Map-based intelligent driving implementation method and device

Abstract

The embodiment of the invention provides an intelligent driving implementation method and device based on a map, a vehicle, a storage medium and a program product, and particularly relates to the technical field of traffic information processing. The method comprises the steps that a vehicle is controlled to conduct auxiliary driving based on lightweight map data obtained from a cloud end, the lightweight map data is lightweight high-precision map data, the lightweight map data provides key map elements needed by auxiliary driving, and the key map elements comprise lane boundaries; when the vehicle is in the auxiliary driving state, the vehicle is controlled to keep driving in the lane based on the lane boundary provided by the lightweight map data, and vehicle control is achieved depending on the lane boundary instead of the lane center line. The method is used for achieving the effect of meeting the requirements of automatic driving and advanced driving assistance systems for the precision and freshness of map data.

Description

Technical Field

[0001] The present application relates to the technical field of traffic information processing, and in particular to a method, device, vehicle, storage medium and program product for realizing map-based intelligent driving. Background Art

[0002] The design of high-definition maps (HD Maps) pursues high precision and rich geographical elements to meet the needs of autonomous driving and advanced driver assistance systems. Among them, high-precision surveying and mapping-level mobile measurement equipment, high-beam laser radar, panoramic imaging equipment, combined inertial navigation systems, etc. can provide high-precision raw data collection. By combining point cloud data and image data, real-world geographical elements such as lane markings, traffic signs, etc. can be abstracted, and a large number of hidden elements can be added manually based on on-site conditions and professional experience. These elements may include information on vulnerable road users (Vulnerable Road Users, VRU), signs of fully enclosed roads, etc., to meet the needs of functional safety and safety redundancy.

[0003] In the field of intelligent driving, the "freshness" or real-time nature of data is becoming increasingly important. With the development of autonomous driving and assisted driving technologies, vehicles need to rely on the latest map data to make accurate decisions. If there is a deviation between the high-precision map and the real-time perception data of the vehicle, it may increase the complexity of the driving algorithm and affect the accuracy of its judgment. The current update speed of high-precision maps is difficult to meet the needs of autonomous driving and advanced driver assistance systems for data real-time and freshness. Summary of the invention

[0004] The embodiments of the present application provide a map-based intelligent driving implementation method, device, vehicle, storage medium and program product to achieve the effect of meeting the accuracy and freshness requirements of autonomous driving and advanced driver assistance systems for map data.

[0005] In a first aspect, an embodiment of the present application provides a map-based intelligent driving implementation method, the method comprising:

[0006] Controlling the vehicle to perform assisted driving based on lightweight map data obtained from the cloud, wherein the lightweight map data is lightweight high-precision map data, and the lightweight map data provides key map elements required for assisted driving, wherein the key map elements include lane boundaries;

[0007] When the vehicle is in the assisted driving state, the vehicle is controlled to keep traveling in the lane based on the lane boundary provided by the lightweight map data, so that the vehicle is controlled by relying on the lane boundary instead of the lane centerline.

[0008] In a possible implementation manner, controlling the vehicle to keep traveling in the lane based on the lane boundary provided by the lightweight map data includes:

[0009] During vehicle driving, extracting a lane boundary of a lane in which the vehicle is currently driving from the lightweight map data;

[0010] Selecting a reference boundary from the lane boundary of the current driving lane, wherein the reference boundary is a reference boundary for controlling the driving trajectory of the vehicle;

[0011] The distance between the vehicle and the reference boundary is controlled to be within a preset distance range, so as to control the vehicle to keep traveling within the lane.

[0012] In a possible implementation, the method further includes:

[0013] When the vehicle is in the assisted driving state, when the planned lane-level driving path indicates that a vehicle lane change is required ahead, determining a current driving lane of the vehicle and a target lane for lane change, wherein the lane-level driving path is generated by lane-level path planning based on the lightweight map data;

[0014] Extracting from the lightweight map data the lane boundary of the current driving lane, the lane boundary of the target lane, and a variable lane position point between the current driving lane and the target lane, the variable lane position point being a position where the lane boundary between the current driving lane and the target lane changes from a solid line to a dotted line;

[0015] Based on the extracted lane-changeable position point, the lane boundary of the current driving lane and the lane boundary of the target lane, a lane-changeable area is defined within the lane area range of the current driving lane and the target lane, wherein the lane-changeable area refers to an area where a vehicle is allowed to change lanes;

[0016] The vehicle is controlled to change lanes within the lane-changeable area.

[0017] In a possible implementation, the method further includes:

[0018] When the vehicle is in the assisted driving state, when the planned lane-level driving path indicates that an intersection needs to be passed ahead, determining the current driving lane of the vehicle and a target entry lane that is allowed to be entered after passing the intersection, wherein the target entry lane includes an outermost lane and an innermost lane that can be entered;

[0019] extracting a lane boundary of the current driving lane, a lane boundary of the outermost lane, and a lane boundary of the innermost lane from the lightweight map data;

[0020] Based on the extracted lane boundary of the current driving lane, the lane boundary of the outermost lane, and the lane boundary of the innermost lane, a traversable area of ​​the intersection is defined within the intersection area, wherein the traversable area of ​​the intersection refers to an area where vehicles are allowed to travel and can safely pass through the intersection;

[0021] The vehicle is controlled to pass through the intersection within a passable area of ​​the intersection.

[0022] In a possible implementation, the method further includes:

[0023] When the vehicle is in the assisted driving state, it is predicted that a side of the vehicle will collide with a front obstacle when the vehicle is traveling along a planned lane-level driving path, and a lane boundary of the current driving lane is extracted from the lightweight map data;

[0024] Based on the extracted lane boundary of the current driving lane and in combination with the detected obstacle position information, a narrow road passable area is defined within the lane area of ​​the current driving lane, wherein the narrow road passable area refers to an area where vehicles are allowed to travel without colliding with obstacles;

[0025] The vehicle is controlled to travel within the narrow road passable area to avoid collision with obstacles.

[0026] In a possible implementation, the method further includes:

[0027] When the vehicle is in the assisted driving state, detecting an actual boundary of a current driving lane through a vehicle sensor;

[0028] extracting a lane boundary of the current driving lane from the lightweight map data;

[0029] Comparing the detected actual boundary with the extracted lane boundary of the current driving lane to obtain a comparison result;

[0030] Based on the comparison result, it is determined whether to generate reminder information, where the reminder information is used to remind the user to take over the vehicle.

[0031] In a possible implementation, the method further includes:

[0032] Vectorize the environmental perception data detected by the vehicle sensor to generate vehicle-side vector data;

[0033] The vehicle-side vector map data is sent to the cloud, and the cloud is used to extract the map elements of the target section based on the multiple vehicle-side vector data after acquiring the multiple vehicle-side vector data of the target section, and update the lightweight map data of the target section based on the extracted map elements.

[0034] In a second aspect, an embodiment of the present application provides a map-based intelligent driving implementation device, the device comprising:

[0035] An assisted driving module, used to control the vehicle to perform assisted driving based on lightweight map data obtained from the cloud, wherein the lightweight map data is lightweight high-precision map data, and the lightweight map data provides key map elements required for assisted driving, wherein the key map elements include lane boundaries;

[0036] A control module is used to control the vehicle to keep driving in the lane based on the lane boundary provided by the lightweight map data when the vehicle is in the assisted driving state, so as to realize vehicle control relying on the lane boundary instead of the lane center line.

[0037] In a possible implementation manner, the control module is specifically used to:

[0038] During vehicle driving, extracting a lane boundary of a lane in which the vehicle is currently driving from the lightweight map data;

[0039] Selecting a reference boundary from the lane boundary of the current driving lane, wherein the reference boundary is a reference boundary for controlling the driving trajectory of the vehicle;

[0040] The distance between the vehicle and the reference boundary is controlled to be within a preset distance range, so as to control the vehicle to keep traveling within the lane.

[0041] In a possible implementation manner, the map-based intelligent driving implementation device is further used to:

[0042] When the vehicle is in the assisted driving state, when the planned lane-level driving path indicates that a vehicle lane change is required ahead, determining a current driving lane of the vehicle and a target lane for lane change, wherein the lane-level driving path is generated by lane-level path planning based on the lightweight map data;

[0043] Extracting from the lightweight map data the lane boundary of the current driving lane, the lane boundary of the target lane, and a variable lane position point between the current driving lane and the target lane, the variable lane position point being a position where the lane boundary between the current driving lane and the target lane changes from a solid line to a dotted line;

[0044] Based on the extracted lane-changeable position point, the lane boundary of the current driving lane and the lane boundary of the target lane, a lane-changeable area is defined within the lane area range of the current driving lane and the target lane, wherein the lane-changeable area refers to an area where a vehicle is allowed to change lanes;

[0045] The vehicle is controlled to change lanes within the lane-changeable area.

[0046] In a possible implementation manner, the map-based intelligent driving implementation device is further used to:

[0047] When the vehicle is in the assisted driving state, when the planned lane-level driving path indicates that an intersection needs to be passed ahead, determining the current driving lane of the vehicle and a target entry lane that is allowed to be entered after passing the intersection, wherein the target entry lane includes an outermost lane and an innermost lane that can be entered;

[0048] extracting a lane boundary of the current driving lane, a lane boundary of the outermost lane, and a lane boundary of the innermost lane from the lightweight map data;

[0049] Based on the extracted lane boundary of the current driving lane, the lane boundary of the outermost lane, and the lane boundary of the innermost lane, a traversable area of ​​the intersection is defined within the intersection area, wherein the traversable area of ​​the intersection refers to an area where vehicles are allowed to travel and can safely pass through the intersection;

[0050] The vehicle is controlled to pass through the intersection within a passable area of ​​the intersection.

[0051] In a possible implementation manner, the map-based intelligent driving implementation device is further used to:

[0052] When the vehicle is in the assisted driving state, it is predicted that a side of the vehicle will collide with a front obstacle when the vehicle is traveling along a planned lane-level driving path, and a lane boundary of the current driving lane is extracted from the lightweight map data;

[0053] Based on the extracted lane boundary of the current driving lane and in combination with the detected obstacle position information, a narrow road passable area is defined within the lane area of ​​the current driving lane, wherein the narrow road passable area refers to an area where vehicles are allowed to travel without colliding with obstacles;

[0054] The vehicle is controlled to travel within the narrow road passable area to avoid collision with obstacles.

[0055] In a possible implementation manner, the map-based intelligent driving implementation device is further used to:

[0056] When the vehicle is in the assisted driving state, detecting an actual boundary of a current driving lane through a vehicle sensor;

[0057] extracting a lane boundary of the current driving lane from the lightweight map data;

[0058] Comparing the detected actual boundary with the extracted lane boundary of the current driving lane to obtain a comparison result;

[0059] Based on the comparison result, it is determined whether to generate reminder information, where the reminder information is used to remind the user to take over the vehicle.

[0060] In a possible implementation manner, the map-based intelligent driving implementation device is further used to:

[0061] Vectorize the environmental perception data detected by the vehicle sensor to generate vehicle-side vector data;

[0062] The vehicle-side vector map data is sent to the cloud, and the cloud is used to extract the map elements of the target section based on the multiple vehicle-side vector data after acquiring the multiple vehicle-side vector data of the target section, and update the lightweight map data of the target section based on the extracted map elements.

[0063] In a third aspect, an embodiment of the present application provides a vehicle, including: a memory, a processor;

[0064] The memory stores computer-executable instructions;

[0065] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0066] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementations of the first aspect.

[0067] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0068] The map-based intelligent driving implementation method, device, vehicle, storage medium and program product provided by the embodiment of the present application can obtain lightweight map data from the cloud, and the vehicle can be controlled to perform assisted driving based on the lightweight map data, wherein the lightweight map data is still a kind of high-precision map data, which can provide key map elements required for assisted driving, and the key map elements include lane boundaries. When the vehicle is in the assisted driving state, the lane boundaries provided by the lightweight map data can control the vehicle to keep driving in the lane, that is, when the vehicle is in the assisted driving state, the system uses the lane boundary information in the lightweight map data to control the vehicle. This method relies on lane boundaries, rather than traditional lane centerlines, to control the vehicle. Lightweight map data provides detailed geographic information at the lane level, which is crucial for autonomous driving and advanced driver assistance systems. Compared with standard definition maps and high-precision maps, lightweight maps strike a balance between data volume and details. Lightweight maps significantly reduce the amount of data by streamlining unnecessary map elements. Despite the small amount of data, lightweight maps still retain the key details required for autonomous driving and advanced driver assistance systems. By streamlining map elements, lightweight maps can be updated faster to ensure the freshness of map data. This is crucial to meeting the real-time and accuracy requirements of autonomous driving and advanced driver assistance systems for map data. This method achieves the effect of meeting the accuracy and freshness requirements of autonomous driving and advanced driver assistance systems for map data by optimizing the structure and content of map data. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0070] Figure 1 A schematic diagram of a scenario for a map-based intelligent driving implementation method provided in this application;

[0071] Figure 2 Schematic diagram of the map-based intelligent driving implementation method provided in this application Figure 1 ;

[0072] Figure 3 Schematic diagram of the map-based intelligent driving implementation method provided in this application Figure 2 ;

[0073] Figure 4 A schematic diagram comparing the lightweight map provided for this application with the standard definition map;

[0074] Figure 5 A schematic diagram of the data structure of the lightweight map provided for this application;

[0075] Figure 6Schematic diagram of the difference in vehicle control between the lightweight map and the traditional high-precision map in the lane change scenario provided by this application;

[0076] Figure 7 Schematic diagram of the difference in vehicle control between the lightweight map and the traditional high-precision map in the intersection scenario provided by this application;

[0077] Figure 8 Schematic diagram of the difference in vehicle control between the lightweight map and the traditional high-precision map in the narrow road traffic scenario provided by this application;

[0078] Fig. 9 Schematic diagram of the difference in the expression of road obstacles between the lightweight map provided for this application and the traditional high-precision map;

[0079] Fig.10 Schematic diagram of the mapping process of the lightweight map provided for this application;

[0080] Fig.11 A schematic diagram of the structure of the map-based intelligent driving implementation device provided in this application;

[0081] Fig.12 A schematic diagram of the structure of the vehicle provided for this application.

[0082] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0083] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0084] In the process of building high-precision maps, both point cloud data and image data play a vital role. Point cloud data is a set of three-dimensional coordinate points obtained by scanning the surrounding environment through sensors such as lidar. These point cloud data can accurately reflect the geometry and structure of the road and its surrounding environment, and are one of the basic data for high-precision map production. Image data mainly comes from visual sensors such as cameras. By capturing images of the road and its surrounding environment, key information such as lane lines, traffic signs, and traffic lights can be extracted to provide rich semantic information for high-precision maps.

[0085] The design of high-precision maps pursues high accuracy and rich geographical elements, but there are the following disadvantages:

[0086] 1. High production cost: The production of high-precision maps requires a large number of mobile surveying equipment, data collection personnel, and in-house cartographers. The investment of these equipment and human resources leads to high production costs.

[0087] 2. Large data volume: High-precision maps are very detailed in their abstract expression of the real world, which is specifically reflected in the large number of elements and the fine classification of elements. The large number of elements is mainly because high-precision maps express many entities that actually exist in the real world, such as traffic signs, tree trunks, light poles, bridge piers, tunnel entrances and exits, no-parking zones, etc. High-precision maps need to express the geometry and attribute information of these geographic elements. The fine classification of elements is mainly because high-precision maps express a large number of fine classifications at the attribute level based on the actual situation on site. For example, road attributes express whether the road is fully closed, whether there are vulnerable road users such as pedestrians and bicycles on the road, and lane boundaries and road boundaries will provide specific detailed classifications such as deceleration markings, short and thick dashed lines, guardrails, curbs, walls, etc., in order to restore the real world as much as possible.

[0088] 3. Low data coverage: The coverage of highways is high, but the coverage of urban roads is still insufficient. And because urban roads are complex and changeable, these factors increase the difficulty of high-precision map production.

[0089] 4. Low freshness: In the field of intelligent driving, the "freshness" or real-time nature of data is becoming increasingly important. If there is a deviation between the high-precision map and the real-time perception data of the vehicle, it may increase the complexity of the driving algorithm and affect the accuracy of its judgment. Maintaining the freshness of map data at the daily and weekly levels is crucial to the stability of the intelligent driving algorithm, but the current update cycle of high-precision maps is long, which does not match the requirements of intelligent driving for data real-time and freshness.

[0090] The map-based intelligent driving implementation method provided by the present application can obtain lightweight map data from the cloud, and the vehicle can be controlled for assisted driving based on the lightweight map data, wherein the lightweight map data is still a kind of high-precision map data, which can provide the key map elements required for assisted driving, and the key map elements include lane boundaries. When the vehicle is in the assisted driving state, the lane boundaries provided by the lightweight map data can control the vehicle to keep driving in the lane, that is, when the vehicle is in the assisted driving state, the system uses the lane boundary information in the lightweight map data to control the vehicle. This method relies on lane boundaries, rather than traditional lane centerlines, to control the vehicle. Lightweight map data provides detailed geographic information at the lane level, which is essential for autonomous driving and advanced driver assistance systems. Compared with standard definition maps and high-precision maps, lightweight maps strike a balance between data volume and details. Lightweight maps significantly reduce the amount of data by streamlining unnecessary map elements. Despite the small amount of data, lightweight maps still retain the key details required for autonomous driving and advanced driver assistance systems. By streamlining map elements, lightweight maps can be updated faster to ensure the freshness of map data. This is crucial to meet the real-time and accuracy requirements of map data for autonomous driving and advanced driver assistance systems. By optimizing the structure and content of map data, this method solves the technical problem that the current update speed of high-precision maps cannot meet the real-time and freshness requirements of autonomous driving and advanced driver assistance systems.

[0091] Figure 1 A schematic diagram of a scenario of a map-based intelligent driving implementation method provided in this application, such as Figure 1 As shown, in this application scenario, vehicle 101 uses visual sensors such as a driving recorder to shoot the road and the surrounding environment to obtain environmental perception data. The collected environmental perception data can be vectorized to generate vehicle-side vector data. Then, the vehicle can send the generated vehicle-side vector data to the cloud 102. After the cloud obtains multiple vehicle-side vector data for the same road section, it can use these vehicle-side vector data to generate map elements of the road section, and then update the lightweight map data of the road section, and then send the updated lightweight map data to other vehicle applications.

[0092] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0093] Figure 2 Schematic diagram of the map-based intelligent driving implementation method provided in this application Figure 1,like Figure 2 As shown, the method may include:

[0094] S201. Control the vehicle for assisted driving based on lightweight map data obtained from the cloud. The lightweight map data is lightweight high-precision map data. The lightweight map data provides key map elements required for assisted driving. The key map elements include lane boundaries.

[0095] Lightweight map data, as the name implies, is map data for lightweight maps. Although lightweight, these map data are still highly accurate and can provide key map elements required for assisted driving. Key map elements may include or only include roads, lane groups, lanes, lane boundaries, traffic lights, and ground markings. Ground markings specifically refer to signs and markings drawn directly on the road surface, such as arrow signs, stop lines, and crosswalk signs. Compared with traditional high-precision maps, lightweight maps have smaller data volumes, reducing the burden of data transmission and processing. Due to the small amount of data, lightweight maps can be updated more frequently to ensure the real-time and accuracy of the data.

[0096] In one possible implementation, the map elements in the lightweight map data are observable elements in the real world, that is, those that can be directly observed by the vehicle's visual sensors, such as lanes, lane boundaries, etc., and do not include unobservable elements. For example, the center line of a lane is usually virtual or inferred and is not directly visible to the visual sensor, and therefore is not included in the lightweight map data.

[0097] In another possible implementation, lightweight map data focuses on expressing feature objects that exist on the road, and does not express roadside objects and elevation objects. Roadside objects (such as roadside traffic signs, tree trunks, guardrails, light poles, etc.) and elevation objects (such as traffic signs above the road, overpasses, bridges, etc.) are not within the expression scope of lightweight map data. This simplification helps reduce the amount and complexity of data. In addition, although traffic lights are three-dimensional objects, they are simplified to two-dimensional objects in lightweight maps. This conversion reduces data complexity while still providing necessary navigation information.

[0098] Compared with high-precision maps, lightweight map data has a smaller data volume, a faster update speed, and can provide key map elements required for assisted driving. The data accuracy of high-precision maps can reach the centimeter level, while the data accuracy of lightweight maps can be sub-meter level. The design goal of lightweight maps is to reduce the data volume and processing complexity while ensuring a certain accuracy, so as to increase the map update speed. The number of map elements provided by lightweight maps is less than that of high-precision maps. For example, high-precision maps may provide hundreds of map elements, while the number of map elements provided by lightweight maps can be less, such as no more than one hundred. In addition, high-precision maps and lightweight maps can also be distinguished by their specific application scenarios and coverage areas. High-precision maps are mainly used on highways, while lightweight maps mainly cover urban roads.

[0099] Assisted driving, also known as automatic navigation assisted driving, automatic intelligent driving, intelligent driving, assisted automatic driving, etc., refers to the use of map data and sensor data, combined with the on-board navigation route, to allow the vehicle to automatically change lanes, automatically enter and exit ramps on the way to the destination. The driver only needs to set the starting and end points of the vehicle, and the vehicle can automatically reach the destination and complete functions such as driving, overtaking, changing lanes, and turning.

[0100] In an embodiment of the present application, assisted driving is performed based on the acquired lightweight map data.

[0101] S202: When the vehicle is in the assisted driving state, the lane boundary provided by the lightweight map data is used to control the vehicle to stay within the lane, thereby achieving vehicle control based on the lane boundary rather than the lane centerline.

[0102] In the assisted driving state, the lane boundary information in the lightweight map data is used to control the vehicle to stay in the lane. The vehicle uses the lane boundary information in the lightweight map data as a reference instead of the traditional lane centerline. This method is closer to the actual road environment because the lane boundary is usually directly detectable by the visual sensor.

[0103] The map-based intelligent driving implementation method provided by the embodiment of the present application can obtain lightweight map data from the cloud, and the vehicle can be controlled to perform assisted driving based on the lightweight map data, wherein the lightweight map data is still a kind of high-precision map data, which can provide key map elements required for assisted driving, and the key map elements include lane boundaries. When the vehicle is in the assisted driving state, the lane boundaries provided by the lightweight map data can control the vehicle to keep driving in the lane, that is, when the vehicle is in the assisted driving state, the system uses the lane boundary information in the lightweight map data to control the vehicle. This method relies on lane boundaries, rather than traditional lane centerlines, to control the vehicle. Lightweight map data provides detailed geographic information at the lane level, which is crucial for autonomous driving and advanced driver assistance systems. Compared with standard definition maps and high-precision maps, lightweight maps strike a balance between data volume and details. Lightweight maps significantly reduce the amount of data by streamlining unnecessary map elements. Despite the small amount of data, lightweight maps still retain the key details required for autonomous driving and advanced driver assistance systems. By streamlining map elements, lightweight maps can be updated faster to ensure the freshness of map data. This is crucial to meet the real-time and accuracy requirements of autonomous driving and advanced driver assistance systems for map data. This method achieves the effect of meeting the accuracy and freshness requirements of autonomous driving and advanced driver assistance systems for map data by optimizing the structure and content of map data.

[0104] Figure 3 Schematic diagram of the map-based intelligent driving implementation method provided in this application Figure 2 ,like Figure 3 As shown, in this embodiment Figure 2 Based on the embodiment, a map-based intelligent driving implementation method is described in detail, and the method may include:

[0105] S301. Control the vehicle for assisted driving based on lightweight map data obtained from the cloud. The lightweight map data is lightweight high-precision map data. The lightweight map data provides key map elements required for assisted driving. The key map elements include lane boundaries.

[0106] Reference Figure 4As shown, a schematic diagram comparing the lightweight map and the standard definition map provided in this application can intuitively understand the difference between the two in expressing the real world. As shown in the figure, the lightweight map provides lane-level geometry and attribute information, which means that every lane, lane boundary and ground marking in the real world is described in detail, and attribute information such as lane type is provided. Standard Definition Map (SD Map) or its variations, such as SD Pro (Standard Definition Map Professional), SD Plus (Standard Definition Map Plus), etc., are usually expressed in units of roads, mainly showing the centerline of the road, and marking key points on the centerline of the road to provide attribute information related to the road, and do not provide detailed information at the lane level.

[0107] In lightweight maps, roads are composed of one or more lane groups. Lane groups are subdivision units of roads, usually used to distinguish sets of lanes with different directions or functions. Lane groups are further subdivided into multiple lanes. Each lane has a clear position and order in a lane group. The topological relationship between lanes describes the relative position and connection method of lanes, including lane merging, separation, and intersection. Lane boundaries define the physical boundaries of lanes, including solid lines, dotted lines, etc. The topological relationship between lanes and lane boundaries describes the legal driving area of ​​​​the lane. The topological relationship between lanes and traffic lights describes which traffic lights control the lanes.

[0108] Reference Figure 5 As shown, this is a schematic diagram of the data structure of the lightweight map provided by this application, which expresses the real-world roads, lanes, ground signs and traffic lights through a road / lane model layer, an independent ground sign layer and a traffic light layer. This structure supports the expression of detailed lane-level information, ground signs and traffic lights, meeting the needs of autonomous driving and advanced driver assistance systems.

[0109] S302: When the vehicle is in an assisted driving state and during the driving of the vehicle, a lane boundary of a lane in which the vehicle is currently traveling is extracted from lightweight map data.

[0110] When the vehicle is in assisted driving state, the lane boundary of the vehicle's current lane can be extracted from the lightweight map data.

[0111] S303: Select a reference boundary from the lane boundary of the current driving lane, where the reference boundary is a reference boundary for controlling the driving trajectory of the vehicle.

[0112] A reference boundary is selected from the lane boundaries of the current driving lane. This reference boundary will be used as a basis for controlling the vehicle's driving trajectory. The selection of the reference boundary may be based on multiple factors, such as lane width, road curvature, and traffic conditions.

[0113] S304. Control the distance between the vehicle and the reference boundary to be within a preset distance range, so as to control the vehicle to stay within the lane, and realize vehicle control based on the lane boundary rather than the lane centerline.

[0114] The distance between the vehicle and the reference boundary is controlled to be maintained within a preset distance range. This ensures that the vehicle can stably stay in the lane. This method relies on the lane boundary rather than the lane centerline for vehicle control, which improves the adaptability in complex or unclear road environments.

[0115] In a possible implementation, the map-based intelligent driving implementation method provided in the embodiment of the present application may further include:

[0116] When the vehicle is in the assisted driving state, when the planned lane-level driving path indicates that the vehicle needs to change lanes ahead, the vehicle's current driving lane and the target lane for lane change are determined. The lane-level driving path is generated by lane-level path planning based on lightweight map data;

[0117] Extracting the lane boundary of the current driving lane, the lane boundary of the target lane, and a variable lane position point between the current driving lane and the target lane from the lightweight map data, wherein the variable lane position point is a position where the lane boundary between the current driving lane and the target lane changes from a solid line to a dotted line;

[0118] Based on the extracted lane-changing position point, the lane boundary of the current driving lane and the lane boundary of the target lane, a lane-changing area is defined within the lane area of ​​the current driving lane and the target lane. The lane-changing area refers to an area where the vehicle is allowed to change lanes.

[0119] Control the vehicle to change lanes within the lane-changeable area.

[0120] Reference Figure 6As shown, it is a schematic diagram of the vehicle control difference between the lightweight map and the traditional high-precision map in the lane change scenario provided by this application. Among them, the left picture is a real scene, the vehicle is driving on a ramp with a large curvature. At this time, the vehicle cannot perceive the changes in the lane ahead through the visual sensor, and needs to rely on map data for lane change planning. The right picture compares the lane change planning differences between the traditional high-precision map and the lightweight map provided by this application. The traditional high-precision map provides a connection relationship between the center lines of the lanes. The vehicle relies on the center line of the lane to change lanes. It drives along the center line of the lane of the current driving lane to the end of the lane center line before planning the lane change. The planned variable lane area (the lane area range corresponding to the red lane center line in the figure) is small, and it is difficult to change lanes when the road traffic is congested. With the lightweight map provided by this application, the vehicle can determine the variable lane position point between the current driving lane of the vehicle and the target lane of the lane change. The variable lane position point is the position where the lane boundary between the current driving lane and the target lane changes from a solid line to a dotted line. The variable lane area is determined based on the position point. The variable lane area planned by this method is large, and the vehicle lane change is more flexible.

[0121] In a possible implementation, the map-based intelligent driving implementation method provided in the embodiment of the present application may further include:

[0122] When the vehicle is in the assisted driving state, when the planned lane-level driving path indicates that an intersection is required ahead, the vehicle's current driving lane and the target entry lane that is allowed after passing the intersection are determined. The target entry lane includes the outermost lane and the innermost lane that can be entered;

[0123] Extracting a lane boundary of a current driving lane, a lane boundary of an outermost lane, and a lane boundary of an innermost lane from the lightweight map data;

[0124] Based on the extracted lane boundaries of the current driving lane, the lane boundaries of the outermost lane, and the lane boundaries of the innermost lane, a passable area of ​​the intersection is defined within the intersection area. The passable area of ​​the intersection refers to an area where vehicles are allowed to travel and can safely pass through the intersection.

[0125] Control vehicles to pass through the intersection within the passable area of ​​the intersection.

[0126] The target entry lanes are all exit lanes that the vehicle is allowed to enter.

[0127] Among them, if the vehicle turns left through the intersection, the innermost lane allowed to enter is the leftmost lane of the target entry lane, and the outermost lane allowed to enter is the rightmost lane of the target entry lane; if the vehicle turns right through the intersection, the innermost lane allowed to enter is the rightmost lane of the target entry lane, and the outermost lane allowed to enter is the leftmost lane of the target entry lane; if the vehicle goes straight through the intersection, the innermost lane allowed to enter is the leftmost lane of the target entry lane, and the outermost lane allowed to enter is the rightmost lane of the target entry lane; if the vehicle makes a U-turn through the intersection, the innermost lane allowed to enter is the leftmost lane of the target entry lane, and the outermost lane allowed to enter is the rightmost lane of the target entry lane. The outermost lane and innermost lane are different depending on the vehicle's driving direction.

[0128] Reference Figure 7 As shown, it is a schematic diagram of the vehicle control difference between the lightweight map and the traditional high-precision map in the intersection scenario provided by this application, wherein the left figure is a real scene, the vehicle turns left at the intersection, and there is no marking on the scene after leaving the waiting area, and the vehicle needs to rely on map data to complete the left turn. The right figure compares the intersection planning differences between the traditional high-precision map and the lightweight map provided by this application. The traditional high-precision map provides the connection relationship between the center lines of the lanes in the intersection. The vehicle relies on the center line of the lane to pass through the intersection. It is usually recommended to drive into the innermost lane. However, if the innermost lane to be driven into is narrowed, this situation may cause the vehicle to need to change lanes immediately after passing the intersection, and the driving experience is not good. With the lightweight map provided by this application, the vehicle can extract the lane boundary of the current driving lane, the lane boundary of the outermost lane that can be driven into, and the lane boundary of the innermost lane from the lightweight map data, and define a passable area of ​​the intersection based on the extracted lane boundary of the current driving lane, the lane boundary of the outermost lane, and the lane boundary of the innermost lane. The drivable area of ​​the intersection leads to all exit lanes that the vehicle can enter after passing through the intersection. Based on the lane width and traffic conditions of each exit lane, the optimal exit lane can be selected as the target exit lane to reduce the subsequent lane change requirements. The intersection traffic path is generated based on the selected target exit lane, and the vehicle is controlled to pass through the intersection according to the intersection traffic path. This method can improve the autonomous driving experience.

[0129] In a possible implementation, the map-based intelligent driving implementation method provided in the embodiment of the present application may further include:

[0130] When the vehicle is in the assisted driving state, it is predicted that the side of the vehicle will collide with the obstacle in front while driving along the planned lane-level driving path, and the lane boundary of the current driving lane is extracted from the lightweight map data;

[0131] Based on the extracted lane boundary of the current driving lane and the detected obstacle position information, a narrow road passable area is defined within the lane area of ​​the current driving lane. The narrow road passable area refers to an area where vehicles are allowed to travel without colliding with obstacles.

[0132] Control the vehicle to travel within the passable area of ​​narrow roads to avoid collision with obstacles.

[0133] Reference Figure 8 As shown, it is a schematic diagram of the difference in vehicle control between the lightweight map and the traditional high-precision map in the narrow road traffic scenario provided by this application. Among them, the left picture is a real scene, and the vehicle in front occupies part of the current driving lane. At this time, it is necessary to combine the map data and environmental perception data to plan the narrow road traffic. The right picture compares the difference in narrow road traffic planning between the traditional high-precision map and the lightweight map provided by this application. The traditional high-precision map provides the center line of the current driving lane. The vehicle relies on the center line of the lane to drive, which may cause a collision accident, or the vehicle exits the assisted driving state, or the vehicle stops here until the vehicle in front leaves. With the lightweight map provided by this application, the vehicle can define a narrow road passable area based on the lane boundary of the current driving lane extracted from the lightweight map data, combined with the detected obstacle location information. Furthermore, a narrow road pass path can be generated in the narrow road passable area based on the vehicle body width, and the vehicle is controlled to drive along the narrow road pass path to avoid collision with obstacles.

[0134] In a possible implementation, the map-based intelligent driving implementation method provided in the embodiment of the present application may further include:

[0135] When the vehicle is in the assisted driving state, the actual boundary of the current driving lane is detected by the vehicle sensor;

[0136] Extracting the lane boundary of the current driving lane from the lightweight map data;

[0137] Comparing the detected actual boundary with the extracted lane boundary of the current driving lane to obtain a comparison result;

[0138] Based on the comparison result, it is determined whether to generate a reminder message, and the reminder message is used to remind the user to take over the vehicle.

[0139] Reference Fig. 9As shown, it is a schematic diagram of the difference in the expression of road obstacles between the lightweight map provided by this application and the traditional high-precision map, wherein the left picture is a real scene, where irregular obstacles are placed on the edge of the road, these obstacles may be construction equipment, parked vehicles, roadblocks, fallen trees, etc. The right picture compares the difference in the expression of road obstacles between the traditional high-precision map and the lightweight map provided by this application. Traditional high-precision maps express road obstacles according to real scenes. These obstacles can be fixed (such as guardrails, curbs) or movable (such as accident vehicles, construction equipment, parked vehicles), and define the edge of the road surface based on the position of the obstacle, helping the vehicle to determine the drivable area and avoid driving off the road or colliding with road obstacles. The lightweight map provided by this application provides the fitted lane boundaries.

[0140] Based on this, when using lightweight map data for assisted driving, a difference threshold can be set in advance to indicate the maximum allowable deviation range. When the vehicle is in the assisted driving state, the actual boundary detected by the vehicle sensor is compared with the lane boundary extracted from the lightweight map data. If the boundary difference in the comparison result exceeds this difference threshold, a reminder message can be generated to remind the user to exit the assisted driving state and take over the vehicle, thereby ensuring driving safety in complex or uncertain environments.

[0141] In a possible implementation, the map-based intelligent driving implementation method provided in the embodiment of the present application may further include:

[0142] Vectorize the environmental perception data detected by the vehicle sensor to generate vehicle-side vector data;

[0143] The vehicle-side vector map data is sent to the cloud. The cloud is used to extract the map elements of the target section based on the multiple vehicle-side vector data after obtaining the multiple vehicle-side vector data of the target section, and update the lightweight map data of the target section based on the extracted map elements.

[0144] Environmental perception data can be image data collected by visual sensors, such as driving recorders. Before using visual sensors for data collection, the vehicle can be calibrated to ensure that the image data collected by the visual sensor can be accurately mapped to the coordinate system of the real world. Among them, environmental perception data can come from operating vehicles such as online car-hailing vehicles and logistics vehicles. By deploying visual sensors on these vehicles, environmental perception data can be continuously collected. In addition, environmental perception data can also come from smart driving vehicles. When smart driving vehicles turn on the smart driving function, their visual sensors will collect environmental perception data.

[0145] In this embodiment, a vectorization algorithm can be deployed on the vehicle, so that the environmental perception data can be vectorized on the vehicle side to generate vehicle-side vector data. Data vectorization can be realized on the vehicle side, which greatly reduces the amount of data returned, and vector data is easier to deflect and encrypt coordinates, which can improve the security of data return.

[0146] Through vectorization algorithms, combined with time synchronization optimization technology, combined navigation technology, and camera adaptive calibration technology, mass production-level on-device mapping capabilities can be achieved. Specifically, the mapping process can be performed in the vehicle's visual sensor.

[0147] Reference Fig.10 As shown, this is a schematic diagram of the drawing process of the lightweight map provided in this application, in which environmental perception data is collected through visual sensors, and the environmental perception data is vectorized on the vehicle to generate vehicle-side vector data. Multiple vehicle-side vector data for the same road section are reported to the cloud, and the cloud generates a lightweight map of the road section based on the multiple vehicle-side vector data. After inspection and verification, the generated lightweight map can be published.

[0148] Compared with high-precision maps, lightweight maps have made corresponding subtractions in data accuracy, included map elements, and production models, but have greatly increased data coverage and freshness, realizing the expansion of NOA (Navigate on Autopilot) functions to urban roads and lane-level rendering and navigation functions.

[0149] The map-based intelligent driving implementation method provided by the embodiment of the present application can obtain lightweight map data from the cloud, and the vehicle can be controlled to perform assisted driving based on the lightweight map data, wherein the lightweight map data is still a kind of high-precision map data, which can provide key map elements required for assisted driving, and the key map elements include lane boundaries. When the vehicle is in the assisted driving state, the lane boundaries provided by the lightweight map data can control the vehicle to keep driving in the lane, that is, when the vehicle is in the assisted driving state, the system uses the lane boundary information in the lightweight map data to control the vehicle. This method relies on lane boundaries, rather than traditional lane centerlines, to control the vehicle. Lightweight map data provides detailed geographic information at the lane level, which is crucial for autonomous driving and advanced driver assistance systems. Compared with standard definition maps and high-precision maps, lightweight maps strike a balance between data volume and details. Lightweight maps significantly reduce the amount of data by streamlining unnecessary map elements. Despite the small amount of data, lightweight maps still retain the key details required for autonomous driving and advanced driver assistance systems. By streamlining map elements, lightweight maps can be updated faster to ensure the freshness of map data. This is crucial to meet the real-time and accuracy requirements of autonomous driving and advanced driver assistance systems for map data. This method achieves the effect of meeting the accuracy and freshness requirements of autonomous driving and advanced driver assistance systems for map data by optimizing the structure and content of map data.

[0150] Fig.11 A schematic diagram of the structure of the map-based intelligent driving implementation device provided in this application, such as Fig.11 As shown, the map-based intelligent driving implementation device 110 provided in this embodiment includes:

[0151] The assisted driving module 1101 is used to control the vehicle to perform assisted driving based on lightweight map data obtained from the cloud, where the lightweight map data is lightweight high-precision map data, and the lightweight map data provides key map elements required for assisted driving, and the key map elements include lane boundaries;

[0152] The control module 1102 is used to control the vehicle to keep driving in the lane based on the lane boundary provided by the lightweight map data when the vehicle is in the assisted driving state, so as to realize vehicle control based on the lane boundary instead of the lane center line.

[0153] In a possible implementation, the control module is specifically used to:

[0154] During the driving process of the vehicle, the lane boundary of the current lane of the vehicle is extracted from the lightweight map data;

[0155] Selecting a reference boundary from the lane boundary of the current driving lane, the reference boundary being a reference boundary for controlling the driving trajectory of the vehicle;

[0156] The distance between the vehicle and the reference boundary is controlled to be within a preset distance range so as to control the vehicle to keep driving in the lane.

[0157] In a possible implementation, the map-based intelligent driving implementation device is further used to:

[0158] When the vehicle is in the assisted driving state, when the planned lane-level driving path indicates that the vehicle needs to change lanes ahead, the vehicle's current driving lane and the target lane for lane change are determined. The lane-level driving path is generated by lane-level path planning based on lightweight map data;

[0159] Extracting the lane boundary of the current driving lane, the lane boundary of the target lane, and a variable lane position point between the current driving lane and the target lane from the lightweight map data, wherein the variable lane position point is a position where the lane boundary between the current driving lane and the target lane changes from a solid line to a dotted line;

[0160] Based on the extracted lane-changing position point, the lane boundary of the current driving lane and the lane boundary of the target lane, a lane-changing area is defined within the lane area of ​​the current driving lane and the target lane. The lane-changing area refers to an area where the vehicle is allowed to change lanes.

[0161] Control the vehicle to change lanes within the lane-changeable area.

[0162] In a possible implementation, the map-based intelligent driving implementation device is further used to:

[0163] When the vehicle is in the assisted driving state, when the planned lane-level driving path indicates that an intersection is required ahead, the vehicle's current driving lane and the target entry lane that is allowed after passing the intersection are determined. The target entry lane includes the outermost lane and the innermost lane that can be entered;

[0164] Extracting a lane boundary of a current driving lane, a lane boundary of an outermost lane, and a lane boundary of an innermost lane from the lightweight map data;

[0165] Based on the extracted lane boundaries of the current driving lane, the lane boundaries of the outermost lane, and the lane boundaries of the innermost lane, a passable area of ​​the intersection is defined within the intersection area. The passable area of ​​the intersection refers to an area where vehicles are allowed to travel and can safely pass through the intersection.

[0166] Control vehicles to pass through the intersection within the passable area of ​​the intersection.

[0167] In a possible implementation, the map-based intelligent driving implementation device is further used to:

[0168] When the vehicle is in the assisted driving state, it is predicted that the side of the vehicle will collide with the obstacle in front while driving along the planned lane-level driving path, and the lane boundary of the current driving lane is extracted from the lightweight map data;

[0169] Based on the extracted lane boundary of the current driving lane and the detected obstacle position information, a narrow road passable area is defined within the lane area of ​​the current driving lane. The narrow road passable area refers to an area where vehicles are allowed to travel without colliding with obstacles.

[0170] Control the vehicle to travel within the passable area of ​​narrow roads to avoid collision with obstacles.

[0171] In a possible implementation, the map-based intelligent driving implementation device is further used to:

[0172] When the vehicle is in the assisted driving state, the actual boundary of the current driving lane is detected by the vehicle sensor;

[0173] Extracting the lane boundary of the current driving lane from the lightweight map data;

[0174] Comparing the detected actual boundary with the extracted lane boundary of the current driving lane to obtain a comparison result;

[0175] Based on the comparison result, it is determined whether to generate a reminder message, and the reminder message is used to remind the user to take over the vehicle.

[0176] In a possible implementation, the map-based intelligent driving implementation device is further used to:

[0177] Vectorize the environmental perception data detected by the vehicle sensor to generate vehicle-side vector data;

[0178] The vehicle-side vector map data is sent to the cloud. The cloud is used to extract the map elements of the target section based on the multiple vehicle-side vector data after obtaining the multiple vehicle-side vector data of the target section, and update the lightweight map data of the target section based on the extracted map elements.

[0179] The map-based intelligent driving implementation device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effects are similar, and this embodiment will not be repeated here.

[0180] Fig.12 The schematic diagram of the structure of the vehicle provided in this application. Fig.12As shown, the vehicle 120 provided in this embodiment includes: at least one processor 1201 and a memory 1202. Optionally, the vehicle 120 also includes a communication component 1203. The processor 1201, the memory 1202 and the communication component 1203 are connected via a bus.

[0181] In a specific implementation process, at least one processor 1201 executes the computer-executable instructions stored in the memory 1202, so that at least one processor 1201 executes the above method.

[0182] The specific implementation process of the processor 1201 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0183] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The steps of the method disclosed in the invention may be directly embodied as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0184] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0185] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.

[0186] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0187] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0188] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0189] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0190] The division of units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0191] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0192] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0193] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0194] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0195] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A map-based intelligent driving implementation method, characterized in that: The method comprises: Controlling the vehicle to perform assisted driving based on lightweight map data obtained from the cloud, wherein the lightweight map data is lightweight high-precision map data, and the lightweight map data provides key map elements required for assisted driving, wherein the key map elements include lane boundaries; When the vehicle is in the assisted driving state, the vehicle is controlled to keep traveling in the lane based on the lane boundary provided by the lightweight map data, so that the vehicle is controlled by relying on the lane boundary instead of the lane centerline.

2. The method according to claim 1, characterized in that The controlling the vehicle to keep traveling in the lane based on the lane boundary provided by the lightweight map data comprises: During vehicle driving, extracting a lane boundary of a lane in which the vehicle is currently driving from the lightweight map data; Selecting a reference boundary from the lane boundary of the current driving lane, wherein the reference boundary is a reference boundary for controlling the driving trajectory of the vehicle; The distance between the vehicle and the reference boundary is controlled to be within a preset distance range, so as to control the vehicle to keep traveling within the lane.

3. The method according to any one of claims 1 to 2, characterized in that: The method further comprises: When the vehicle is in the assisted driving state, when the planned lane-level driving path indicates that a vehicle lane change is required ahead, determining a current driving lane of the vehicle and a target lane for lane change, wherein the lane-level driving path is generated by lane-level path planning based on the lightweight map data; Extracting from the lightweight map data the lane boundary of the current driving lane, the lane boundary of the target lane, and a variable lane position point between the current driving lane and the target lane, the variable lane position point being a position where the lane boundary between the current driving lane and the target lane changes from a solid line to a dotted line; Based on the extracted lane-changeable position point, the lane boundary of the current driving lane and the lane boundary of the target lane, a lane-changeable area is defined within the lane area of ​​the current driving lane and the target lane, wherein the lane-changeable area refers to an area where a vehicle is allowed to change lanes; The vehicle is controlled to change lanes within the lane-changeable area.

4. The method according to any one of claims 1 to 2, characterized in that: The method further comprises: When the vehicle is in the assisted driving state, when the planned lane-level driving path indicates that an intersection needs to be passed ahead, determining the current driving lane of the vehicle and a target entry lane that is allowed to be entered after passing the intersection, wherein the target entry lane includes an outermost lane and an innermost lane that can be entered; extracting a lane boundary of the current driving lane, a lane boundary of the outermost lane, and a lane boundary of the innermost lane from the lightweight map data; Based on the extracted lane boundary of the current driving lane, the lane boundary of the outermost lane, and the lane boundary of the innermost lane, a traversable area of ​​the intersection is defined within the intersection area, wherein the traversable area of ​​the intersection refers to an area where vehicles are allowed to travel and can safely pass through the intersection; The vehicle is controlled to pass through the intersection within a passable area of ​​the intersection.

5. The method according to any one of claims 1-2, characterized in that: The method further comprises: When the vehicle is in the assisted driving state, it is predicted that a side of the vehicle will collide with a front obstacle when the vehicle is traveling along a planned lane-level driving path, and a lane boundary of a current driving lane is extracted from the lightweight map data; Based on the extracted lane boundary of the current driving lane and in combination with the detected obstacle position information, a narrow road passable area is defined within the lane area of ​​the current driving lane, wherein the narrow road passable area refers to an area where vehicles are allowed to travel without colliding with obstacles; The vehicle is controlled to travel within the narrow road passable area to avoid collision with obstacles.

6. The method according to any one of claims 1-2, characterized in that: The method further comprises: When the vehicle is in the assisted driving state, detecting an actual boundary of a current driving lane through a vehicle sensor; extracting a lane boundary of the current driving lane from the lightweight map data; Comparing the detected actual boundary with the extracted lane boundary of the current driving lane to obtain a comparison result; Based on the comparison result, it is determined whether to generate reminder information, where the reminder information is used to remind the user to take over the vehicle.

7. The method according to any one of claims 1-2, characterized in that: The method further comprises: Vectorize the environmental perception data detected by the vehicle sensor to generate vehicle-side vector data; The vehicle-side vector map data is sent to the cloud, and the cloud is used to extract the map elements of the target section based on the multiple vehicle-side vector data after acquiring the multiple vehicle-side vector data of the target section, and update the lightweight map data of the target section based on the extracted map elements.

8. A map-based intelligent driving implementation device, characterized in that: The device comprises: An assisted driving module, used to control the vehicle to perform assisted driving based on lightweight map data obtained from the cloud, wherein the lightweight map data is lightweight high-precision map data, and the lightweight map data provides key map elements required for assisted driving, wherein the key map elements include lane boundaries; A control module is used to control the vehicle to keep driving in the lane based on the lane boundary provided by the lightweight map data when the vehicle is in the assisted driving state, so as to realize vehicle control relying on the lane boundary instead of the lane center line.

9. A vehicle, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 7.

10. A computer-readable storage medium / computer program product, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor; and / or, The computer program product comprises a computer program, which implements the method according to any one of claims 1 to 7 when executed by a processor.

Images