in a parking lot.

[0300] refer to Figure 37The vehicle parking module first identifies the current location, and the English letters can be identified as their corresponding numbers in the English alphabet.

[0301] When the system detects a zone number equal to the user's command, it enters that zone. If no zone number is detected, the onboard parking module attempts to identify nearby zoning signs. If a sign is present, the vehicle moves in the direction indicated by the sign; otherwise, the onboard parking module directs the vehicle to continue along the current road. At intersections, the system attempts to identify the zone number of the area enclosed by the edges of each intersection and veers towards the side closer to the specified zone number. This process continues until the specified zone number is found.

[0302] Once the vehicle enters the designated area number, or if the user does not provide an area number command, the on-board parking module will determine whether the user has provided a specific parking space number.

[0303] If the user only provides the area number instruction but not the parking space number instruction, then refer to step S1433, first try to identify the parking space line, find the nearest empty space to park according to the parking space line, if there is no parking space line, then use ultrasonic radar to measure distance, find at least one and a half car body vacancy between two vehicles parked side by side, and park the vehicle in the vacant area.

[0304] If the user provides a specific parking space number, the system converts the indicated parking space number into a number and simultaneously identifies nearby parking space numbers. It then proceeds forward along the current road, attempting to identify parking space number signs when encountering a fork in the road. If a parking space number sign exists, it determines whether the target parking space number matches the sign's numerical range. If it does, it turns in the direction indicated by the sign. If there is no sign, it identifies the nearest parking space number to the vehicle's current position on each fork in the road and selects the direction with the smaller absolute value of the difference between the target parking space number and the current one. If the system detects that the vehicle has already entered an area, it prioritizes turning in the direction not previously selected. If the target parking space is successfully identified and is available, the system controls the vehicle to enter the parking space. If the above process results in an infinite loop, or if the target parking space is found to be obstructed after successful identification, the onboard parking module generates a parking failure message in step S1436 and sends it to the user terminal device.

[0305] In step S1435, the vehicle parking module controls the vehicle to move forward along the current road, while simultaneously determining whether there is a sudden change in the width of the feasible area ahead, as per step S1434. If a sudden change in the width of the feasible area ahead is detected, and it is not at a fork in the road, and the width after the change is more than a preset multiple of the current road width (e.g., more than three times), and there are stationary vehicles in the area after the road width change, then it is determined that the vehicle has arrived at a parking space / parking plaza, and then parking is performed as per S1433.

[0306] In step S1400 of this disclosure, the parking control process executed by the on-board parking module based on the historical parking target location selection instruction received by the terminal parking module includes the following steps:

[0307] S1441: Compare the current position of the vehicle with at least one historical trajectory of the vehicle that includes the selected historical parking target position, and obtain the optimal historical trajectory based on the number of times the vehicle's historical trajectory has been driven and the distance between the current position of the vehicle and the historical trajectory of the vehicle.

[0308] Preferably, the vehicle parking module initially obtains the vehicle's current location via GPS, and simultaneously retrieves and analyzes the location information of historical trajectory points. When multiple different historical trajectories exist near the vehicle for the same selected historical destination, the historical trajectory with more trips is selected as the optimal historical trajectory; if multiple historical trajectories with the same number of trips exist, the historical trajectory closer to the current location is selected as the optimal historical trajectory.

[0309] S1442: Determine whether the minimum distance between the optimal historical trajectory and the current position of the vehicle exceeds the preset distance (e.g., 5 meters). If it exceeds, proceed to step S1443; if it does not exceed, proceed to step S1444.

[0310] S1443: The on-board parking module controls the vehicle to move forward along the current feasible domain and continuously marks the historical trajectory point closest to the vehicle's current position to find the optimal historical trajectory. Whenever the vehicle encounters a fork in the road, the on-board parking module controls the vehicle to turn into the fork in the road that the line connecting the marked point points points in. This continues until the minimum distance between the vehicle's current position and the optimal historical trajectory is less than or equal to a preset distance, and then proceeds to step S1444.

[0311] Figure 38 This is an example diagram for finding the optimal historical trajectory.

[0312] S1444: The on-board parking module controls the vehicle to reposition itself in order to move the vehicle to the optimal historical trajectory and align it with the historical pose.

[0313] Preferably, repositioning is performed through the following steps. Figure 39 It is a repositioning trajectory point matching map.

[0314] The vehicle's current location and surrounding environment are obtained through the vehicle's onboard camera. The image is captured and its features are extracted. Then, the historical trajectory points within a preset distance range (e.g., 5 meters) of the current location of the vehicle are retrieved from the historical map data. If multiple historical trajectory points are available for retrieval, a limited number of points, such as 2 to 4 points, that are closest to and furthest from the vehicle are retrieved.

[0315] The mapping information of the retrieved trajectory points is compared with the mapping information of the vehicle's current position. If there are not enough matching feature points, the relocation is determined to be unsuccessful. The vehicle is then controlled to continue driving along the current feasible region, constantly approaching the optimal historical trajectory, and continuously attempting relocation and feature point matching.

[0316] Once the onboard parking module finds a sufficient number of matching feature points and relocalization is successful, the vehicle pose can be solved using RANSAC iteration and the PnP algorithm, based on the surrounding environment of the vehicle's current position. The image range is measured to correct the vehicle's current position information, and the vehicle is controlled to move to the historical trajectory point and align with the historical pose.

[0317] S1445: Plan the optimal global route from the optimal historical trajectory, and control the vehicle to travel to the target location based on the optimal global route.

[0318] After the vehicle arrives at the target location, parking can be processed based on historical parking orders and in accordance with the parking processing methods (S1431 to S1436) described above in this disclosure.

[0319] Preferably, the onboard parking module retrieves the optimal route planned from historical map data and drives at a constant low speed in the direction shown by the optimal route. During driving, local path planning is performed simultaneously, using a positioning system such as GPS for positioning and SLAM mapping of the vehicle's surrounding environment. When the location information contained in the historical trajectory points matching the current location does not match the location information calculated by the vehicle's own inertia, the location information of the previous matching point and the current matching point is updated (detour points taken for obstacle avoidance are ignored here).

[0320] The specific update method is as follows: Let the coordinates of the previous matching point be... The historical coordinate information of the current matching point is The current location's coordinates are calculated using inertial navigation. Due to historical points The location information is based on GPS, which can have a very large error. In contrast, the relative motion vector calculated using inertial navigation... Its error is negligible over short distances compared to errors such as those of GPS; therefore, the updated coordinates of the new points... and Must meet Search and So that in and When the true coordinates of two points are given, the observed values ​​are... and The probability is highest at that time.

[0321] More preferably, this disclosure provides an iterative update method. Based on the weak law of large numbers, it is assumed that the observed coordinates of GPS follow an independent two-dimensional Gaussian distribution. Therefore, this disclosure performs L2 optimization on the point coordinates.

[0322] The result of the optimization is the new coordinates of the previous point. New current point coordinates .

[0323] This process continuously updates the historical trajectory point location information and surrounding environmental features. After traversing the trajectory multiple times, the resulting trajectory coordinate function will converge to the true trajectory coordinate function. Furthermore, other position coordinate optimization methods and other probabilistic cost functions are also applicable to this step.

[0324] If the planned route is interrupted during the vehicle's journey, the onboard parking module can attempt to retrieve a suboptimal route from historical map data and re-enter along that route. If retrieving the suboptimal route fails or the suboptimal route does not exist, the parking is deemed a failure. If the vehicle successfully enters the historical parking destination area, the module checks whether the parking area number and parking space number are specified in the historical parking command, and controls the vehicle to park in a nearby empty / designated parking space, referring to the content described above.

[0325] In step S1436 described above, post-processing for successful / failed parking is preferably performed based on the following steps.

[0326] A. Parking successful:

[0327] The system collects images of the parking area and generates parking success information, which is then sent to the user's terminal device.

[0328] Received confirmation information from the user's terminal device;

[0329] Record the parking process data into the parking map data and plan at least the optimal route;

[0330] Parking is complete, and the on-board parking module is turned off.

[0331] B. Parking failed:

[0332] The on-board parking module pauses, generates a parking failure message, and sends it to the user terminal device to request further instructions.

[0333] If no further instructions are received, a control signal is generated to control the vehicle to return to its initial parking position.

[0334] Upon receiving further instructions, control signals are generated based on these instructions to control the vehicle.

[0335] In this disclosure, the vehicle parking module has two scenarios: one is that the vehicle arrives at the destination normally and parks successfully; the other is that the parking fails due to road errors / malfunctions, such as driving into a dead end, failing to identify a parking lot, having no parking space, and various road blockages.

[0336] More specifically, upon successful parking, the vehicle can take 360-degree surround photos of the vehicle body and then send the 360-degree scene images to the user's terminal device. If the user confirms that the parking is correct, the parking process data is processed. If this trip is a new trip without historical memory, the on-board parking module stores the trajectory points of this trip, including the vehicle's position and attitude information with six degrees of freedom, and aligned with the GPS information, along with the environmental features and distance measurements of the environmental features in the SLAM mapping of the trajectory points, into the historical map data, and generates a trajectory curve through clustering.

[0337] Then, the locations and instructions of all intersections are marked and recorded. By analyzing the geometry of the route, all possible routes with an overlap measure of 0 are extracted, and the total length is compared to store these routes as the optimal to the second-best routes in the vehicle's storage system or the cloud storage system. If this trip is a historical parking mode, the current trip is overwritten with the original trip, with overlapping parts completely updated and non-overlapping parts retained.

[0338] If parking fails or a rejection message is received from the user's terminal device, the onboard parking module instructs the vehicle to move to the roadside within the currently feasible area, stop, activate hazard lights, and send a 360-degree photo of the surrounding environment to the user's terminal device. Simultaneously, the onboard parking module displays the current driving trajectory and GPS navigation location to the user's terminal device. If the current parking trip is a historical parking mode, the historical trajectory map is also displayed to the user's terminal device; previously traveled historical trajectories are marked with solid lines, and untraced historical trajectories are marked with dashed lines. After an error is reported, the module requests further instructions from the user. If the user does not respond for an extended period, the vehicle will (by default) automatically return to the parking start position; this default setting can be modified by the user. Further instructions received by the onboard parking module from the user include: returning to the parking start point, parking on the roadside ahead, and new instructions within the parking lot.

[0339] For example, when the user instructs the vehicle to return to the initial parking position, the onboard parking module directs the vehicle to make a U-turn on the current road and return to the initial position along the original route. After successfully returning to the initial parking position and repositioning, parking ends, and the onboard parking module shuts down.

[0340] When the user instructs the vehicle to park on the side of the road ahead, the on-board parking module directs the vehicle to move forward and identifies the current road width. When the feasible road width ahead is more than a preset number, such as four vehicle widths or more, and there is no no-parking sign or parking space lines, the on-board parking module directs the vehicle to park in the vacant area on the right side of the road.

[0341] When parking fails and the vehicle is in the parking lot, the user can provide new instructions within the parking lot, including a new zone number, parking space number, or parking in any location.

[0342] This disclosure also provides a route map-based parking system, including:

[0343] Terminal parking module 1000 generates and sends out a start control signal;

[0344] The vehicle parking module 2000 receives the start control signal and starts.

[0345] The terminal parking module 1000 includes:

[0346] The parking exploration command generation module 1002 generates exploration range commands, exploration distance commands, and parking location exploration commands in response to user operations.

[0347] The vehicle parking module 2000 includes:

[0348] The historical trajectory output module 2002 is used to output the historical trajectory information of the vehicle that is associated with the current position of the vehicle and includes at least one historical parking target position.

[0349] The parking processing module 2004 generates parking exploration instructions or selects historical parking target locations and their corresponding historical trajectory information based on the terminal parking module 1000 and executes the parking control process.

[0350] Among them, the parking processing module 2004 of the vehicle parking module 2000 explores parking locations based on the exploration range and exploration distance limits imposed by the exploration range command and the exploration distance command.

[0351] Preferably, the terminal parking module 1000 further includes:

[0352] The voice signal output module 1006 outputs a first voice signal for interaction (asking whether to use the parking system).

[0353] The voice signal receiving module 1008 is used to receive the user's voice signal.

[0354] The first characteristic speech signal acquisition module 1010 determines whether a first characteristic speech signal in response to a first speech signal is received.

[0355] The touch command receiving module 1012 is used at least to receive historical parking target location selection commands.

[0356] Figure 40 and Figure 41 The structural schematic block diagrams of the terminal parking module 1000 and the vehicle parking module 2000 are shown respectively.

[0357] The terminal parking module 1000 disclosed herein can be installed in a user terminal device (such as a mobile phone) through a computer software program architecture, and the vehicle parking module 2000 disclosed herein can be installed in a vehicle processing system with hardware devices such as memory and processor through a computer software program architecture.

[0358] This disclosure also provides an electronic device, including: a memory storing execution instructions; and a processor executing the execution instructions stored in the memory, causing the processor to perform a method according to any embodiment of this disclosure.

[0359] This disclosure also provides a readable storage medium storing executable instructions, which, when executed by a processor, are used to implement the method of any embodiment of this disclosure.

[0360] This disclosure also provides a computer program product, including a computer program / instructions, and a method for implementing any embodiment of this disclosure when the computer program / instructions are executed by a processor.

[0361] In the description of this specification, the references to terms such as "one embodiment / mode," "some embodiments / modes," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment / mode or example is included in at least one embodiment / mode or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment / mode or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments / modes or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments / modes or examples described in this specification, as well as the features of different embodiments / modes or examples.

[0362] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0363] Those skilled in the art should understand that the above embodiments are merely for illustrating the present disclosure and are not intended to limit the scope of the disclosure. Those skilled in the art can make other changes or modifications based on the above disclosure, and these changes or modifications still fall within the scope of the present disclosure.

Claims

1. A parking method having an autonomous exploration mode, characterized by, Comprising: S1100: The vehicle-mounted parking module receives the start control signal generated and sent by the terminal parking module, and starts based on the start control signal; And S1400: The terminal parking module generates a parking exploration instruction in response to the user's operation and sends it to the vehicle-mounted parking module, and the vehicle-mounted parking module executes the parking control process in the autonomous exploration mode based on the parking exploration instruction; or, the vehicle-mounted parking module sends the vehicle historical trajectory information associated with the current position of the vehicle and including at least one historical parking target position to the terminal parking module, and executes the parking control process based on the historical parking target position selection instruction received by the terminal parking module; Wherein, the parking exploration instruction includes exploration range instruction, exploration distance instruction and parking site exploration instruction, and the vehicle-mounted parking module performs parking site exploration within the exploration range and exploration distance limited by the exploration range instruction and the exploration distance instruction; The parking exploration instruction is generated by the following steps: Generate a four-quadrant interface, each quadrant region of the four-quadrant interface can be selected in response to the user's operation to generate an exploration range instruction, and the exploration range instruction is an exploration range instruction with one or more than two quadrant regions as the exploration range; Generate an exploration distance adjustment interface that can adjust the exploration distance in response to the user's operation to generate an exploration distance instruction; Generate a parking site adjustment interface that can adjust the parking site in response to the user's operation to generate a parking site exploration instruction.

2. The parking method with autonomous exploration mode according to claim 1, characterized in that, Between step S1100 and step S1400, the parking method further comprises: S1200: The terminal parking module outputs a first voice signal for interaction; and S1300: The terminal parking module determines whether a first characteristic voice signal is received in response to the first voice signal.

3. The parking method with autonomous exploration mode according to claim 2, characterized in that, In step S1200, it further comprises: The vehicle-mounted parking module acquires the current position of the vehicle and determines whether there is vehicle historical trajectory information associated with the current position of the vehicle and including at least one historical parking target position.

4. The parking method with autonomous exploration mode according to claim 3, characterized in that, The vehicle-mounted parking module acquires the current position of the vehicle and determines whether there is vehicle historical trajectory information associated with the current position of the vehicle and including at least one historical parking target position, and the determination result is that there is no vehicle historical trajectory information associated with the current position of the vehicle and including at least one historical parking target position, then step S1400 comprises: S1402a: The terminal parking module outputs an operation interface, and the terminal parking module receives the user's operation based on the operation interface; and S1404a: The terminal parking module generates a parking exploration instruction including an exploration range instruction, an exploration distance instruction and a parking site exploration instruction in response to the received user's operation for the user to confirm.

5. The parking method with autonomous exploration mode according to claim 4, characterized in that, Step S1400 further comprises: S1406a: The terminal parking module receives the confirmation instruction, and the terminal parking module sends the parking exploration instruction to the vehicle-mounted parking module to execute the parking control process in the autonomous exploration mode.

6. The parking method with autonomous exploration mode according to claim 4, characterized in that, Step S1400 further includes: S1406b: The terminal parking module does not receive the confirmation instruction or receives the denial instruction, and then the control interface is output again to receive the user's control to execute step S1404a again.

7. The parking method with autonomous exploration mode according to claim 3, characterized in that, The vehicle-mounted parking module acquires the current position of the vehicle and judges whether there is historical trajectory information of the vehicle associated with the current position of the vehicle and including at least one historical parking target position, and the result of the judgment is that there is historical trajectory information of the vehicle associated with the current position of the vehicle and including at least one historical parking target position, then step S1400 includes: S1402c: The vehicle-mounted parking module outputs the historical trajectory information of the vehicle associated with the current position of the vehicle and including at least one historical parking target position to the terminal parking module for selection.

8. The parking method with autonomous exploration mode according to claim 7, characterized in that, Step S1400 further includes: S1404c: The terminal parking module receives the historical parking target position selection instruction, and the vehicle-mounted parking module executes the parking control process based on the historical parking target position selection instruction.

9. The parking method with autonomous exploration mode according to claim 7, characterized in that, Step S1400 further includes: S1404d: The terminal parking module does not receive the historical parking target position selection instruction, and then the terminal parking module outputs a control interface, and the control interface is used to receive the user's control; and S1406d: The terminal parking module generates a parking exploration instruction including an exploration range instruction, an exploration distance instruction, and a parking spot exploration instruction in response to the user's control received by the control interface, for user confirmation.

10. The parking method with autonomous exploration mode according to claim 9, characterized in that, Step S1400 further includes: S1408d: The terminal parking module receives the confirmation instruction, and the terminal parking module sends the parking exploration instruction to the vehicle-mounted parking module to execute the parking control process in the autonomous exploration mode.

11. The parking method with autonomous exploration mode according to claim 9, characterized in that, Step S1400 further includes: S1408e: The terminal parking module does not receive the confirmation instruction or receives the denial instruction, and then the control interface is output again to execute step S1406d again.

12. The parking method with autonomous exploration mode according to any one of claims 1 to 11, characterized in that, The parking control process in the autonomous exploration mode based on the parking exploration instruction includes: The vehicle-mounted parking module judges whether the vehicle meets the starting condition, and generates a control signal to control the vehicle to start after meeting the starting condition; The vehicle-mounted parking module acquires the feasible domain of the current road based on the exploration range and the exploration distance limited by the exploration range instruction and the exploration distance instruction, performs local path planning in real time according to the road conditions, and generates a control signal to control the vehicle to travel along the feasible domain; and The vehicle-mounted parking module explores the parking spot, and then executes the parking process based on the explored parking spot.

13. The parking method with autonomous exploration mode according to claim 12, characterized in that, Further includes: The vehicle-mounted parking module acquires the fork information, and then performs local path planning based on the fork information to update the feasible domain.

14. The parking method with autonomous exploration mode according to claim 13, characterized in that, The local path planning based on the fork information to update the feasible domain includes: A two-dimensional grid map is constructed for the fork; Based on each edge line of the intersection, all grids covering the edge line are assigned an initial value, and the grids adjacent to all the assigned grids in the feasible region in the two-dimensional grid map are assigned, the edge grids are assigned a first value, the corner grids are assigned a second value, if a grid is adjacent to multiple edge grids or corner grids, the grid is assigned the minimum value of the first value or the second value, until all the grids in the feasible region are assigned, and the grid assignment based on each edge line of the intersection is completed; The grid assignment results based on each edge line are superimposed based on the grid map assignment, and an initial superimposed assignment map is obtained; The center point of the grid with the minimum superimposed assignment in the initial superimposed assignment map is taken as the center point of the intersection, and if there are multiple grids with the minimum superimposed assignment, the geometric center of the grids is taken as the center point of the intersection; Based on the intersection center point, the intersection passing region and each sub-path are obtained; and The feasible region is updated based on the intersection passing region and each sub-path.

15. The parking method with autonomous exploration mode according to any one of claims 1 to 11, characterized in that, Based on the historical parking target position selection instruction, a parking control process is executed, including: S1441: comparing the current position of the vehicle with at least one historical trajectory of the vehicle containing the selected historical parking target position, and obtaining an optimal historical trajectory based on the number of times the vehicle travels along the historical trajectory and the distance between the current position of the vehicle and the historical trajectory of the vehicle; S1442: determining whether the minimum distance between the optimal historical trajectory and the current position of the vehicle exceeds a preset distance, if yes, executing step S1443, and if no, executing step S1444; S1443: the vehicle-mounted parking module controls the vehicle to move forward along the current feasible region, and continuously marks the historical trajectory point closest to the current position of the vehicle to find the optimal historical trajectory, and whenever the vehicle encounters an intersection, the vehicle-mounted parking module controls the vehicle to turn into the branch pointed by the direction of the line connecting the marked point, until the minimum distance between the current position of the vehicle and the optimal historical trajectory is less than or equal to the preset distance, and step S1444 is entered; S1444: the vehicle-mounted parking module controls the vehicle to reposition to control the vehicle to move to the optimal historical trajectory and align the historical pose; and S1445: an optimal global route is planned from the optimal historical trajectory, and the vehicle is controlled to travel to the target location based on the optimal global route.

16. The parking method with autonomous exploration mode according to any one of claims 1 to 11, characterized in that, Step S1400 further includes the following parking processing process: S1431: the vehicle-mounted parking module explores to a parking site; S1432: if the parking site is a roadside parking area, step S1433 is executed, if the parking site is a parking lot, step S1434 is executed, and if the parking site is an empty land / square, step S1435 is executed; S1433: the roadside parking area is identified, and the vehicle-mounted parking module generates a control signal to control the vehicle to enter the roadside parking area; S1434: the parking lot is identified, and then the parking area is identified, and then the parking space number is identified, and then the vehicle-mounted parking module generates a control signal to control the vehicle to enter the parking space; S1435: identifying an empty lot / square, and the vehicle-mounted parking module generates a control signal to control the vehicle to enter the empty lot / square; and S1436: when the step S1433 / S1434 / S1435 is successfully executed, the vehicle-mounted parking module generates a control signal to control the vehicle to park, and generates a parking success information and sends it to the user terminal device; when the step S1433 / S1434 / S1435 fails, the vehicle-mounted parking module generates a parking failure information and sends it to the mobile device.

17. A parking system with an autonomous exploration mode, characterized in that Comprising: a terminal parking module, which generates and sends a start control signal; and a vehicle-mounted parking module, which receives the start control signal and starts; wherein the terminal parking module comprises: a parking exploration instruction generation module, which generates an exploration range instruction, an exploration distance instruction and a parking site exploration instruction in response to the user's operation; the vehicle-mounted parking module comprises: a historical trajectory output module, which outputs the historical trajectory information of the vehicle associated with the current position of the vehicle and including at least one historical parking target position; a parking processing module, which executes a parking control process based on the parking exploration instruction generated by the terminal parking module or the selected historical parking target position and its corresponding historical trajectory information of the vehicle; wherein the parking processing module of the vehicle-mounted parking module limits the exploration range and exploration distance based on the exploration range instruction and the exploration distance instruction to explore the parking site; the parking exploration instruction is generated by the following steps: generating a four-quadrant interface, each quadrant area of the four-quadrant interface can be selected in response to the user's operation to generate an exploration range instruction, and the exploration range instruction is an exploration range instruction with one or more than two quadrant areas as the exploration range; generating an exploration distance adjustment interface, which can adjust the exploration distance in response to the user's operation to generate an exploration distance instruction; generating a parking site adjustment interface, which can adjust the parking site in response to the user's operation to generate a parking site exploration instruction.

18. The parking system with autonomous exploration mode according to claim 17, characterized in that, The terminal parking module further comprises: a voice signal output module, which outputs a first voice signal for interaction; a voice signal receiving module, which is used to receive the user's voice signal; a first feature voice signal acquisition module, which judges whether a first feature voice signal is received in response to the first voice signal.

19. The parking system with autonomous exploration mode according to claim 17, characterized in that, The terminal parking module further comprises: a touch instruction receiving module, which is used to receive at least a historical parking target position selection instruction.

20. An electronic device, comprising: Comprising: a memory, which stores an execution instruction; and a processor, which executes the execution instruction stored in the memory, so that the processor executes the parking method of any one of claims 1-16.

21. A readable storage medium, characterized by, The readable storage medium stores execution instructions, and the execution instructions are used for realizing the parking method in any one of claims 1 to 16 when executed by the processor.

22. A computer program product comprising a computer program, characterized in that, The computer program realizes the parking method in any one of claims 1 to 16 when executed by the processor.

Images