Fusion map generation method and intelligent car control method

Abstract

The invention provides a fusion map generation method and an intelligent car control method. The fused map generation method comprises the following steps: mapping original map data of a plurality of navigation modes into a plurality of logic sub-layers of a fused map, and generating standard map data in different logic sub-layers; when the navigation mode is the laser radar navigation mode, generating standard map data in the background sub-layer; and when the navigation mode is the linear navigation mode or the matrix two-dimensional code navigation mode, generating standard map data in the path sub-layer. Original map data from different sources are fused into unified standard map data in a plurality of logic sub-layers of a fused map through a mapping relation, isomerism is digested through the fused map, and the problem of complexity of an original system is solved. According to the method, the existing facts are respected, the data consistency and uniformity are realized, the universality and usability of the fused map are enhanced, and the adaptability of the fused map is improved.

Description

Technical Field

[0001] This application relates to the field of intelligent control technology, and more specifically, to a method for controlling an intelligent vehicle that integrates a map generation method and an intelligent vehicle control method. Background Technology

[0002] Intelligent Guided Vehicles (IGVs) are unmanned transport devices used in automated terminals and smart factories. They achieve autonomous navigation and obstacle avoidance through technologies such as the BeiDou Navigation Satellite System, LiDAR, and visual SLAM. They possess high flexibility, intelligent navigation, and superior performance. IGVs typically follow navigation information provided by QR code strips or matrix QR codes.

[0003] The current production line uses various IGV systems, including magnetic navigation, QR code navigation, and laser SLAM navigation, each operating independently. This results in issues such as non-standard maps, incompatible scheduling, complex traffic control, and poor system scalability, leading to low overall logistics efficiency and becoming a production bottleneck.

[0004] Therefore, this application provides a method for generating fused maps to solve the above-mentioned technical problems. Summary of the Invention

[0005] The purpose of this application is to provide a method that integrates map generation and intelligent vehicle control, thereby solving at least one of the aforementioned technical problems. The specific solution is as follows: According to a specific embodiment of this application, in a first aspect, this application provides a method for generating a fused map, comprising: Obtain raw map data for multiple navigation modes, including: linear navigation mode, matrix QR code navigation mode, and LiDAR navigation mode; The original map data of the multiple navigation modes are mapped to multiple logical sub-layers of the fused map, and standard map data is generated in different logical sub-layers, wherein the multiple logical sub-layers include a background sub-layer and a path sub-layer; When the navigation mode is the LiDAR navigation mode, standard map data is generated in the background sub-layer; When the navigation mode is the linear navigation mode or the matrix QR code navigation mode, standard map data is generated in the path sub-layer.

[0006] Optionally, when the navigation mode is the LiDAR navigation mode, standard map data is generated on the background sub-layer, including: When the navigation mode is the LiDAR navigation mode, the raster data and / or point cloud data of the LiDAR navigation mode are mapped to the background sub-layer to generate standard map data that constitutes the environment model.

[0007] Optionally, when the navigation mode is the linear navigation mode or the matrix QR code navigation mode, generating standard map data in the path sub-layer includes: The original map data of each of the various linear navigation modes and the matrix QR code navigation modes are mapped to the path sub-layer, and standard map data including virtual reference lines and network graphs are generated, wherein the network graphs include key edge information and key node information.

[0008] Optionally, the original map data of the multiple navigation modes are mapped to multiple logical sub-layers of the merged map, including: The standard map data in the background sublayer and the path sublayer are mapped to the traffic sublayer, which includes standard map data of various key traffic areas defined based on a preset traffic strategy.

[0009] Optionally, after mapping the original map data of the multiple navigation modes to multiple logical sub-layers of the fused map, the method further includes: Based on the scope of the application scenario, determine the standard map data for at least two logical sub-layers among the plurality of logical sub-layers; The standard map data from the at least two logical sub-layers are overlaid to generate a basic application map; Based on the standard map data and functional requirements in the aforementioned basic application map, extract multiple map annotation information and multiple functional annotation information; The basic application map is annotated based on the multiple functional annotation information and the multiple map annotation information to generate a virtual application map.

[0010] Optionally, the method further includes: Establish a one-to-one mapping relationship between the standard map data in the at least two logical sub-layers and the standard map data in the virtual application map, as well as the logical relationship between the standard map data in the virtual application map and the map annotation information; When any standard map data in any logical sub-layer changes, the corresponding standard map data in the virtual application map is triggered to change based on the one-to-one mapping relationship, and the corresponding map annotation information is also triggered to change based on the logical relationship.

[0011] Optionally, the original map data of the multiple navigation modes are mapped to multiple logical sub-layers of the merged map, including: After aligning the original map data of the multiple navigation modes with coordinates, they are mapped to multiple logical sub-layers of the fused map.

[0012] According to a specific embodiment of this application, in a second aspect, this application provides a control method for an intelligent vehicle, comprising: The target task and real-time traffic conditions of the target intelligent vehicle are obtained, as well as the traffic sub-layer of the fused map in the method of claim 4; A virtual traffic map is generated based on the target task and the traffic sub-layer; The real-time traffic conditions are mapped onto the virtual traffic map to obtain the real-time traffic map; The navigation path for the target intelligent vehicle to perform the target task is determined based on the real-time traffic map.

[0013] Optionally, generating a virtual traffic map based on the target task and the traffic sub-layer includes: The target standard map data in the traffic sub-layer is determined according to the task scope of the target task, wherein the target standard map data includes standard map data of the environment model related to the target task, standard map data of the virtual reference line and network diagram, and standard map data of various key traffic areas. Based on the target standard map data, a basic traffic map is generated; Based on the target standard map data in the basic traffic map and the target task, extract multiple traffic environment annotation information and multiple traffic control annotation information; The basic traffic map is annotated based on the multiple traffic environment annotations and the multiple traffic control annotations to generate a virtual traffic map.

[0014] Optionally, after determining the navigation path for the target intelligent vehicle to perform the target task, the method further includes: When the target smart car is in the standard pickup ready state, traffic conditions are predicted for the navigation route based on the real-time traffic map. Based on the prediction results and the supply time window in the target task, the passage time window reservation is made for the intersection points in the navigation path, and the passage time window of each successfully reserved intersection point is declared. Based on the passage time window of each intersection point in the navigation path, a control command for the target intelligent vehicle is generated and sent to the target intelligent vehicle. The control command for the target intelligent vehicle includes controlling the target intelligent vehicle to travel at a set speed between two adjacent intersection points and to pass through each intersection point within the passage time window.

[0015] Optionally, after determining the navigation path for the target intelligent vehicle to perform the target task, the method further includes: In the real-time traffic map, when it is determined that the first intelligent vehicle and the second intelligent vehicle are heading towards the target intersection of the navigation path via different paths, the meeting of the first intelligent vehicle and the second intelligent vehicle is predicted based on their respective current standard driving status information, wherein one of the first intelligent vehicle and the second intelligent vehicle is the target intelligent vehicle; When the current standard driving status information of the first intelligent vehicle and the second intelligent vehicle respectively meets the preset impending encounter condition, control commands for the first intelligent vehicle and the second intelligent vehicle to avoid each other are generated based on their respective current standard driving status information, the priority of their respective target tasks, and the standard position information of the target intersection point. These commands include: When the priority of the target task of the first intelligent vehicle is higher than the priority of the target task of the second intelligent vehicle, the current standard driving speed value and current standard position information of the first intelligent vehicle are obtained. The current standard remaining distance value is obtained based on the current standard location information of the first intelligent vehicle and the standard location information of the target intersection point; Based on the current time point, the current first standard remaining distance value, and the current standard driving speed value of the first intelligent vehicle, a first standard passage time window is obtained for the first intelligent vehicle to pass through the target intersection point; The second standard passage time window for the second intelligent vehicle to pass through the target intersection point is obtained based on the first standard passage time window of the first intelligent vehicle and the preset safe avoidance time interval. The current standard distance value of the second intelligent vehicle is obtained based on the current standard location information of the second intelligent vehicle and the standard location information of the target intersection point; The standard driving speed of the second intelligent vehicle is obtained based on the current remaining distance value of the second standard and the second standard travel time window. The avoidance control command includes controlling the first intelligent vehicle to pass through the intersection at its standard driving speed within the first standard traffic time window, and controlling the second intelligent vehicle to pass through the intersection at its standard driving speed within the second standard traffic time window.

[0016] Optionally, control commands for the intelligent vehicle are generated, including: Generate standard control commands for the intelligent vehicle; The standard control commands are converted into dedicated control commands that can control the smart car.

[0017] Optionally, before obtaining the real-time traffic map based on the real-time traffic conditions and the virtual traffic map, the method further includes: Based on the virtual traffic map, a one-to-one conversion relationship is defined between the standard control commands for each working mode and the exclusive control commands for the intelligent vehicle in the corresponding working mode; The step of converting the standard control commands into dedicated control commands that can control the smart car includes converting the standard control commands into dedicated control commands according to the conversion relationship.

[0018] Compared with the prior art, the above-described solutions of this application have at least the following beneficial effects: This application provides a method for generating a fused map and a control method for an intelligent vehicle. The fused map generation method maps raw map data from multiple navigation modes to multiple logical sub-layers of the fused map, and generates standard map data in different logical sub-layers. When the navigation mode is the LiDAR navigation mode, standard map data is generated in the background sub-layer; when the navigation mode is the linear navigation mode or the matrix QR code navigation mode, standard map data is generated in the path sub-layer. By fusing raw map data from different sources into unified standard map data in multiple logical sub-layers of the fused map through mapping relationships, the heterogeneity of the fused map is absorbed, solving the complexity problem of the original system. It respects existing facts, achieves data consistency and uniformity, enhances the versatility and usability of the fused map, and improves its adaptability.

[0019] The control method for the intelligent vehicle generates a virtual traffic map based on the target task and traffic sub-layers; obtains a real-time traffic map through real-time traffic conditions and the virtual traffic map; and finally determines the navigation path for the target intelligent vehicle to perform the target task based on the real-time traffic map. By combining the digitally constructed virtual traffic map with real-time traffic conditions, a high-fidelity digital twin of the production line logistics, synchronized in real time with the physical world, is created. This achieves rapid real-time data linkage and millisecond-level rapid simulation. By sensing the dynamics of the target intelligent vehicle and road changes in real time, data is provided for rapid prediction of traffic changes, improving the efficiency of dynamic perception and decision-making. Attached Figure Description

[0020] Figure 1 A flowchart of a fused map generation method according to an embodiment of this application is shown; Figure 2 A flowchart of a control method for an intelligent vehicle according to an embodiment of this application is shown. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the application. The singular forms “a,” “said,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0023] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0024] It should be understood that although the terms first, second, third, etc., may be used in the embodiments of this application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, first may also be referred to as second without departing from the scope of the embodiments of this application, and similarly, second may also be referred to as first.

[0025] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”

[0026] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0027] It should be noted that any symbols and / or numbers present in the specification that are not marked in the accompanying drawings are not reference numerals.

[0028] The optional embodiments of this application are described in detail below with reference to the accompanying drawings.

[0029] Example 1 The following is combined Figure 1 The embodiments of this application will be described in detail.

[0030] This application provides a method for generating a fused map, including: Step S101: Obtain the raw map data for multiple navigation modes.

[0031] The multiple navigation modes include: linear navigation mode, matrix QR code navigation mode, and lidar navigation mode.

[0032] Linear navigation modes include magnetic navigation mode and linear QR code navigation mode.

[0033] The intelligent vehicle can be an Automated Guided Vehicle (AGV) or an Infinite Vehicle (IGV).

[0034] The matrix QR code navigation mode distributes matrix QR codes at various navigation points on the ground. The intelligent car reads the information of one matrix QR code and drives to the next matrix QR code. This process is repeated until the navigation destination is reached. The LiDAR navigation mode is a method in which the intelligent vehicle uses LiDAR to scan the surrounding terrain to obtain point cloud information, matches it with the original map data of the LiDAR navigation mode, identifies the road to travel, and continues until it reaches the navigation destination. In magnetic navigation mode, a magnetic navigation strip is placed on the ground, and the intelligent car is guided by the magnetic navigation strip to reach the navigation destination. The linear QR code navigation mode involves placing a QR code navigation strip on the ground or in the air. The intelligent vehicle reaches the navigation destination by reading the guidance information on the QR code on the navigation strip.

[0035] The original map data includes: grid data and / or point cloud data of the LiDAR navigation mode, guidance information of the QR code on the QR code navigation strip of the linear QR code navigation mode, guidance information recorded on the magnetic navigation strip of the magnetic navigation mode, and guidance information of the matrix QR code of the matrix QR code navigation mode.

[0036] Step S102: Map the original map data of the multiple navigation modes to multiple logical sub-layers of the fused map, and generate standard map data in different logical sub-layers.

[0037] The plurality of logical sub-layers include a background sub-layer and a path sub-layer.

[0038] Because the original map data comes from different sources and has different representations in the original electronic maps, when mapping the original map data from the original electronic maps to the merged map, the original map data is transformed into standard map data with a unified representation format in the merged map. For example, a global world coordinate system is established with a fixed corner of the production line (such as the geodetic datum) as the origin, and all standard map data uses this coordinate system as a reference. This unifies the original map data under different coordinate systems into standard map data under the global world coordinate system.

[0039] The standard map data includes: standard map data for the environment model, standard map data for the virtual reference lines and network graph, wherein the network graph includes key edge information and key node information.

[0040] In some specific embodiments, mapping the original map data of the multiple navigation modes to multiple logical sub-layers of the fused map includes: Step S102a: After aligning the original map data of the multiple navigation modes with coordinates, map them to multiple logical sub-layers of the fused map.

[0041] Since the original map data originates from different electronic maps, and these electronic maps use different coordinate systems, in order to transform the original map data into standard map data, the original map data needs to be aligned to the coordinate system of the fused map, and then mapped to multiple logical sub-layers of the fused map. This solves the problem of differences in original map data for different navigation modes, enables real-time data fusion, and provides efficient and unified basic data support for various application scenarios.

[0042] Step S103a: When the navigation mode is the LiDAR navigation mode, standard map data is generated in the background sub-layer.

[0043] In some specific embodiments, the specific components include: Step S103aa: When the navigation mode is the LiDAR navigation mode, the raster data and / or point cloud data of the LiDAR navigation mode are mapped to the background sub-layer to generate standard map data constituting the environment model.

[0044] Among them, the environmental model is a twin digital model of the factory environment constructed using standard map data, which is used to determine the impact of the factory environment on the operation of the intelligent vehicle.

[0045] Step S103b: When the navigation mode is the linear navigation mode or the matrix QR code navigation mode, standard map data is generated in the path sub-layer.

[0046] In some specific embodiments, the specific components include: Step S103ba: Map the original map data of each of the various linear navigation modes and the matrix QR code navigation mode to the path sub-layer, and generate standard map data including virtual reference lines and network graphs.

[0047] The network graph includes key edge information and key node information.

[0048] Virtual reference lines are used to represent navigation paths in linear navigation mode or matrix QR code navigation mode, and to mark the driving information of the smart car on the navigation path.

[0049] For example, in the path sub-layer, a series of continuous virtual reference lines are extracted from the center line of the magnetic navigation strip. Each virtual reference line has a unique identifier and is labeled with driving information (such as allowed driving direction, maximum driving speed, etc.) on the virtual navigation line. The physical coordinates of each QR code are accurately measured and mapped to the world coordinates in the world coordinate system to establish a mapping table between "QR code - unique identifier" and world coordinates.

[0050] Key nodes refer to critical locations in practical applications, such as QR code points (e.g., positioning and navigation nodes for smart cars), intersections, turning points, and work sites; key node information refers to the information representing key nodes in the path sub-layer. A critical edge is a path that connects key nodes in a practical application; critical edge information refers to the information in the path sublayer that represents the critical edge, such as virtual navigation lines or drivable areas in free space.

[0051] In some specific embodiments, mapping the original map data of the multiple navigation modes to multiple logical sub-layers of the fused map includes: Step S111: Map the standard map data in the background sublayer and the path sublayer to the traffic sublayer. The traffic sublayer includes standard map data of various key traffic areas defined based on a preset traffic strategy.

[0052] Preset traffic strategies are designed to delineate key traffic areas within a site based on production needs and site conditions. These strategies are applied to the traffic sublayer to define the locations of these key traffic areas. For example, key traffic areas include: one-way streets, two-way streets, speed-limited zones, no-stopping zones, intersections, charging areas, buffer zones, and process areas. One-way or two-way streets can be defined based on virtual reference lines mapped from the background sublayer to the traffic sublayer; intersections can be defined based on the environment model mapped from the path sublayer to the traffic sublayer. This provides the traffic sublayer with basic traffic area divisions.

[0053] This application's embodiments fuse original map data from different sources into unified standard map data across multiple logical sub-layers of a fused map through mapping relationships. This fused map mitigates heterogeneity and resolves the complexity issues of the original system. It respects existing facts while achieving data consistency and uniformity, enhancing the versatility and usability of the fused map and improving its adaptability.

[0054] Based on the fused map, a virtual application map suitable for the application can be generated according to the application scenario and functional requirements. In some specific embodiments, after mapping the original map data of the multiple navigation modes to multiple logical sub-layers of the fused map, the method further includes: Step S121: Determine the standard map data of at least two logical sub-layers among the multiple logical sub-layers according to the scenario scope of the application scenario.

[0055] The standard map data for at least two logical sub-layers is data within the scope of the application scenario. For example, if the scenario scope is only used for smart cars to deliver goods within a workshop, then only the standard map data for that workshop is obtained; if the scenario scope is used for smart cars to deliver goods across workshops, then standard map data for multiple workshops is obtained.

[0056] Step S122: Overlay the standard map data from the at least two logical sub-layers to generate a basic application map.

[0057] Step S123: Determine multiple map annotation information and multiple function annotation information based on the standard map data and functional requirements in the basic application map.

[0058] Map annotation information refers to the annotation of geographical information on standard map data. For example, if the purpose is to deliver goods by a smart car, various types of map annotation information such as process islands, material warehouses, driving routes, road segments, and intersections are extracted from the basic application map.

[0059] Functional labeling information refers to the labeling of functions and uses of standard map data. For example, if it is for smart car delivery, functional labeling information such as one-way streets, two-way streets, driving time limits, and driving speed limits are extracted from the basic application map.

[0060] Step S124: Based on the multiple functional annotation information and the multiple map annotation information, annotate the basic application map to generate a virtual application map.

[0061] For example, if the purpose is to deliver goods by a smart car, the map annotation information of the process island in the environment model of the basic application map is used to annotate the virtual reference line, the map annotation information of the driving path is used to annotate the key edge, the map annotation information of the road segment is used to annotate the key node, and the map annotation information of the intersection point is used to annotate the key node.

[0062] For example, if the purpose is for smart vehicles to deliver goods, the system would indicate whether the first road segment is a one-way or two-way street; or indicate that the first road segment is a one-way street in the first time period and a two-way street in the second time period; or indicate that the first road segment travels at normal speed in the third time period and at a speed limit in the fourth time period. When the smart vehicle is driving, it can achieve autonomous driving by using the map annotations and functional annotations in the virtual application map.

[0063] This specific embodiment creates a high-fidelity digital twin that is synchronized with the physical world in real time. That is, based on the merged map, it can generate a virtual application map suitable for the application scenario and functional requirements. The merged map simplifies the process of generating virtual application maps, enabling users to quickly achieve their application goals.

[0064] In some specific embodiments, the method further includes: Step S131: Establish a one-to-one mapping relationship between the standard map data in the at least two logical sub-layers and the standard map data in the virtual application map, as well as the logical relationship between the standard map data in the virtual application map and the map annotation information.

[0065] Step S132: When any standard map data in any logical sub-layer changes, the corresponding standard map data in the virtual application map is triggered to change based on the one-to-one mapping relationship, and the corresponding map annotation information is also triggered to change based on the logical relationship.

[0066] This specific embodiment establishes a one-to-one mapping relationship between the virtual application map and the data of at least two associated logical sub-layers. If the standard map data in a logical sub-layer changes, the related standard map data in the virtual application map, as well as other logically related data, also change accordingly. This maintains consistency between the original data and the application data, improving situational awareness and response capabilities.

[0067] This application's embodiments map raw map data from multiple navigation modes to multiple logical sub-layers of the fused map, and generate standard map data in different logical sub-layers. When the navigation mode is the LiDAR navigation mode, standard map data is generated in the background sub-layer; when the navigation mode is the linear navigation mode or the matrix QR code navigation mode, standard map data is generated in the path sub-layer. By fusing raw map data from different sources into unified standard map data in multiple logical sub-layers of the fused map through mapping relationships, the heterogeneity is absorbed through the fused map, solving the complexity problem of the original system. It respects existing facts, achieves data consistency and uniformity, enhances the versatility and usability of the fused map, and improves its adaptability.

[0068] Example 2 like Figure 2 As shown in the figure, this application provides a control method for an intelligent vehicle, including: Step S201: Obtain the target task and real-time traffic conditions of the target intelligent vehicle, as well as the traffic sub-layer of the fused map in the above fused map generation method.

[0069] In this embodiment, the group control device generates target tasks for the target intelligent vehicle based on the production requirements defined by MES / WMS or manually. For example, a material delivery task from point A to point B.

[0070] Real-time traffic data is collected by sensors (such as cameras, vehicle sensors, and infrared sensors) on-site to track the real-time travel path of intelligent vehicles. This data can include traffic data at intersections and turning points, queuing data at work sites, and traffic congestion data.

[0071] Step S203: Generate a virtual traffic map based on the target task and the traffic sub-layer.

[0072] In some specific embodiments, generating a virtual traffic map based on the target task and the traffic sub-layer includes: Step S203-1: Determine the target standard map data in the traffic sub-layer according to the task scope of the target task.

[0073] The target standard map data includes standard map data of the environmental model within the task scope of the target task, standard map data of virtual reference lines and network diagrams, and standard map data of various key traffic areas.

[0074] The target task includes a starting point and an ending point. These locations determine the scope of the target task. For example, if the task scope is limited to within a workshop, only the target standard map data for that workshop is acquired; if the task scope spans multiple workshops, target standard map data for multiple workshops within the task scope is acquired. The target standard map data can be limited to the specific target task, or its redundancy can be appropriately increased to expand the acquisition range, enabling timely responses based on dynamic changes in actual road conditions.

[0075] Step S203-2: Generate a basic traffic map based on the target standard map data.

[0076] In this specific embodiment, the target standard map data in the basic traffic map is the foundational data for performing the target task. Various virtual traffic maps suitable for specific needs can be generated based on the basic traffic map.

[0077] Step S203-3: Extract multiple traffic environment annotation information and multiple traffic control annotation information based on the target standard map data in the basic traffic map and the target task.

[0078] This specific embodiment extracts multiple traffic environment annotations and multiple traffic control annotations required for the target task from the basic traffic map.

[0079] Traffic environment labeling information refers to the annotation of geographic information on target standard map data. For example, if the target task is intelligent vehicle delivery, various types of traffic environment labeling information such as process islands, material warehouses, driving routes, road segments, and intersections are extracted from the basic traffic map.

[0080] Traffic control labeling information refers to the information that labels traffic and control functions on target standard map data. For example, if the target task is for a smart car to deliver goods, traffic control labeling information such as one-way streets, two-way streets, travel time restrictions, and travel speed restrictions will be extracted from the basic traffic map.

[0081] Step S203-4: Based on the multiple traffic environment annotation information and the multiple traffic control annotation information, the basic traffic map is annotated to generate a virtual traffic map.

[0082] For example, if the target task is to deliver goods by a smart car, then the traffic environment annotation information of the process island in the environment model of the basic traffic map is annotated, the traffic environment annotation information of the driving path is annotated for the virtual reference line, the traffic environment annotation information of the road segment is annotated for the key edge, and the traffic environment annotation information of the intersection point is annotated for the key node.

[0083] For example, if the target task is for the intelligent vehicle to deliver goods, traffic control information could be added to indicate whether the first road segment is a one-way or two-way street; or to indicate that the first road segment is a one-way street in the first time period and a two-way street in the second time period; or to indicate that the first road segment travels at normal speed in the third time period and is subject to speed limits in the fourth time period. When the intelligent vehicle performs the target task, it can achieve autonomous driving by using the traffic environment and traffic control information in the virtual traffic map.

[0084] Step S205: Obtain a real-time traffic map based on the real-time traffic conditions and the virtual traffic map.

[0085] The real-time traffic map includes standard driving status information of the smart car.

[0086] Standard driving status information is a dedicated representation of driving status information used in real-time traffic maps. Standard driving status information includes standard driving speed values ​​and standard location information. The standard driving speed value is the numerical representation of the intelligent vehicle's real-time driving speed in the real-time traffic map, and the standard location information is the numerical representation of the intelligent vehicle's actual location in the real-time traffic map.

[0087] This specific embodiment combines a digitally constructed virtual traffic map with real-time traffic conditions to obtain a real-time traffic map, creating a high-fidelity digital twin of the production line logistics that is synchronized with the physical world in real time.

[0088] In some specific embodiments, obtaining the real-time traffic map based on the real-time traffic conditions and the virtual traffic map includes: Step S205a: Map the real-time traffic conditions onto the virtual traffic map to obtain the real-time traffic map.

[0089] In this specific embodiment, real-time traffic conditions are mapped onto the virtual traffic map, enabling rapid real-time data linkage and millisecond-level simulation. By sensing the dynamics of the target intelligent vehicle and road changes in real time, data is provided for rapid prediction of traffic changes, improving the efficiency of dynamic perception and decision-making.

[0090] Step S207: Determine the navigation path for the target intelligent vehicle to perform the target task based on the real-time traffic map.

[0091] Multiple feasible navigation routes are determined based on real-time traffic conditions, traffic environment annotations, and traffic control annotations; the optimal navigation route is determined based on preset optimal navigation conditions.

[0092] The preset optimal navigation conditions include: shortest path, least congestion, and / or highest score.

[0093] For example, using graph search algorithms (such as...) Dijkstra's algorithm (or its variants) eliminates impassable road segments or turns affected by traffic control and the environment in the road network graph, generating multiple feasible paths, such as obtaining a candidate set through the K-shortest path algorithm or path enumeration.

[0094] Shortest path: Shortest static geographical distance.

[0095] Least congestion: Calculate the estimated travel time based on real-time traffic conditions and select the route with the shortest travel time.

[0096] The highest score is likely a comprehensive score (including road safety, number of traffic lights, road grade, user preferences, etc.).

[0097] For methods of determining the optimal navigation path, please refer to existing technologies; the embodiments in this application will not be described in detail here.

[0098] To ensure the unordered movement of the intelligent vehicle during delivery, in some specific embodiments, after determining the navigation path for the target intelligent vehicle to perform the target task, the method further includes: Step S209a-1: When the target smart car is in the standard pickup ready state, traffic conditions are predicted for the navigation route based on the real-time traffic map.

[0099] The standard pickup ready state is a dedicated representation used by real-time traffic maps to indicate the pickup ready state of intelligent vehicles. When an intelligent vehicle completes pickup, it is marked as this state on the real-time traffic map.

[0100] Predicting traffic conditions for the navigation route based on the real-time traffic map refers to predicting traffic conditions based on real-time traffic information in the real-time traffic map and the annotation information in the virtual traffic map. For example, real-time traffic information includes the speed and location of other smart cars, and the traffic environment information annotated in the virtual traffic map includes information about roads and surrounding areas that can affect the smart car's movement (such as process islands, buffer zones, intersections, and smart car queues). The navigation route is divided into multiple segments using process islands, buffer zones, and intersections. The average speed of each smart car on each segment is used to determine the speed of that segment. The travel time of the corresponding segment is obtained based on the quotient of the length of each segment and the average speed of the smart cars on that segment. The total travel time of the navigation route can be determined by the speed and travel time of each segment.

[0101] For example, using graph search algorithms (such as...) Dijkstra's algorithm (or its variants) eliminates impassable road segments or turns affected by traffic control and the environment in the road network graph, generating multiple feasible paths. For example, a candidate set can be obtained through the K-shortest path algorithm or path enumeration. The estimated travel time is calculated based on real-time traffic conditions, and the path with the shortest travel time is determined to be the smooth-traffic path. The estimated travel time is calculated based on real-time traffic conditions, and the path with the longest travel time is determined to be the congested path. When no intelligent vehicle lists an intersection point as a travel path within a preset idle time, the intersection point is determined to be idle.

[0102] Step S209a-2: Based on the prediction results and the supply time window in the target task, reserve passage time windows for intersection points in the navigation path, and declare the passage time window for each successfully reserved intersection point.

[0103] The target task includes a supply time window. The group control device generates the supply time window in the target task of the target intelligent vehicle based on production demand.

[0104] For example, after predicting traffic conditions on the navigation path, intersection points are extracted from the navigation path. Based on the supply time window and the predicted traffic conditions (including driving speed and travel time on the road segment), a predicted arrival time for each intersection point is generated. The supply time window refers to the time window for replenishing materials to the process island. The predicted arrival time for each intersection point is coordinated with the travel time windows of other smart cars that have successfully booked in the reservation queue of the corresponding intersection point. Based on the predicted arrival time of the target smart car, a time interval including the predicted arrival time is formed, and the upper and lower limits of the time interval are adjusted so that the time interval of the target smart car does not overlap with the travel time windows of other smart cars that have successfully booked in the reservation queue, thus forming the travel time window of the target smart car. The travel time window of the target smart car at each intersection point is added to the reservation queue of the corresponding intersection point, declaring that the target smart car has been successfully booked at the corresponding intersection point.

[0105] This specific embodiment locks in the time for the intelligent vehicle performing the target task to pass through the intersection by reserving a passage time window at the intersection, thereby avoiding conflicts with other intelligent vehicles at the intersection.

[0106] Step S209a-3: Based on the passage time window of each intersection point in the navigation path, generate control commands for the target intelligent vehicle to drive, and send them to the target intelligent vehicle.

[0107] The control commands for the target intelligent vehicle include controlling the target intelligent vehicle to travel at a set speed between two adjacent intersections and to pass through each intersection within a travel time window. The set speed value can be obtained by subtracting the target intelligent vehicle's current time from the minimum time point of the travel time window to obtain the travel time, calculating the distance from the target intelligent vehicle's current position to the intersection, and then dividing by the travel time.

[0108] In this specific embodiment, predictive traffic control is introduced. Based on the traffic conditions of the navigation route predicted in the real-time traffic map, resources on the navigation route are pre-allocated through reservation, upgrading the original control to guidance, avoiding disorderly driving of intelligent vehicles, backlog at key nodes, and indefinite delays in tasks, thus improving the scheduling efficiency and task execution efficiency of intelligent vehicles. Conflicts are avoided by declaring passage time windows at intersections.

[0109] To address the issue of overlapping or conflicting time windows during actual driving of intelligent vehicles, in some specific embodiments, after determining the navigation path for the target intelligent vehicle to execute the target task, the method further includes: Step S209b-1: In the real-time traffic map, when it is determined that the first intelligent vehicle and the second intelligent vehicle are traveling towards the target intersection of the navigation path on different paths, the meeting of the first intelligent vehicle and the second intelligent vehicle is predicted based on their respective current standard driving status information.

[0110] Wherein, one of the first intelligent vehicle and the second intelligent vehicle includes the target intelligent vehicle.

[0111] Standard driving status is a specialized representation of the driving status of intelligent vehicles used in real-time traffic maps. Standard driving status information includes standard driving speed values ​​and standard location information.

[0112] Step S209b-2: When the current standard driving status information of the first intelligent vehicle and the second intelligent vehicle respectively meets the preset conditions for an impending encounter, control commands for the first intelligent vehicle and the second intelligent vehicle to avoid each other are generated based on the current standard driving status information of the first intelligent vehicle and the second intelligent vehicle respectively, the priority of their respective target tasks, and the standard position information of the target intersection point.

[0113] The preset conditions for an impending encounter include determining that the time taken for the first intelligent vehicle and the second intelligent vehicle to reach the target intersection point is the same, based on their respective current positions and current speeds.

[0114] The avoidance control command includes controlling the first intelligent vehicle to pass through the intersection at its standard driving speed within the first standard traffic time window, and controlling the second intelligent vehicle to pass through the intersection at its standard driving speed within the second standard traffic time window.

[0115] The control commands for the first and second intelligent vehicles to avoid obstacles are generated based on their respective current standard driving status information, the priority of their respective target tasks, and the standard position information of the target intersection point. These commands include: Step S209b-21: When the priority of the target task of the first intelligent vehicle is higher than the priority of the target task of the second intelligent vehicle, obtain the current standard driving speed value and current standard position information of the first intelligent vehicle.

[0116] For example, the first intelligent vehicle has the highest priority for its target task, while the second intelligent vehicle has a normal priority for its target task. The first intelligent vehicle's target task has a higher priority than the second intelligent vehicle's target task. The first intelligent vehicle's current standard driving speed is 2 m / s, and its current standard position information is (0, 18). The second intelligent vehicle's current standard driving speed is 3 m / s, and its current standard position information is (12, 0). The standard position information of the target intersection is (12, 18). According to the current information of the first and second intelligent vehicles, the two intelligent vehicles will meet at the target intersection in 6 seconds.

[0117] Step S209b-22: Obtain the current first standard remaining distance value based on the current standard position information of the first intelligent vehicle and the standard position information of the target intersection point. The first standard remaining distance value is the distance between the current position of the first intelligent vehicle and the target intersection.

[0118] For example, continuing the above example, since the vertical coordinate of the current standard position information of the first intelligent vehicle is the same as the vertical coordinate of the standard position information of the target intersection point (i.e., 18), we only need to calculate the difference between the horizontal coordinates of the two positions to obtain the distance between the two positions. The current remaining distance value of the first standard is |12-0|=12m.

[0119] Step S209b-23: Based on the current time point, the current first standard remaining distance value, and the current standard driving speed value of the first intelligent vehicle, obtain the first standard passage time window for the first intelligent vehicle to pass through the target intersection point.

[0120] Specifically, the start time of the first standard passage time window is the current time plus the time taken for the first intelligent vehicle to reach the target intersection, and the end time is the start time plus the preset time taken for the first intelligent vehicle to pass through the target intersection.

[0121] For example, continuing the above example, the current time is 8:00:00, and the time it takes for the first intelligent vehicle to reach the target intersection is 12m / (2m / s) = 6s. The preset passage time is 1 second, so the first standard passage time window of the target intersection is [8:00:06, 8:00:07].

[0122] Steps S209b-24 involve obtaining a second standard passage time window for the second intelligent vehicle to pass through the target intersection point based on the first standard passage time window of the first intelligent vehicle and a preset safe avoidance time interval, and obtaining a current second standard remaining distance value based on the current standard position information of the second intelligent vehicle and the standard position information of the target intersection point. The second standard remaining distance value is the distance between the current position of the second intelligent vehicle and the target intersection.

[0123] For example, continuing the above example, the preset safe avoidance time interval is 1 second, and the second standard passage time window is [8:00:08, 8:00:09]. Since the horizontal coordinate of the current standard position information of the second intelligent vehicle is the same as the horizontal coordinate of the standard position information of the target intersection point (i.e., 12), we only need to calculate the difference between the vertical coordinates of the two positions to obtain the distance between the two positions. The current second standard remaining distance value = |18-0| = 18m.

[0124] Step S209b-25: Obtain the standard driving speed value of the second intelligent vehicle based on the current second standard remaining distance value and the second standard travel time window. That is, the standard driving speed value of the second intelligent vehicle is the quotient of the second standard remaining distance value and the duration of the second standard travel time window.

[0125] For example, continuing the above example, the time it takes for the second intelligent vehicle to reach the target intersection should be 8:00:08 - 8:00:00 = 8s, then the standard driving speed of the second intelligent vehicle should be reduced to 18m / 8s = 2.25 m / s.

[0126] If overlapping or conflicting time windows occur during actual driving, arbitration is conducted based on task priority and a first-come, first-served principle. Higher-priority intelligent vehicles maintain their original driving status, while lower-priority vehicles wait or slow down. The lower-priority vehicle proceeds to the target intersection only after successfully passing through it, or a new route is planned. This avoids the two intelligent vehicles meeting at the target intersection and ensures that the higher-priority vehicle has priority in passing through the target intersection.

[0127] Because all data in the merged map and its derivative maps are represented using a unified standard map data format, which differs from the original map data, the standard control commands generated in the merged map and its derivative maps differ from the proprietary control commands actually used by the intelligent vehicle. Therefore, in some specific embodiments, the generated control commands for the intelligent vehicle include: Step S211: Generate standard control commands for the intelligent vehicle.

[0128] The standard control commands for intelligent vehicles include at least the standard control commands for driving and the standard control commands for avoiding obstacles.

[0129] Step S212: Convert the standard control commands into exclusive control commands that can control the smart car.

[0130] Accordingly, the dedicated control commands for the intelligent vehicle include at least dedicated control commands for driving and dedicated control commands for avoiding obstacles.

[0131] The standard control commands for driving an intelligent vehicle are represented by dedicated driving control commands used in real-time traffic maps. Dedicated driving control commands, on the other hand, are represented by the actual driving control commands used by the target intelligent vehicle. For example, to control an intelligent vehicle to travel straight, the standard control command generated by the group control system based on standard map data is "originating point number 'd001', destination number 'd002', path number 'L001-L003-L005'", while the dedicated driving control command is "originating point number 'F101', destination number 'F203', path number 'R002-R004-R006'". In practical applications, the standard driving control commands issued from the real-time traffic map need to be converted into dedicated driving control commands for the intelligent vehicle to truly control its movement.

[0132] The intelligent vehicle's obstacle avoidance control command is the representation of the dedicated obstacle avoidance control command used by the real-time traffic map; while the dedicated obstacle avoidance control command is the representation of the actual obstacle avoidance control command used by the target intelligent vehicle.

[0133] In some specific embodiments, before step S205, the following steps are also included: Based on the virtual traffic map, a one-to-one conversion relationship is defined between the standard control commands for each working mode and the exclusive control commands for the intelligent vehicle in the corresponding working mode; the conversion of the standard control commands into exclusive control commands that can control the intelligent vehicle includes converting the standard control commands into exclusive control commands according to the conversion relationship.

[0134] The conversion relationships are implemented manually and stored in a conversion relationship dataset. For example, the conversion relationship dataset can be a data table in a database, a worksheet in an Excel workbook, or a configuration file. This specific embodiment is not limited to these. In use, the conversion relationship dataset allows for quick retrieval of definition information to facilitate the conversion between standard control commands and proprietary control commands. For example, continuing the above example, the standard control command for the intelligent vehicle is "origin number 'd001', destination number 'd002', path number 'L001-L003-L005'", while the dedicated driving control command is "origin number 'F101', destination number 'F203', path number 'R002-R004-R006'". The conversion relationship between the standard control command and the dedicated driving control command is stored in the first data table of the first database. In practical applications, when the standard driving control command is issued in the real-time traffic map, the dedicated driving control command for the intelligent vehicle, "origin number 'F101', destination number 'F203', path number 'R002-R004-R006'", is obtained from the first data table based on the standard control command and obtained through the conversion relationship in the first data table. Only then can the intelligent vehicle be truly controlled.

[0135] This specific embodiment abstracts the states, functions, and actions of different navigation modes of intelligent vehicles into standard state machines, standard functions, and standard action interfaces applied to a virtual traffic map. This eliminates application differences, such as differences in working states, control commands, switching commands, and switching conditions. A unified standard information is used to represent these differences in the virtual traffic map, enabling the application layer to call intelligent vehicles with different dedicated navigation modes without distinction. For example, for process intelligent vehicles and logistics intelligent vehicles, the existing system uses the number "1" to represent the idle state for process intelligent vehicles and the number "2" for logistics intelligent vehicles. However, the fused map and its derivative maps uniformly use the letter "a" to represent the idle state. This ensures the reusability of the existing system, breaks down data silos between different dedicated navigation modes, enables the uploading and command distribution of state data in a unified format in the fused map and its derivative maps, and improves the usability of the existing system.

[0136] In some specific embodiments, the standard operating states of the various operating modes of the intelligent vehicle defined based on the virtual traffic map and the intelligent vehicle's capability information include: Step S204a: Define the standard working state of the intelligent vehicle under normal task execution mode based on the virtual traffic map and the capability information of the intelligent vehicle.

[0137] The standard working states under the normal task execution mode include: standard waiting to load state, standard loading state, standard loading ready state, standard driving state, standard waiting to unload state, standard unloading state, and standard unloading ready state.

[0138] For example, the capability information of the intelligent vehicle includes the maximum carrying capacity, maximum driving speed, and minimum driving speed. The embodiments in this application are not limited to this.

[0139] Step S204b: Define the standard working state of the intelligent vehicle in idle standby mode based on the virtual traffic map and the intelligent vehicle's capability information.

[0140] The standard operating states in the idle standby mode include: standard docking state, standard return state, standard low power state, and standard resource ready state.

[0141] Step S204c: Define the standard operating state of the intelligent vehicle in the avoidance mode based on the virtual traffic map and the intelligent vehicle's capability information.

[0142] The standard operating states under the avoidance mode include: standard parking state, standard obstacle avoidance state, standard standby state, and standard recovery state.

[0143] Step S204d: Define the standard working state of the intelligent vehicle in the precise positioning mode based on the virtual traffic map and the capability information of the intelligent vehicle.

[0144] The standard operating states under the precise positioning mode include: standard speed switching state, standard fine-tuning state, and standard mode switching state.

[0145] The information defined based on the virtual traffic map and the intelligent vehicle's capability information is implemented through manual settings. The defined information is stored in a definition dataset, which can be a data table in a database, a worksheet in an Excel workbook, or a configuration file. This specific embodiment is not limited to these. In use, the definition dataset allows for quick retrieval of definition information, facilitating the conversion between information from the virtual traffic map and actual application information.

[0146] In this specific embodiment, multiple operating modes include: normal task execution mode, idle standby mode, avoidance mode, and precise positioning mode. By defining standard operating states for each of the various operating modes of the intelligent vehicle for the virtual traffic map, comprehensive and seamless information exchange between the application layer and the execution layer is enabled, improving the usability of the existing system.

[0147] This application's embodiments generate a virtual traffic map based on the target task and traffic sub-layers; obtain a real-time traffic map through real-time traffic conditions and the virtual traffic map; and finally determine the navigation path for the target intelligent vehicle to perform the target task based on the real-time traffic map. By combining the digitally constructed virtual traffic map with real-time traffic conditions, a high-fidelity digital twin of the production line logistics, synchronized in real time with the physical world, is created. This achieves rapid real-time data linkage and millisecond-level rapid simulation. By sensing the dynamics of the target intelligent vehicle and road changes in real time, data is provided for rapidly predicting traffic changes, improving the efficiency of dynamic perception and decision-making.

[0148] Finally, it should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems or apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and relevant parts can be referred to the method section.

[0149] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for generating fused maps, characterized in that, include: Obtain raw map data for multiple navigation modes, including: linear navigation mode, matrix QR code navigation mode, and LiDAR navigation mode; The original map data of the multiple navigation modes are mapped to multiple logical sub-layers of the fused map, and standard map data is generated in different logical sub-layers, wherein the multiple logical sub-layers include a background sub-layer and a path sub-layer; When the navigation mode is the LiDAR navigation mode, standard map data is generated in the background sub-layer; When the navigation mode is the linear navigation mode or the matrix QR code navigation mode, standard map data is generated in the path sub-layer.

2. The method according to claim 1, characterized in that, When the navigation mode is the LiDAR navigation mode, standard map data is generated in the background sub-layer, including: When the navigation mode is the LiDAR navigation mode, the raster data and / or point cloud data of the LiDAR navigation mode are mapped to the background sub-layer to generate standard map data that constitutes the environment model.

3. The method according to claim 2, characterized in that, When the navigation mode is the linear navigation mode or the matrix QR code navigation mode, generating standard map data in the path sub-layer includes: The original map data of each of the various linear navigation modes and the matrix QR code navigation modes are mapped to the path sub-layer, and standard map data including virtual reference lines and network graphs are generated, wherein the network graphs include key edge information and key node information.

4. The method according to claim 3, characterized in that, Mapping the original map data of the multiple navigation modes to multiple logical sub-layers of the merged map includes: The standard map data in the background sublayer and the path sublayer are mapped to the traffic sublayer, which includes standard map data of various key traffic areas defined based on a preset traffic strategy.

5. The method according to claim 1, characterized in that, After mapping the original map data of the multiple navigation modes to multiple logical sub-layers of the fused map, the process also includes: Based on the scope of the application scenario, determine the standard map data for at least two logical sub-layers among the plurality of logical sub-layers; The standard map data from the at least two logical sub-layers are overlaid to generate a basic application map; Based on the standard map data and functional requirements in the aforementioned basic application map, extract multiple map annotation information and multiple functional annotation information; The basic application map is annotated based on the multiple functional annotation information and the multiple map annotation information to generate a virtual application map.

6. The method according to claim 5, characterized in that, The method further includes: Establish a one-to-one mapping relationship between the standard map data in the at least two logical sub-layers and the standard map data in the virtual application map, as well as the logical relationship between the standard map data in the virtual application map and the map annotation information; When any standard map data in any logical sub-layer changes, the corresponding standard map data in the virtual application map is triggered to change based on the one-to-one mapping relationship, and the corresponding map annotation information is also triggered to change based on the logical relationship.

7. The method according to claim 1, characterized in that, Mapping the original map data of the multiple navigation modes to multiple logical sub-layers of the merged map includes: After aligning the original map data of the multiple navigation modes with coordinates, they are mapped to multiple logical sub-layers of the fused map.

8. A control method for an intelligent vehicle, characterized in that, include: The target task and real-time traffic conditions of the target intelligent vehicle are obtained, as well as the traffic sub-layer of the fused map in the method of claim 4; A virtual traffic map is generated based on the target task and the traffic sub-layer; The real-time traffic conditions are mapped onto the virtual traffic map to obtain the real-time traffic map; The navigation path for the target intelligent vehicle to perform the target task is determined based on the real-time traffic map.

9. The method according to claim 8, characterized in that, The process of generating a virtual traffic map based on the target task and the traffic sub-layer includes: The target standard map data in the traffic sub-layer is determined according to the task scope of the target task, wherein the target standard map data includes standard map data of the environment model related to the target task, standard map data of the virtual reference line and network diagram, and standard map data of various key traffic areas. Based on the target standard map data, a basic traffic map is generated; Based on the target standard map data in the basic traffic map and the target task, extract multiple traffic environment annotation information and multiple traffic control annotation information; The basic traffic map is annotated based on the multiple traffic environment annotations and the multiple traffic control annotations to generate a virtual traffic map.

10. The method according to claim 8, characterized in that, After determining the navigation path for the target intelligent vehicle to perform the target task, the method further includes: When the target smart car is in the standard pickup ready state, traffic conditions are predicted for the navigation route based on the real-time traffic map. Based on the prediction results and the supply time window in the target task, the passage time window reservation is made for the intersection points in the navigation path, and the passage time window of each successfully reserved intersection point is declared. Based on the passage time window of each intersection point in the navigation path, a control command for the target intelligent vehicle is generated and sent to the target intelligent vehicle. The control command for the target intelligent vehicle includes controlling the target intelligent vehicle to travel at a set speed between two adjacent intersection points and to pass through each intersection point within the passage time window.

11. The method according to claim 8, characterized in that, After determining the navigation path for the target intelligent vehicle to perform the target task, the method further includes: In the real-time traffic map, when it is determined that the first intelligent vehicle and the second intelligent vehicle are heading towards the target intersection of the navigation path via different paths, the meeting of the first intelligent vehicle and the second intelligent vehicle is predicted based on their respective current standard driving status information, wherein one of the first intelligent vehicle and the second intelligent vehicle is the target intelligent vehicle; When the current standard driving status information of the first intelligent vehicle and the second intelligent vehicle respectively meets the preset impending encounter condition, control commands for the first intelligent vehicle and the second intelligent vehicle to avoid each other are generated based on their respective current standard driving status information, the priority of their respective target tasks, and the standard position information of the target intersection point. These commands include: When the priority of the target task of the first intelligent vehicle is higher than the priority of the target task of the second intelligent vehicle, the current standard driving speed value and current standard position information of the first intelligent vehicle are obtained. The current standard remaining distance value is obtained based on the current standard location information of the first intelligent vehicle and the standard location information of the target intersection point; Based on the current time point, the current first standard remaining distance value, and the current standard driving speed value of the first intelligent vehicle, a first standard passage time window is obtained for the first intelligent vehicle to pass through the target intersection point; The second standard passage time window for the second intelligent vehicle to pass through the target intersection point is obtained based on the first standard passage time window of the first intelligent vehicle and the preset safe avoidance time interval. The current standard distance value of the second intelligent vehicle is obtained based on the current standard location information of the second intelligent vehicle and the standard location information of the target intersection point; The standard driving speed of the second intelligent vehicle is obtained based on the current remaining distance value of the second standard and the second standard travel time window. The avoidance control command includes controlling the first intelligent vehicle to pass through the intersection at its standard driving speed within the first standard traffic time window, and controlling the second intelligent vehicle to pass through the intersection at its standard driving speed within the second standard traffic time window.

12. The method according to claim 10 or 11, characterized in that, Generate control commands for the intelligent vehicle, including: Generate standard control commands for the intelligent vehicle; The standard control commands are converted into dedicated control commands that can control the smart car.

13. The method according to claim 12, characterized in that, Before obtaining the real-time traffic map based on the real-time traffic conditions and the virtual traffic map, the method further includes: Based on the virtual traffic map, a one-to-one conversion relationship is defined between the standard control commands for each working mode and the exclusive control commands for the intelligent vehicle in the corresponding working mode; The step of converting the standard control commands into dedicated control commands that can control the smart car includes converting the standard control commands into dedicated control commands according to the conversion relationship.

Images