Vehicle control method, device and equipment based on shared map and storage medium

Abstract

The invention discloses a vehicle control method based on a shared map. The method comprises the steps of obtaining current position information of a vehicle and current environment information of an environment where the vehicle is located; generating a parking map acquisition instruction based on the current position information; in response to the parking map acquisition instruction, acquiring initial parking lot map data corresponding to the coordinate information of the target parking lot from a shared map; performing feature matching processing based on the current environment information and the initial parking lot map data to obtain target parking lot map data; target parking position information is determined based on the target parking lot map data, path planning processing is performed based on the current position information, the target parking lot map data and the target parking position information, and a target parking path is determined; and controlling the vehicle to move to a parking space corresponding to the target parking position information according to the target parking path. The current vehicle obtains the map data of the current parking lot environment through the shared map, and the problem of improving the accuracy of the parking lot map in intelligent driving is solved.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular to a vehicle control method, device, equipment and storage medium based on a shared map. Background Art

[0002] In the field of autonomous driving, Automated Valet Parking (AVP) is a common scenario in the application of driverless technology. The basic requirement of the parking function is that the user gives instructions to the vehicle at a designated location, and the vehicle can respond to the user's instructions and automatically drive to the parking space in the parking lot or from the parking space in the parking lot to the designated location. The key to realizing AVP is to build a parking map for positioning, navigation and vehicle location display.

[0003] In the prior art, parking lot data maps for AVP mainly rely on map suppliers to use special parking lot map collection vehicles to collect parking lot data in key cities and key areas across the country, and provide them to OEMs (Original Equipment Manufacturers) in the form of data packets. However, the map suppliers have low collection efficiency, high collection costs, and low map freshness, resulting in most cities' parking lots not yet being fully covered by high-precision maps. Therefore, how to improve the accuracy of parking lot maps has become an urgent problem to be solved. Summary of the invention

[0004] In order to solve the problems of the prior art, the present application provides a vehicle control method, device, equipment and storage medium based on a shared map. A shared map is constructed by constructing environmental information of the parking environment collected by historical vehicles. The current vehicle obtains map data of the current parking environment through the shared map, so that the current vehicle plans a parking path based on the map data of the current parking lot, moves to the parking space, and achieves efficient parking, thereby solving the problem of improving the accuracy of parking lot maps in intelligent driving.

[0005] According to one aspect of the present application, a vehicle control method based on a shared map is provided, the method comprising:

[0006] Obtaining current location information of the vehicle and current environment information of the environment in which the vehicle is located;

[0007] generating a parking map acquisition instruction based on the current position information; the parking map acquisition instruction includes coordinate information of a target parking lot that meets a preset parking condition;

[0008] In response to the parking map acquisition instruction, initial parking lot map data corresponding to the coordinate information of the target parking lot is acquired from a shared map; the shared map is map data of a pre-constructed parking lot environment stored in a cloud server;

[0009] Perform feature matching processing based on the current environment information and the initial parking lot map data to obtain target parking lot map data;

[0010] Determining target parking location information based on the target parking lot map data, performing path planning processing based on the current location information, the target parking lot map data and the target parking location information to determine a target parking path;

[0011] The vehicle is controlled to move to a parking space corresponding to the target parking position information according to the target parking path.

[0012] In a possible implementation, the acquiring the current position information of the vehicle and the current environment information of the environment in which the vehicle is located includes:

[0013] In response to a parking instruction issued by an object, determining position information of the vehicle based on a positioning system of the vehicle, and determining the position information as the current position information;

[0014] After determining the current position information, the image acquisition device of the vehicle is controlled to acquire images of the environment around the vehicle, and the acquired image data is used as the current environment information.

[0015] In a possible implementation, the current location information includes the coordinates of the vehicle.

[0016] The generating of the parking map acquisition instruction based on the current position information includes:

[0017] Based on the coordinates of the vehicle and the positioning system of the vehicle, searching for a plurality of parking lots to be selected whose distances from the vehicle meet a preset distance condition;

[0018] Obtain the number of available parking spaces corresponding to each of the plurality of parking lots to be selected;

[0019] Determine a candidate parking lot with a number of vacant parking spaces greater than a preset number as a target parking lot;

[0020] Acquire the coordinates of the target parking lot, and determine the coordinates of the target parking lot as the target parking lot information;

[0021] The parking map acquisition instruction is generated based on the target parking lot information.

[0022] In a possible implementation, the acquiring, from a shared map, initial parking lot map data corresponding to the target parking lot information includes:

[0023] Based on the coordinates of the target parking lot, map data corresponding to the coordinates of the target parking lot is acquired from the shared map, and the map data corresponding to the coordinates of the target parking lot is determined as initial parking lot map data.

[0024] In a possible implementation, performing feature matching processing based on the current environment information and the initial parking lot map data to obtain target parking lot map data includes:

[0025] Extracting features from the current environment information to obtain current environment features, and performing matching processing on the initial parking lot map data based on the current environment features to obtain matching parking lot map data;

[0026] Performing three-dimensional modeling based on the matched parking lot map data to obtain three-dimensional parking lot map data to display the floors, entrances and exits, and parking spaces of the target parking lot;

[0027] The three-dimensional parking lot map data is determined as the target parking lot map data.

[0028] In a possible implementation, extracting features from the current environment information to obtain current environment features, and performing matching processing on the initial parking lot map data based on the current environment features to obtain matching parking lot map data includes:

[0029] Performing semantic segmentation processing on the current environment information to obtain semantic features; the semantic features are used to characterize the location information of lane lines, gates, speed bumps, signboards and parking spaces in the environment where the vehicle is located;

[0030] Performing coordinate calculation based on the semantic features and the coordinates of the vehicle to obtain a coordinate calculation result;

[0031] The initial parking lot map data is searched based on the coordinate calculation result to obtain the matching parking lot map data.

[0032] In a possible implementation, the method for constructing the shared map includes:

[0033] In the case where the historical vehicle is located in a historical parking environment, an image acquisition device based on the historical vehicle acquires images of the environment around the historical vehicle to obtain historical parking image data;

[0034] Uploading the historical parking lot image data and the current location information of the historical vehicle to a cloud server, wherein the current location information includes the coordinates of the historical vehicle;

[0035] Performing semantic segmentation on the historical parking lot image data based on a cloud server to obtain historical parking lot image features;

[0036] Perform feature matching processing based on the current position information of the historical vehicle, the image features of the historical parking lot, and the positioning system to generate historical parking lot fragment map data;

[0037] fusing the historical parking lot fragment map data corresponding to the historical parking lot to obtain historical parking lot map data;

[0038] The shared map is constructed based on the historical parking lot map data.

[0039] In a possible implementation, after controlling the vehicle to move to the parking space corresponding to the target parking position information, the method includes:

[0040] Uploading the current location information of the vehicle and the current environment information to a cloud server;

[0041] Performing semantic segmentation on the current environment information based on the cloud server to obtain environment image features;

[0042] Perform feature matching processing based on the current location information, the environmental image features and the positioning system to generate environmental fragment map data;

[0043] Based on the environmental fragment map data, the map data corresponding to the current location information in the shared map is updated.

[0044] In another aspect, a vehicle control device based on a shared map is provided, the device comprising:

[0045] A current information acquisition module is used to acquire the current position information of the vehicle and the current environment information of the vehicle;

[0046] An instruction generation module, configured to generate a parking map acquisition instruction based on the current position information; the parking map acquisition instruction includes coordinate information of a target parking lot that meets a preset parking condition;

[0047] an initial map data acquisition module, configured to acquire initial parking lot map data corresponding to the coordinate information of the target parking lot from a shared map in response to the parking map acquisition instruction; the shared map is map data of a pre-constructed parking lot environment stored in a cloud server;

[0048] A target map data acquisition module, configured to perform feature matching processing based on the current environment information and the initial parking lot map data to obtain target parking lot map data;

[0049] a target parking path determination module, configured to determine target parking position information based on the target parking lot map data, and to perform path planning processing based on the current position information, the target parking lot map data, and the target parking position information to determine the target parking path;

[0050] The parking control module is used to control the vehicle to move to the parking space corresponding to the target parking position information according to the target parking path.

[0051] In a possible implementation, the current information acquisition module is used to:

[0052] In response to a parking instruction issued by an object, determining position information of the vehicle based on a positioning system of the vehicle, and determining the position information as the current position information;

[0053] After determining the current position information, the image acquisition device of the vehicle is controlled to acquire images of the environment around the vehicle, and the acquired image data is used as the current environment information.

[0054] In a possible implementation, the current position information includes the coordinates of the vehicle, and the instruction generating module is used to:

[0055] Based on the coordinates of the vehicle and the positioning system of the vehicle, searching for a plurality of parking lots to be selected whose distances from the vehicle meet a preset distance condition;

[0056] Obtain the number of available parking spaces corresponding to each of the plurality of parking lots to be selected;

[0057] Determine a candidate parking lot with a number of vacant parking spaces greater than a preset number as a target parking lot;

[0058] Acquire the coordinates of the target parking lot, and determine the coordinates of the target parking lot as the target parking lot information;

[0059] The parking map acquisition instruction is generated based on the target parking lot information.

[0060] In a possible implementation, the initial map data acquisition module is used to:

[0061] Based on the coordinates of the target parking lot, map data corresponding to the coordinates of the target parking lot is acquired from the shared map, and the map data corresponding to the coordinates of the target parking lot is determined as initial parking lot map data.

[0062] In a possible implementation, the target map data acquisition module is used to:

[0063] Extracting features from the current environment information to obtain current environment features, and performing matching processing on the initial parking lot map data based on the current environment features to obtain matching parking lot map data;

[0064] Performing three-dimensional modeling based on the matched parking lot map data to obtain three-dimensional parking lot map data to display the floors, entrances and exits, and parking spaces of the target parking lot;

[0065] The three-dimensional parking lot map data is determined as the target parking lot map data.

[0066] In a possible implementation, the target map data acquisition module includes a matching unit, and the matching unit is used to:

[0067] Performing semantic segmentation processing on the current environment information to obtain semantic features; the semantic features are used to characterize the location information of lane lines, gates, speed bumps, signboards and parking spaces in the environment where the vehicle is located;

[0068] Performing coordinate calculation based on the semantic features and the coordinates of the vehicle to obtain a coordinate calculation result;

[0069] The initial parking lot map data is searched based on the coordinate calculation result to obtain the matching parking lot map data.

[0070] In a possible implementation, the device further includes a shared map construction module, and the shared map construction module is used to:

[0071] In the case where the historical vehicle is located in a historical parking environment, an image acquisition device based on the historical vehicle acquires images of the environment around the historical vehicle to obtain historical parking image data;

[0072] Uploading the historical parking lot image data and the current location information of the historical vehicle to a cloud server, wherein the current location information includes the coordinates of the historical vehicle;

[0073] Performing semantic segmentation on the historical parking lot image data based on a cloud server to obtain historical parking lot image features;

[0074] Perform feature matching processing based on the current position information of the historical vehicle, the image features of the historical parking lot, and the positioning system to generate historical parking lot fragment map data;

[0075] fusing the historical parking lot fragment map data corresponding to the historical parking lot to obtain historical parking lot map data;

[0076] The shared map is constructed based on the historical parking lot map data.

[0077] In a possible implementation, the device further includes an updating module.

[0078] After controlling the vehicle to move to the parking space corresponding to the target parking position information, the updating module is used to:

[0079] Uploading the current location information of the vehicle and the current environment information to a cloud server;

[0080] Performing semantic segmentation on the current environment information based on the cloud server to obtain environment image features;

[0081] Perform feature matching processing based on the current location information, the environmental image features and the positioning system to generate environmental fragment map data;

[0082] Based on the environmental fragment map data, the map data corresponding to the current location information in the shared map is updated.

[0083] On the other hand, an electronic device is provided, comprising a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement any of the above-mentioned aspects of the vehicle control method based on a shared map.

[0084] On the other hand, a computer-readable storage medium is provided, in which at least one instruction or at least one program is stored, and the at least one instruction or the at least one program is loaded and executed by a processor to implement a vehicle control method based on a shared map as described in any of the above aspects.

[0085] On the other hand, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device performs any of the above-mentioned vehicle control methods based on a shared map.

[0086] The embodiment of the present application obtains the current position information of the vehicle and the current environment information of the environment in which the vehicle is located; generates a parking map acquisition instruction based on the current position information; the parking map acquisition instruction includes the coordinate information of the target parking lot that meets the preset parking conditions; in response to the parking map acquisition instruction, obtains the initial parking lot map data corresponding to the coordinate information of the target parking lot from the shared map; the shared map is the map data of the pre-constructed parking environment stored in the cloud server; performs feature matching processing based on the current environment information and the initial parking lot map data to obtain the target parking lot map data; determines the target parking position information based on the target parking lot map data, performs path planning processing based on the current position information, the target parking lot map data and the target parking position information to determine the target parking path; controls the vehicle to move to the parking space corresponding to the target parking position information according to the target parking path. The shared map is constructed by the environmental information of the parking lot environment collected by the historical vehicles, and the current vehicle obtains the map data of the current parking lot environment through the shared map, so that the current vehicle plans the parking path based on the map data of the current parking lot, moves to the parking space, and achieves efficient parking, thereby solving the problem of improving the accuracy of the parking lot map in intelligent driving. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0088] Figure 1 is a flow chart of a vehicle control method based on a shared map provided in an embodiment of the present application;

[0089] Figure 2 It is a structural block diagram of a vehicle-side system provided in an embodiment of the present application;

[0090] Figure 3 It is a flowchart of a method for generating a parking map acquisition instruction provided in an embodiment of the present application;

[0091] Figure 4 It is a flowchart of a method for constructing a shared map provided in an embodiment of the present application;

[0092] Figure 5 It is a structural block diagram of a cloud server provided in an embodiment of the present application;

[0093] Figure 6 It is a flowchart of a method for confirming target parking lot map data provided by an embodiment of the present application;

[0094] Figure 7 It is an interactive schematic diagram provided in an embodiment of the present application;

[0095] Figure 8 This is a structural block diagram of the APP mobile terminal provided in the embodiment of the present application;

[0096] Fig. 9 It is a structural block diagram of a vehicle control device based on a shared map provided in an embodiment of the present application. DETAILED DESCRIPTION

[0097] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0098] With the development of intelligent connected vehicles, intelligent assisted driving on highways and in urban areas has developed rapidly, and automated valet parking (AVP) to solve the last mile of intelligent driving is also accelerating. The typical AVP function complete process is: (1) When the user arrives at the entrance or near the parking lot, he can leave the vehicle after turning on the AVP function through the mobile phone or the vehicle; (2) The vehicle receives the AVP function activation command, enters the parking lot by itself, finds an available parking space in the parking lot and parks, and then sends the vehicle location to the user; (3) When the user needs to use the car, he can find the car by himself according to the vehicle location, or send a vehicle call through the mobile phone to specify the location where he needs to get on the car; (4) After receiving the call command, the vehicle leaves the parking space by itself, arrives at the designated location, and the user gets on the car. Building a parking map for positioning, navigation and vehicle location display is the key to realizing AVP. At present, the parking data map used for AVP in the industry mainly relies on map suppliers to use special parking map collection vehicles to collect parking data in key cities and key areas across the country, and provide it to OEMs (Original Equipment Manufacturers) in the form of data packets.

[0099] The current construction of AVP parking lot maps mainly relies on map vendors using special equipment to collect data. However, the map vendors' collection efficiency is low, the collection cost is high, and the map freshness is low. As a result, parking lots in most cities have not yet achieved full coverage of high-precision maps. AVP can only support parking lots in key places in some cities.

[0100] The embodiment of the present application provides a vehicle control method based on a shared map. Through crowdsourcing maps, a large amount of parking map data is obtained at a relatively low cost. The data is integrated through a cloud service platform to construct a parking map with high coverage and freshness. The map is distributed to the vehicle side for positioning navigation and parking lot selection, and to the mobile phone side for vehicle running trajectory and position display, thereby providing an AVP map system solution. Figure 1 The vehicle control method based on the shared map includes steps S101 to S1011.

[0101] In step S101, the current position information of the vehicle and the current environment information of the environment where the vehicle is located are obtained.

[0102] In a possible implementation, the acquiring the current position information of the vehicle and the current environment information of the environment in which the vehicle is located includes:

[0103] In response to a parking instruction issued by an object, determining position information of the vehicle based on a positioning system of the vehicle, and determining the position information as the current position information;

[0104] After determining the current position information, the image acquisition device of the vehicle is controlled to acquire images of the environment around the vehicle, and the acquired image data is used as the current environment information.

[0105] In a possible implementation, the vehicle includes a vehicle-side system, such as Figure 2 As shown, the vehicle-side system includes a collection unit, a communication unit and a map display unit.

[0106] In one possible implementation, the acquisition unit includes an AI processing chip, a front camera, an IMU (Inertial measurement unit) and a GNSS (Global Navigation Satellite System).

[0107] In one possible implementation, the communication unit includes a vehicle-mounted TBOX (Telematics BOX, telematics processor) configured with a 4G / 5G module for uploading map data and obtaining the latest map data from a server.

[0108] In a possible implementation, the map display unit includes a vehicle-mounted TBOX and a display screen.

[0109] In a possible implementation, the object issues the parking instruction through a pre-developed mobile APP terminal. The mobile APP terminal includes functions such as user information registration, modification, and cancellation.

[0110] In a possible implementation, in response to a parking instruction issued by an object, the position information of the vehicle is determined based on the GNSS of the vehicle.

[0111] In a possible implementation, the image acquisition device of the vehicle may be at least one of a radar and a camera, and at least one of the radar or the camera is controlled to detect the environment around the vehicle, and the data obtained by the detection is used as the current environmental information.

[0112] In a possible implementation, the image acquisition device of the vehicle is a front camera. After determining the current position information, the front camera is controlled to shoot the environment around the vehicle, and the captured video data is used as the current environment information.

[0113] In this implementation, the current position information is determined based on the positioning system of the vehicle, and the vehicle is controlled to collect current environmental information. At the same time, the vehicle's position information and surrounding environmental information are collected, thereby improving the accuracy of vehicle positioning.

[0114] In step S103, a parking map acquisition instruction is generated based on the current position information; the parking map acquisition instruction includes coordinate information of a target parking lot that meets a preset parking condition.

[0115] In a possible implementation, the current location information includes the coordinates of the vehicle.

[0116] The generating of the parking map acquisition instruction based on the current position information includes:

[0117] Based on the coordinates of the vehicle and the positioning system of the vehicle, searching for a plurality of parking lots to be selected whose distances from the vehicle meet a preset distance condition;

[0118] Obtain the number of available parking spaces corresponding to each of the plurality of parking lots to be selected;

[0119] Determine a candidate parking lot with a number of vacant parking spaces greater than a preset number as a target parking lot;

[0120] Acquire the coordinates of the target parking lot, and determine the coordinates of the target parking lot as the target parking lot information;

[0121] The parking map acquisition instruction is generated based on the target parking lot information.

[0122] In one possible implementation, Figure 3 As shown, the method for generating the parking map acquisition instruction includes steps S301 to S309.

[0123] In step S301, based on the coordinates of the vehicle and the positioning system of the vehicle, a plurality of parking lots to be selected whose distances from the vehicle meet a preset distance condition are searched.

[0124] In a possible implementation, the current location information includes coordinates of the vehicle, and the coordinates of the vehicle are obtained based on GNSS detection.

[0125] In a possible implementation, based on the coordinates of the vehicle and the GNSS system of the vehicle, a parking lot near the vehicle is searched, specifically, a parking lot within a distance of 3 kilometers near the vehicle is searched.

[0126] In step S303, the number of available parking spaces corresponding to each of the plurality of parking lots to be selected is obtained.

[0127] In one possible implementation, the display unit of the mobile APP can display the nearest available parking lot. The mobile APP sends instructions to the server, and the server obtains the number of vacant parking spaces corresponding to each of the multiple parking lots to be selected. The mobile APP can display the time and location of the last vacant parking space scan.

[0128] In step S305, a to-be-selected parking lot having a number of available parking spaces greater than a preset number is determined as a target parking lot.

[0129] In a possible implementation, in order to improve the convenience of parking, the candidate parking lot with the largest number of available parking spaces is preferentially selected as the target parking lot.

[0130] In a possible implementation, the target parking lot can also be determined by manually operating the mobile APP.

[0131] In step S307, the coordinates of the target parking lot are acquired, and the coordinates of the target parking lot are determined as the target parking lot information.

[0132] In a possible implementation manner, after the target parking lot is determined, the coordinates of the target parking lot are acquired through a positioning system.

[0133] In a possible implementation, the coordinates of the target parking lot may also be acquired through historical record information of the target parking lot on a cloud server.

[0134] In step S309, the parking map acquisition instruction is generated based on the target parking lot information.

[0135] In a possible implementation, the parking map acquisition instruction is used to acquire map data of the target parking lot, so that the vehicle can park based on the map data of the target parking lot.

[0136] In this implementation, based on the coordinates of the vehicle and the positioning system of the vehicle, a target parking lot is searched for, and a parking map acquisition instruction is generated based on the target parking lot information, so that an available parking lot near the vehicle can be quickly located, and a parking map acquisition instruction is generated based on the parking lot information, so that map data can be quickly acquired according to the instruction later.

[0137] In step S105 , in response to the parking map acquisition instruction, initial parking lot map data corresponding to the coordinate information of the target parking lot is acquired from a shared map; the shared map is map data of a pre-constructed parking lot environment stored in a cloud server.

[0138] In a possible implementation, the acquiring, from a shared map, initial parking lot map data corresponding to the target parking lot information includes:

[0139] Based on the coordinates of the target parking lot, map data corresponding to the coordinates of the target parking lot is acquired from the shared map, and the map data corresponding to the coordinates of the target parking lot is determined as initial parking lot map data.

[0140] In a possible implementation manner, the shared map includes coordinates, and based on the coordinates of the target parking lot, map data corresponding to the coordinates of the target parking lot is acquired from the shared map.

[0141] In a possible implementation, the map data corresponding to the coordinates of the target parking lot includes map data of all coordinates whose distances from the coordinates of the target parking lot meet a preset distance range.

[0142] In a possible implementation manner, the map data corresponding to the coordinates of the target parking lot is map data including the coordinates of the vehicle and the range where the coordinates of the target parking lot are located.

[0143] In this implementation, based on the coordinates of the target parking lot, map data corresponding to the coordinates of the target parking lot is obtained from the shared map. Since the shared map is fresh and has a wide coverage, the map data of the target parking lot obtained from the shared map is highly accurate, thereby improving the convenience and safety of parking.

[0144] In a possible implementation, the method for constructing the shared map includes:

[0145] In the case where the historical vehicle is located in a historical parking environment, an image acquisition device based on the historical vehicle acquires images of the environment around the historical vehicle to obtain historical parking image data;

[0146] Uploading the historical parking lot image data and the current location information of the historical vehicle to a cloud server, wherein the current location information includes the coordinates of the historical vehicle;

[0147] Performing semantic segmentation on the historical parking lot image data based on a cloud server to obtain historical parking lot image features;

[0148] Perform feature matching processing based on the current position information of the historical vehicle, the image features of the historical parking lot, and the positioning system to generate historical parking lot fragment map data;

[0149] fusing the historical parking lot fragment map data corresponding to the historical parking lot to obtain historical parking lot map data;

[0150] The shared map is constructed based on the historical parking lot map data.

[0151] In one possible implementation, Figure 4 As shown, the method for constructing a shared map includes steps S401 to S4011.

[0152] In step S401, when a historical vehicle is located in a historical parking environment, an image acquisition device based on the historical vehicle acquires images of the environment around the historical vehicle to obtain historical parking image data.

[0153] In a possible implementation, the historical vehicle is a vehicle that was located in a parking lot environment at a historical time, and the historical vehicle also includes a collection unit, a communication unit, and a map display unit.

[0154] In a possible implementation, there are multiple historical vehicles, there are multiple historical parking lot environments, and the historical parking lot image data is data collected from different vehicles in different parking lots at multiple different historical times.

[0155] In step S403, the historical parking lot image data and the current location information of the historical vehicle are uploaded to a cloud server, wherein the current location information includes the coordinates of the historical vehicle.

[0156] In a possible implementation, the image acquisition device of the historical vehicle may be at least one of a radar and a camera, which controls at least one of the radar or the camera to detect the environment around the vehicle, and uses the detected data as the historical parking lot image data.

[0157] In a possible implementation, the image acquisition device of the historical vehicle is a front camera. After determining the current position information of the historical vehicle, the front camera is controlled to shoot the environment around the vehicle, and the captured video data is used as the historical parking lot image data.

[0158] In step S405, semantic segmentation is performed on the historical parking lot image data based on the cloud server to obtain historical parking lot image features.

[0159] In a possible implementation, an AI processing device of a visual language model similar to LVMs (Large Visual Models) carried on a vehicle is used to perform semantic segmentation and semantic recognition on the historical parking lot image data collected by the camera to obtain the historical parking lot image features, which represent the elements of the historical parking lot environment. Specifically, the elements of the historical parking lot environment include key features such as gates, lane lines, parking space lines, pillars, road signs, speed bumps, and ground direction guidance signs.

[0160] In step S407, feature matching processing is performed based on the current position information of the historical vehicle, the image features of the historical parking lot, and the positioning system to generate historical parking lot fragment map data.

[0161] In one possible implementation, based on the body coordinates of the historical vehicle and utilizing the camera ranging principle, the local coordinates of the elements of the historical parking environment are calculated, and a VIO (visual-inertial) fusion algorithm is performed through vision combined with IMU (Inertial Measurement Unit), and the vehicle-mounted GNSS is used to perform vehicle positioning. Combined with the local coordinates of the elements of the historical parking environment, historical parking lot fragment map data is generated.

[0162] In one possible implementation, a VIO fusion algorithm is used to estimate the pose and position of the vehicle.

[0163] In one possible implementation, target tracking is performed on key road features in the historical parking lot fragment map data in combination with the vehicle posture, the historical parking lot fragment map data is optimized based on the target tracking result, and the optimized historical parking lot fragment map data is used as the historical parking lot fragment map data.

[0164] In step S409, the historical parking lot fragment map data corresponding to the historical parking lot is fused to obtain historical parking lot map data.

[0165] In one possible implementation, the historical parking lot image data is clustered and fused to obtain location information and feature information of the historical parking lot image data, and multiple historical parking lot fragment map data are matched and classified using these location and feature information. By performing data fitting on multiple historical parking lot fragment map data at the same location, more accurate historical parking lot map data is obtained.

[0166] In step S4011, the shared map is constructed based on the historical parking map data.

[0167] In a possible implementation, data splicing is performed based on all historical parking lot map data, and related historical parking lot map data are associated to construct a complete parking lot map.

[0168] In a possible implementation, all parking lot maps are spliced ​​to obtain the shared map. As the amount of data increases, the area coverage and accuracy of the parking lots in the shared map also increase.

[0169] In one possible implementation, the map data of the parking lot in the shared map is lightweighted based on a cloud server, and different types of parking lot map data, such as semantic maps and topological maps, can be constructed. When the parking lot information is called on the vehicle side, it can be downloaded with lower traffic.

[0170] In one possible implementation, Figure 5 As shown, the cloud server includes a data clustering and fusion module, a data splicing module and a data compression and distribution module. Specifically, the cloud server is used to construct the shared map. The data clustering and fusion module is used to perform semantic segmentation on the historical parking lot image data to obtain historical parking lot image features; perform feature matching processing based on the current position information of the historical vehicle, the historical parking lot image features and the positioning system to generate historical parking lot fragment map data. The data splicing module is used to perform fusion processing on the historical parking lot fragment map data corresponding to the historical parking lot to obtain historical parking lot map data; and construct the shared map based on the historical parking lot map data. The data compression and distribution module is used to perform lightweight processing on the map data of the parking lot in the shared map.

[0171] In this implementation, parking lot data is collected on the vehicle side, and a shared map is built on the cloud server based on the collected parking lot data. This can quickly collect map data of various parking lots in the city and build a map for driving, reducing the cost of parking lot map data collection and maintenance, and keeping the map fresh, saving parking lot map data collection time and accelerating the implementation of AVP.

[0172] In step S107, feature matching processing is performed based on the current environment information and the initial parking lot map data to obtain target parking lot map data.

[0173] In a possible implementation, performing feature matching processing based on the current environment information and the initial parking lot map data to obtain target parking lot map data includes:

[0174] Extracting features from the current environment information to obtain current environment features, and performing matching processing on the initial parking lot map data based on the current environment features to obtain matching parking lot map data;

[0175] Performing three-dimensional modeling based on the matched parking lot map data to obtain three-dimensional parking lot map data to display the floors, entrances and exits, and parking spaces of the target parking lot;

[0176] The three-dimensional parking lot map data is determined as the target parking lot map data.

[0177] In one possible implementation, Figure 6 As shown, the method for determining the target parking lot map data includes steps S601 to S605.

[0178] In step S601, feature extraction is performed on the current environment information to obtain current environment features, and matching processing is performed on the initial parking lot map data based on the current environment features to obtain matching parking lot map data.

[0179] In a possible implementation, extracting features from the current environment information to obtain current environment features, and performing matching processing on the initial parking lot map data based on the current environment features to obtain matching parking lot map data includes:

[0180] Performing semantic segmentation processing on the current environment information to obtain semantic features; the semantic features are used to characterize the location information of lane lines, gates, speed bumps, signboards and parking spaces in the environment where the vehicle is located;

[0181] Performing coordinate calculation based on the semantic features and the coordinates of the vehicle to obtain a coordinate calculation result;

[0182] The initial parking lot map data is searched based on the coordinate calculation result to obtain the matching parking lot map data.

[0183] In a possible implementation, the semantic segmentation process is performed based on a semantic segmentation model, which may be based on U-Net (Convolutional Networks for Biomedical Image Segmentation

[0184] , deep learning segmentation network), SegNet (A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation, a deep fully convolutional neural network for pixel-level semantic segmentation), SegFormer (a semantic segmentation model based on Transformer), and other networks are designed, and the models are trained based on public data sets or collected data sets.

[0185] In a possible implementation, feature extraction is performed on the current environment information to obtain current environment features, and the current environment features represent elements of the parking lot environment that the vehicle is currently in. Specifically, the elements of the parking lot environment include key features such as gates, lane lines, parking space lines, pillars, road guide signs, speed bumps, and ground direction guide signs.

[0186] In a possible implementation, the initial parking lot map data is searched according to the elements of the parking lot environment in the current environment feature, and the map data corresponding to the current environment feature in the initial parking lot map data is determined as the matching parking lot map data.

[0187] In step S603, three-dimensional modeling is performed based on the matched parking lot map data to obtain three-dimensional parking lot map data to display the floors, entrances and exits, and parking spaces of the target parking lot.

[0188] In a possible implementation, three-dimensional modeling is performed on the matching parking lot map data to determine information such as entrances and exits, parking lot floors, lanes, and parking spaces in the matching parking lot map data.

[0189] In step S605, the three-dimensional parking lot map data is determined to be the target parking lot map data.

[0190] In a possible implementation, the three-dimensional modeled map data matching the parking lot map data is displayed on the vehicle-side system to facilitate the vehicle user to select a parking space.

[0191] In this implementation, feature matching processing is performed based on the current environment information and the initial parking lot map data to obtain target parking lot map data. The target parking lot map data is determined by feature matching, which reduces the amount of calculation and improves the speed of determining parking lot data.

[0192] In step S109, target parking position information is determined based on the target parking lot map data, and path planning processing is performed based on the current position information, the target parking lot map data, and the target parking position information to determine a target parking path.

[0193] In a possible implementation, after determining the target parking location information, path planning is performed based on the current location information, the target parking lot map data and the target parking location information. The path planning algorithm may be a Dijkstra algorithm, a Floyd-Warshall algorithm, a D* (D-star) algorithm, etc.

[0194] In step S1011, the vehicle is controlled to move to a parking space corresponding to the target parking position information according to the target parking path.

[0195] In a possible implementation, the current position information of the vehicle indicates that the vehicle is located in a parking space in a parking lot. When a user of the vehicle needs to use the vehicle, the vehicle can be controlled to move from the parking space to the location of the user.

[0196] In a possible implementation, the target parking path is displayed by a vehicle-side system.

[0197] In a possible implementation, after controlling the vehicle to move to the parking space corresponding to the target parking position information, the method includes:

[0198] Uploading the current location information of the vehicle and the current environment information to a cloud server;

[0199] Performing semantic segmentation on the current environment information based on the cloud server to obtain environment image features;

[0200] Perform feature matching processing based on the current location information, the environmental image features and the positioning system to generate environmental fragment map data;

[0201] Based on the environmental fragment map data, the map data corresponding to the current location information in the shared map is updated.

[0202] In one possible implementation, after controlling the vehicle to move to the parking space corresponding to the target parking position information, the current position information of the vehicle and the current environment information are uploaded to the cloud server as historical parking lot image data, and the shared map is updated to ensure the freshness of the shared map.

[0203] In this implementation, by continuously collecting map data from the vehicle side and uploading it to the cloud service platform for data integration, the parking lot map data coverage is improved and can be kept fresh, thereby reducing the cost of parking lot mapping and improving positioning accuracy.

[0204] In one possible implementation, Figure 7 As shown, the object sends the parking instruction to the cloud server through the mobile APP, obtains the current location information of the vehicle and the current environment information of the environment where the vehicle is located based on the vehicle's vehicle-side system, and uploads the current location information and the current environment information to the cloud server through the vehicle's communication unit. The cloud server determines the target parking lot map data based on the current location information and the current environment information, and transmits the target parking lot map data to the vehicle-side system. The vehicle-side system realizes parking based on the target parking lot map data.

[0205] In a possible implementation manner, the object may be a driver of the vehicle or other vehicle user.

[0206] In one possible implementation, Figure 8 As shown, the mobile APP is a city parking APP, which includes a user management unit, a parking information determination unit, a vehicle position determination unit and a track display unit. Specifically, the user management unit is used for user information registration, user information modification and user information deletion; the parking information determination unit is used for parking lot position display and parking lot three-dimensional modeling; the vehicle position determination unit is used to determine the position of the vehicle with the assistance of a camera, GNSS positioning and a gyroscope; the track display unit is used to display the vehicle movement track.

[0207] In one possible implementation, the object may park according to the target parking path displayed on the vehicle-side system, and the vehicle may also be controlled by the vehicle-side system to perform automatic parking.

[0208] Fig. 9 The structure diagram of a vehicle control device 900 based on a shared map provided in an embodiment of the present application is shown. The device has the function of implementing the vehicle control method based on a shared map in the above method embodiment. The function can be implemented by hardware or by hardware executing corresponding software. Fig. 9 As shown, the device may include:

[0209] The current information acquisition module 901 is used to acquire the current position information of the vehicle and the current environment information of the vehicle;

[0210] The instruction generation module 902 is used to generate a parking map acquisition instruction based on the current position information; the parking map acquisition instruction includes coordinate information of a target parking lot that meets a preset parking condition;

[0211] The initial map data acquisition module 903 is used to obtain the initial parking lot map data corresponding to the coordinate information of the target parking lot from the shared map in response to the parking map acquisition instruction; the shared map is the map data of the pre-constructed parking lot environment stored in the cloud server;

[0212] A target map data acquisition module 904 is used to perform feature matching processing based on the current environment information and the initial parking lot map data to obtain target parking lot map data;

[0213] A target parking path determination module 905 is used to determine target parking position information based on the target parking lot map data, and to perform path planning processing based on the current position information, the target parking lot map data and the target parking position information to determine the target parking path;

[0214] The parking control module 906 is used to control the vehicle to move to the parking space corresponding to the target parking position information according to the target parking path.

[0215] In a possible implementation, the current information acquisition module 901 is used to:

[0216] In response to a parking instruction issued by an object, determining position information of the vehicle based on a positioning system of the vehicle, and determining the position information as the current position information;

[0217] After determining the current position information, the image acquisition device of the vehicle is controlled to acquire images of the environment around the vehicle, and the acquired image data is used as the current environment information.

[0218] In a possible implementation, the current location information includes the coordinates of the vehicle, and the instruction generation module 902 is used to:

[0219] Based on the coordinates of the vehicle and the positioning system of the vehicle, searching for a plurality of parking lots to be selected whose distances from the vehicle meet a preset distance condition;

[0220] Obtain the number of available parking spaces corresponding to each of the plurality of parking lots to be selected;

[0221] Determine a candidate parking lot with a number of vacant parking spaces greater than a preset number as a target parking lot;

[0222] Acquire the coordinates of the target parking lot, and determine the coordinates of the target parking lot as the target parking lot information;

[0223] The parking map acquisition instruction is generated based on the target parking lot information.

[0224] In a possible implementation, the initial map data acquisition module 903 is used to:

[0225] Based on the coordinates of the target parking lot, map data corresponding to the coordinates of the target parking lot is acquired from the shared map, and the map data corresponding to the coordinates of the target parking lot is determined as initial parking lot map data.

[0226] In a possible implementation, the target map data acquisition module 904 is used to:

[0227] Extracting features from the current environment information to obtain current environment features, and performing matching processing on the initial parking lot map data based on the current environment features to obtain matching parking lot map data;

[0228] Performing three-dimensional modeling based on the matched parking lot map data to obtain three-dimensional parking lot map data to display the floors, entrances and exits, and parking spaces of the target parking lot;

[0229] The three-dimensional parking lot map data is determined as the target parking lot map data.

[0230] In a possible implementation, the target map data acquisition module 904 includes a matching unit, and the matching unit is used to:

[0231] Performing semantic segmentation processing on the current environment information to obtain semantic features; the semantic features are used to characterize the location information of lane lines, gates, speed bumps, signboards and parking spaces in the environment where the vehicle is located;

[0232] Performing coordinate calculation based on the semantic features and the coordinates of the vehicle to obtain a coordinate calculation result;

[0233] The initial parking lot map data is searched based on the coordinate calculation result to obtain the matching parking lot map data.

[0234] In a possible implementation, the device further includes a shared map construction module 907, and the shared map construction module 907 is configured to:

[0235] In the case where the historical vehicle is located in a historical parking environment, an image acquisition device based on the historical vehicle acquires images of the environment around the historical vehicle to obtain historical parking image data;

[0236] Uploading the historical parking lot image data and the current location information of the historical vehicle to a cloud server, wherein the current location information includes the coordinates of the historical vehicle;

[0237] Performing semantic segmentation on the historical parking lot image data based on a cloud server to obtain historical parking lot image features;

[0238] Perform feature matching processing based on the current position information of the historical vehicle, the image features of the historical parking lot, and the positioning system to generate historical parking lot fragment map data;

[0239] fusing the historical parking lot fragment map data corresponding to the historical parking lot to obtain historical parking lot map data;

[0240] The shared map is constructed based on the historical parking lot map data.

[0241] In a possible implementation, the device further includes an updating module 908,

[0242] After controlling the vehicle to move to the parking space corresponding to the target parking position information, the updating module is used to:

[0243] Uploading the current location information of the vehicle and the current environment information to a cloud server;

[0244] Performing semantic segmentation on the current environment information based on the cloud server to obtain environment image features;

[0245] Perform feature matching processing based on the current location information, the environmental image features and the positioning system to generate environmental fragment map data;

[0246] Based on the environmental fragment map data, the map data corresponding to the current location information in the shared map is updated.

[0247] It should be noted that the device provided in the above embodiment, when implementing its functions, is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0248] An embodiment of the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement any one of the shared map-based vehicle control methods provided in the above method embodiments.

[0249] The memory can be used to store software programs and modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, application programs required for functions, etc.; the data storage area can store data created according to the use of the device, etc. In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory can also include a memory controller to provide the processor with access to the memory.

[0250] An embodiment of the present application also provides a computer-readable storage medium, which can be set in an electronic device to store at least one instruction or at least one program related to implementing a vehicle control method based on a shared map. The at least one instruction or the at least one program is loaded and executed by the processor to implement any one of the vehicle control methods based on a shared map provided in the above method embodiments.

[0251] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0252] It should be noted that the above-mentioned sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above-mentioned specific embodiments of this specification are described. Other embodiments are within the scope of the attached claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0253] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0254] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0255] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A vehicle control method based on a shared map, characterized in that: The method comprises: Acquiring current location information of the vehicle and current environment information of the environment in which the vehicle is located; generating a parking map acquisition instruction based on the current position information; the parking map acquisition instruction includes coordinate information of a target parking lot that meets a preset parking condition; In response to the parking map acquisition instruction, initial parking lot map data corresponding to the coordinate information of the target parking lot is acquired from a shared map; the shared map is map data of a pre-constructed parking lot environment stored in a cloud server; Perform feature matching processing based on the current environment information and the initial parking lot map data to obtain target parking lot map data; Determining target parking location information based on the target parking lot map data, performing path planning processing based on the current location information, the target parking lot map data and the target parking location information to determine a target parking path; The vehicle is controlled to move to a parking space corresponding to the target parking position information according to the target parking path.

2. The vehicle control method based on a shared map according to claim 1, characterized in that: The obtaining of the current position information of the vehicle and the current environment information of the environment in which the vehicle is located includes: In response to a parking instruction issued by an object, determining position information of the vehicle based on a positioning system of the vehicle, and determining the position information as the current position information; After determining the current position information, the image acquisition device of the vehicle is controlled to acquire images of the environment around the vehicle, and the acquired image data is used as the current environment information.

3. The vehicle control method based on a shared map according to claim 1, characterized in that: The current location information includes the coordinates of the vehicle, The generating of the parking map acquisition instruction based on the current position information includes: Based on the coordinates of the vehicle and the positioning system of the vehicle, searching for a plurality of parking lots to be selected whose distances from the vehicle meet a preset distance condition; Obtain the number of available parking spaces corresponding to each of the plurality of parking lots to be selected; Determine a candidate parking lot with a number of vacant parking spaces greater than a preset number as a target parking lot; Acquire the coordinates of the target parking lot, and determine the coordinates of the target parking lot as the target parking lot information; The parking map acquisition instruction is generated based on the target parking lot information.

4. The vehicle control method based on a shared map according to claim 3, characterized in that: The obtaining the initial parking lot map data corresponding to the target parking lot information from the shared map includes: Based on the coordinates of the target parking lot, map data corresponding to the coordinates of the target parking lot is acquired from the shared map, and the map data corresponding to the coordinates of the target parking lot is determined as initial parking lot map data.

5. The vehicle control method based on a shared map according to claim 1, characterized in that: The performing feature matching processing based on the current environment information and the initial parking lot map data to obtain target parking lot map data includes: Extracting features from the current environment information to obtain current environment features, and performing matching processing on the initial parking lot map data based on the current environment features to obtain matching parking lot map data; Performing three-dimensional modeling based on the matched parking lot map data to obtain three-dimensional parking lot map data to display the floors, entrances and exits, and parking spaces of the target parking lot; The three-dimensional parking lot map data is determined as the target parking lot map data.

6. The vehicle control method based on a shared map according to claim 5, characterized in that: The extracting features of the current environment information to obtain current environment features, and performing matching processing on the initial parking lot map data based on the current environment features to obtain matching parking lot map data, includes: Performing semantic segmentation processing on the current environment information to obtain semantic features; the semantic features are used to characterize the location information of lane lines, gates, speed bumps, signboards and parking spaces in the environment where the vehicle is located; Performing coordinate calculation based on the semantic feature and the coordinates of the vehicle to obtain a coordinate calculation result; The initial parking lot map data is searched based on the coordinate calculation result to obtain the matching parking lot map data.

7. The vehicle control method based on a shared map according to claim 1, characterized in that: The method for constructing the shared map includes: In the case where the historical vehicle is located in a historical parking environment, an image acquisition device based on the historical vehicle acquires images of the environment around the historical vehicle to obtain historical parking image data; Uploading the historical parking lot image data and the current location information of the historical vehicle to a cloud server, wherein the current location information includes the coordinates of the historical vehicle; Performing semantic segmentation on the historical parking lot image data based on a cloud server to obtain historical parking lot image features; Perform feature matching processing based on the current position information of the historical vehicle, the image features of the historical parking lot, and the positioning system to generate historical parking lot fragment map data; fusing the historical parking lot fragment map data corresponding to the historical parking lot to obtain historical parking lot map data; The shared map is constructed based on the historical parking lot map data.

8. The vehicle control method based on a shared map according to claim 1, characterized in that: After controlling the vehicle to move to the parking space corresponding to the target parking position information, the method includes: Uploading the current location information of the vehicle and the current environment information to a cloud server; Performing semantic segmentation on the current environment information based on the cloud server to obtain environment image features; Perform feature matching processing based on the current location information, the environmental image features and the positioning system to generate environmental fragment map data; Based on the environmental fragment map data, the map data corresponding to the current location information in the shared map is updated.

9. A vehicle control device based on a shared map, characterized in that: The device comprises: A current information acquisition module is used to acquire the current position information of the vehicle and the current environment information of the vehicle; An instruction generation module, configured to generate a parking map acquisition instruction based on the current position information; the parking map acquisition instruction includes coordinate information of a target parking lot that meets a preset parking condition; an initial map data acquisition module, configured to acquire initial parking lot map data corresponding to the coordinate information of the target parking lot from a shared map in response to the parking map acquisition instruction; the shared map is map data of a pre-constructed parking lot environment stored in a cloud server; A target map data acquisition module, configured to perform feature matching processing based on the current environment information and the initial parking lot map data to obtain target parking lot map data; a target parking path determination module, configured to determine target parking position information based on the target parking lot map data, and to perform path planning processing based on the current position information, the target parking lot map data, and the target parking position information to determine the target parking path; The parking control module is used to control the vehicle to move to the parking space corresponding to the target parking position information according to the target parking path.

10. An electronic device, characterized in that: It includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the vehicle control method based on a shared map as described in any one of claims 1 to 8.

Images