Vehicle control method, vehicle-mounted equipment, readable storage medium and program product

Abstract

The invention relates to the technical field of auxiliary driving, in particular to a vehicle control method, vehicle-mounted equipment, a readable storage medium and a program product. The vehicle control method comprises the steps that road information of a vehicle on a current driving road is obtained, the current driving road comprises multiple lanes, and the road information at least comprises lane line position data of the multiple lanes; acquiring a navigation path, wherein the navigation path is at least used for indicating the next driving road of the current driving road; according to the navigation path and the road information, a navigation decision scheme for indicating vehicle driving is determined, the navigation decision scheme is at least used for indicating the vehicle to drive from the current position to the next driving road through a target lane, and the target lane is a lane determined from multiple lanes. Furthermore, vehicle driving can be controlled according to the navigation decision scheme. Therefore, navigation based on the navigation decision scheme can be realized under the condition of not depending on a high-precision map, and the driving experience of a user is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of assisted driving, in particular to a vehicle control method, a vehicle-mounted device, a readable storage medium and a program product. BACKGROUND

[0002] Navigation on autopilot (NOA) usually relies on a high-definition map (HD Map). The HD Map provides high-precision road geometry information and rich traffic environment information to assist the safe driving of vehicles.

[0003] However, the HD Map has a problem of freshness difference. The HD Map cannot update the changed road set information and traffic environment information in time, and the update cost of the HD Map is high. For example, if the closure information of a road in the HD Map is not updated in time, the navigation system may still guide the vehicle to drive on the closed road, which brings a driving safety hazard and affects the driving experience of the user. SUMMARY

[0004] The present application provides a vehicle control method, a vehicle-mounted device, a readable storage medium and a program product.

[0005] In a first aspect, the present application provides a vehicle control method applied to a vehicle-mounted device, wherein the vehicle-mounted device comprises a map; the method comprises: obtaining road information of a current driving road of a vehicle, wherein the current driving road comprises a plurality of lanes, and the road information at least comprises lane line position data of the plurality of lanes; obtaining a navigation path, the navigation path being used at least to indicate a next driving road of the current driving road; determining a navigation decision scheme indicating driving of the vehicle according to the navigation path and the road information, the navigation decision scheme being used at least to indicate driving of the vehicle from a current position to the next driving road via a target lane, wherein the target lane is a lane determined from the plurality of lanes; and controlling driving of the vehicle according to the navigation decision scheme.

[0006] In some embodiments of the present application, the navigation path can be obtained based on a map, which can be a standard definition map (SD Map) pre-stored in the vehicle-mounted device.

[0007] Based on the above method, in-vehicle equipment can perceive road information such as lane line position data in real time. Then, by combining road information and navigation path, it determines a navigation decision scheme to guide the vehicle's movement and controls the vehicle's movement based on the navigation decision scheme. In this way, it is possible to provide accurate navigation decision schemes for the vehicle by using cameras to perceive road information and navigation path in real time without relying on high-definition maps (HD Maps) or LiDAR. This allows the vehicle to reach its destination safely and efficiently based on the navigation decision scheme, improving the user's driving experience. Since it does not rely on HD Maps, it reduces the cost required for map updates.

[0008] Furthermore, since the in-vehicle device in this application can perceive the road information of the vehicle's current driving route in real time, and combine the real-time perceived road information with the navigation path obtained through SD Map to make navigation decision planning, it breaks through the freshness and cost limitations of HD Map, enabling the navigation system to better cope with changes in road and traffic conditions, enhancing the robustness of the navigation system, and supporting a wider range of application scenarios.

[0009] In one possible implementation of the first aspect above, obtaining road information of the vehicle on the current driving road includes any one or more of the following: obtaining traffic flow data of multiple lanes based on traffic flow scanning; obtaining lane line position data of multiple lanes based on lane scanning; obtaining passable area data of multiple lanes based on passable area scanning.

[0010] In one possible implementation of the first aspect above, acquiring traffic flow data for multiple lanes based on traffic flow scanning includes: clustering and trajectory fitting the dynamic obstacle information of each lane perceived by the vehicle's sensors through a traffic flow scanning module to obtain traffic flow data for each lane; wherein, the traffic flow data for each lane includes the identifier, average speed, and orientation of at least one dynamic obstacle.

[0011] In some embodiments of this application, the traffic flow data for each lane is the collection of traffic flow data for each vehicle in that lane. The traffic flow data for each vehicle can consist of a series of scatter points P_i[x,y,yaw,avg,v,max_v,min_v,touch_num], where (x,y) can represent the vehicle's position, yaw can represent the vehicle's orientation, (avg_v,max_v,min_v) can represent the vehicle's average speed, maximum speed, and minimum speed, respectively, and touch_num can represent the number of vehicles passing through a preset position.

[0012] In one possible implementation of the first aspect above, acquiring passable area data for multiple lanes includes: projecting the spatial occupancy information of the current driving road sensed by the sensor onto a preset scan line through a passable area scanning module; and determining the boundary of the passable area of ​​the current driving road and the passable area defined by the boundary based on the scan line.

[0013] This is understandable; road information can be perceived based on visual sensors such as cameras.

[0014] In one possible implementation of the first aspect above, the vehicle's current lane is different from the target lane, and the navigation decision scheme includes a lane-change instruction to switch from the current lane to the target lane; based on the navigation path and road information, a navigation decision scheme is determined to instruct the vehicle to travel, including: determining the target driving direction from the current lane to the target lane based on the navigation path, and determining the lane-change direction in the lane-change instruction based on the target driving direction; determining the target lane-change time for the vehicle to switch from the current lane to the target lane based on a dynamic path planning algorithm, and determining the lane-change time in the lane-change instruction based on the target lane-change time.

[0015] In one possible implementation of the first aspect above, the navigation path is used to indicate the first intersection encountered when traveling from the current driving road to the next driving road; the road information also includes intersection information of the first intersection, which includes the type of the first intersection, stop line position data, zebra crossing position data, guide line position data and / or traffic light status; the navigation decision scheme also includes the target intersection driving path passing through the first intersection; based on the navigation path and the road information, a navigation decision scheme is determined to instruct the vehicle to travel, including: determining the target intersection driving path using a Bézier curve algorithm based on the intersection information of the first intersection.

[0016] In one possible implementation of the first aspect above, the intersection elements of the first intersection include stop lines, zebra crossings, guide lines, and / or traffic lights; the method for determining the intersection information of the first intersection includes: clustering the intersection elements of the first intersection to obtain at least one set of intersection elements, wherein each set of intersection elements corresponds to a type of intersection element; filtering the at least one set of intersection elements to remove element points that do not meet the preset stability conditions, thereby obtaining the intersection information of the first intersection.

[0017] In one possible implementation of the first aspect above, determining the target intersection driving path using the Bézier curve algorithm includes: determining at least one candidate intersection driving path based on the intersection information of the first intersection using the Bézier curve algorithm, wherein the at least one candidate intersection driving path is used to indicate that the vehicle passes through the first intersection; evaluating each candidate intersection driving path based at least on the traffic flow data and passable area data of each candidate intersection driving path, determining the candidate intersection driving path with the highest score among the candidate intersection driving paths, and obtaining the target intersection driving path.

[0018] In one possible implementation of the first aspect mentioned above, the navigation decision scheme further includes a continuous and smooth path from the current driving road to the next driving road via the first intersection; determining a navigation decision scheme to instruct the vehicle to drive based on the navigation path and road information, further including: determining a target connecting path that matches the target intersection driving path based on the lane change instruction; and smoothly fitting the target connecting path with the target intersection driving path to generate a continuous and smooth path from the current driving road to the next driving road via the first intersection.

[0019] In one possible implementation of the first aspect above, determining the target connecting path that matches the driving path at the target intersection based on the lane change instruction includes: using a recursive algorithm to determine at least one candidate connecting path based on road information, and determining the target connecting path from the at least one candidate connecting path according to the lane change instruction.

[0020] In one possible implementation of the first aspect mentioned above, the navigation decision scheme further includes the traffic status of a vehicle entering the first intersection from its current driving road; determining a navigation decision scheme to instruct the vehicle to travel based on the navigation path and road information, and further includes: associating the traffic lights at the first intersection with at least one intersection element in the intersection information through an association algorithm to determine the vehicle's driving direction, and determining the target traffic light based on the vehicle's driving direction and the target intersection driving path; and generating a traffic control command corresponding to acceleration, constant speed, deceleration, or braking based on the state of the target traffic light.

[0021] In one possible implementation of the first aspect above, the navigation decision scheme further includes a target speed limit for the current driving road; when the road information includes speed limit sign information for the current driving road, a navigation decision scheme for instructing the vehicle to travel is determined based on the navigation path and the road information, including: processing the speed limit sign information based on an association algorithm to obtain reference speed limit sign information; calibrating the speed limit corresponding to the navigation path based on the reference speed limit sign information to obtain a target speed limit, wherein the target speed limit is a speed limit that is consistent with or has an error within a preset range corresponding to the speed limit corresponding to the reference speed limit sign information.

[0022] In one possible implementation of the first aspect above, when the road information does not include speed limit sign information of the current road, determining a navigation decision scheme to instruct the vehicle to travel based on the navigation route and the road information further includes: using the speed limit corresponding to the navigation route as the target speed limit.

[0023] Secondly, this application also provides a vehicle control device, including a perception module, a decision module, and a control module; the perception module is used to acquire road information of the vehicle on the current driving road, wherein the current driving road includes multiple lanes, and the road information includes at least lane line position data of the multiple lanes; the perception module is used to acquire a navigation path, the navigation path being used to indicate at least the next driving road of the current driving road; the decision module is used to determine a navigation decision scheme to instruct the vehicle to drive based on the navigation path and the road information, the navigation decision scheme being used to instruct the vehicle to drive from the current position to the next driving road via a target lane, wherein the target lane is a lane determined from multiple lanes; the control module is used to control the vehicle to drive according to the navigation decision scheme.

[0024] Thirdly, this application also provides an in-vehicle device, including: one or more memories and one or more processors, the memories being coupled to the processors; the memories storing one or more programs; when the stored one or more programs are executed by one or more processors, the in-vehicle device causes the in-vehicle device to perform the methods mentioned in the first aspect and any possible implementation of the first aspect.

[0025] Fourthly, this application also provides a vehicle that may include the vehicle-mounted equipment provided in the second aspect above.

[0026] Fifthly, this application also provides a readable storage medium storing instructions that, when executed on an in-vehicle device, cause the in-vehicle device to perform the methods mentioned in the first aspect and any possible implementation thereof.

[0027] Sixthly, this application also provides a program product that, when run on an in-vehicle device, causes the in-vehicle device to perform the methods mentioned in the first aspect and any possible implementation thereof.

[0028] The beneficial effects of the second to sixth aspects mentioned above can be referred to the relevant descriptions in the first aspect and any possible implementation of the first aspect, which will not be repeated here. Attached Figure Description

[0029] Figure 1 According to an embodiment of this application, a schematic diagram of the structure of an environment model is shown;

[0030] Figure 2 According to an embodiment of this application, a traffic flow scan map is shown;

[0031] Figure 3A According to an embodiment of this application, a scan map of a passable area is shown;

[0032] Figure 3B According to an embodiment of this application, a boundary contour map of a passable area is shown;

[0033] Figure 3C According to an embodiment of this application, a schematic diagram of an occupied grid for a passable area is shown;

[0034] Figure 4A According to an embodiment of this application, a lane scanning topology map is shown;

[0035] Figure 4B According to an embodiment of this application, a schematic diagram of a traffic road is shown;

[0036] Figure 5 According to an embodiment of this application, a schematic diagram of a vehicle changing lanes is shown;

[0037] Figure 6 According to an embodiment of this application, a schematic diagram of a traffic scenario at an intersection is shown;

[0038] Figure 7 According to an embodiment of this application, a schematic diagram of a path search and connection scenario is shown;

[0039] Figure 8 According to an embodiment of this application, a schematic diagram showing the association between intersection path decision-making and traffic lights is shown;

[0040] Figure 9 According to an embodiment of this application, a schematic flowchart of a vehicle control method is shown;

[0041] Figure 10 According to an embodiment of this application, a structural schematic diagram of a vehicle control device 800 is shown;

[0042] Figure 11 According to an embodiment of this application, a functional framework diagram of a vehicle 100 is shown. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0044] The vehicle control method provided in this application can be applied to devices including, but not limited to, telematics devices, vehicle control units (VCUs), advanced driver assistance systems (ADAS), etc. This application does not limit the specific form of the device.

[0045] As mentioned earlier, Navigation-Assisted Driving (NOA) relies on high-resolution (HD) maps. However, HD maps suffer from poor timeliness; they cannot update changed road and traffic information in a timely manner, and updating HD maps is costly. For example, if the HD map does not update the closure information of a certain road, the navigation system may still guide the vehicle onto that closed road, posing a driving safety hazard and affecting the user's driving experience. Furthermore, updating HD maps is costly.

[0046] To address the aforementioned problems, this application provides a vehicle control method, which specifically includes: acquiring road information of the vehicle on the current driving road, wherein the current driving road includes multiple lanes, and the road information includes at least lane line position data of the multiple lanes; acquiring a navigation path, wherein the navigation path is used to indicate at least the next driving road of the current driving road; determining a navigation decision scheme to instruct the vehicle to travel based on the navigation path and the road information, wherein the navigation decision scheme is used to instruct the vehicle to travel from the current position to the next driving road via a target lane, wherein the target lane is a lane determined from the multiple lanes; and controlling the vehicle to travel according to the navigation decision scheme.

[0047] In some embodiments of this application, navigation routes can be obtained based on a map, which can be a standard definition map (SD Map) pre-stored in the vehicle device. This map may include basic road information such as road network topology (e.g., road connections, intersections, forks, etc.) and road types (e.g., highways and urban roads). The navigation system can employ a path planning algorithm to plan a navigation route from the starting point to the destination based on the map and the user-inputted starting and ending points. It should be understood that a navigation route is essentially composed of a series of continuous points (or nodes), which represent key locations the vehicle needs to pass through on the map, such as intersections, forks, and the start and end points of roads. The path planning algorithm calculates the shortest or optimal path between these points, connects them to form a continuous route, thereby guiding the vehicle from the starting point to the destination.

[0048] The method provided in this application embodiment allows the in-vehicle device to perceive road information such as lane line position data in real time. By combining road information and navigation path, a navigation decision scheme is determined to guide the vehicle's movement, and the vehicle's movement is controlled based on this navigation decision scheme. In this way, without relying on high-resolution maps (HDMap) or LiDAR, accurate navigation decision schemes can be provided for the vehicle's movement by using a camera to perceive road information and navigation path in real time. This allows the vehicle to reach its destination safely and efficiently based on the navigation decision scheme, improving the user's driving experience.

[0049] Furthermore, since the in-vehicle device in this application can perceive the road information of the vehicle's current driving route in real time, and combine the real-time perceived road information with the navigation path obtained through SD Map to make navigation decision planning, it breaks through the freshness and cost limitations of HD Map, enabling the navigation system to better cope with changes in road and traffic conditions, enhancing the robustness of the navigation system, and supporting a wider range of application scenarios.

[0050] In some embodiments of this application, the in-vehicle device may include an environment model, which may include an environment scanning layer and an environment decision layer. Each module in the environment decision layer can determine the vehicle's navigation decision scheme based on data from the environment scanning layer.

[0051] For example, Figure 1 A schematic diagram of the structure of an environment model is shown.

[0052] refer to Figure 1 The environmental scanning layer may include a traffic flow scanning module, a lane scanning module, and a passable area scanning module. The environmental decision layer may include a navigation guidance decision module, an intersection route decision module, a route search decision module, a traffic light passage decision module, and a speed limit decision module.

[0053] In this embodiment, the vehicle-mounted device can perceive road information of the current driving route based on various modules in the environmental scanning layer. The road information of the vehicle on the current driving route may include any one or more of the following: traffic flow data of multiple vehicles obtained by the traffic flow scanning module; lane line position data of multiple lanes obtained by the lane scanning module; and passable area data of multiple lanes obtained by the passable area scanning module.

[0054] In some embodiments of this application, obtaining traffic flow data for multiple lanes may include: clustering and trajectory fitting the dynamic obstacle information of each lane perceived by the vehicle's sensors through a traffic flow scanning module to obtain traffic flow data for each lane; wherein, the traffic flow data for each lane includes the identifier, average speed, and orientation of at least one dynamic obstacle.

[0055] In some embodiments of this application, traffic flow data can be obtained by the traffic flow scanning module using the k-means clustering algorithm to process the dynamic obstacle information of each lane perceived by the vehicle's sensors. That is, the sampling k-means clustering algorithm clusters and fits the current state, historical trajectory and predicted driving trajectory of the dynamic obstacles perceived by the camera to obtain traffic flow data for each dynamic obstacle.

[0056] For example, Figure 2 A traffic flow scan map is shown. The traffic flow scan map can be obtained using a traffic flow scanning module. Based on the traffic flow scan map, traffic flow data for multiple lanes on the current road can be acquired. (Reference) Figure 2 The diagram shows traffic flow data for multiple lanes in the current travel road R1. The traffic flow data for each lane is the collection of traffic flow data for each vehicle in that lane. The traffic flow data for the three lanes of the current travel road R1 includes dynamic obstacles such as vehicles A to E. The combination of the traffic flow data for each vehicle constitutes the traffic flow data for the current travel road R1.

[0057] It should be understood that the traffic flow data for each vehicle can include information such as the identification of various dynamic obstacles passing through this traffic flow, average speed, and orientation. Specifically, the traffic flow data for each vehicle can be composed of a series of scattered points P_i[x, y, yaw, avg_v, max_v, min_v, touch_num], where (x, y) can represent the vehicle's position, yaw can represent the vehicle's orientation, (avg_v, max_v, min_v) can represent the vehicle's average speed, maximum speed, and minimum speed, respectively, and touch_num can represent the number of vehicles passing through the preset position.

[0058] It should be understood that the traffic flow scanning module can send traffic flow data to the lane scanning module.

[0059] In some embodiments of this application, obtaining passable area data for multiple lanes may include: projecting the spatial occupancy information of the current driving road sensed by the sensor onto a preset scan line through a passable area scanning module; and determining the boundary of the passable area of ​​the current driving road and the passable area defined by the boundary based on the scan line.

[0060] For example, Figure 3A A scan map of a passable area is shown. Figure 3B A traversable area boundary contour map is shown. The traversable area scanning module can, according to... Figure 3A and Figure 3B This involves determining the passable areas and searchable routes within the vehicle's current driving path. Figure 3AThe passable area scanning module uses vehicle sensors (such as cameras) to perceive the current road space the vehicle is traveling on. It converts the perception results into an occupancy grid, thus identifying which areas are occupied (obstacles) and which areas are free (passable). Then, it determines the passable area based on the identified free areas. (Reference) Figure 3B , Figure 3B 35 is the boundary line of the passable area, that is, the edge of the area where vehicles can drive safely. 36 is the boundary line, that is, the farthest distance or maximum range that the vehicle sensors can detect. This boundary line defines the effective detection range of the vehicle sensors.

[0061] Figure 3C A schematic diagram of an occupied grid for a passable area is shown. (Reference) Figure 3C The passable area scanning module uses vehicle sensors to perceive the current road space the vehicle is traveling on, converting the perception results into an occupancy grid. The occupancy grid is a two-dimensional grid map where each grid cell indicates whether the area is occupied (obstacles present) or vacant (passable). The perceived results are then projected onto scan lines, with nodes set at specific intervals along the scan lines. These nodes represent possible travel path points for the vehicle. The passable area scanning module can determine the boundaries of passable areas by finding the intersections of rays emanating from the vehicle and the scan axis. Furthermore, the passable area scanning module can use path search algorithms (such as the A* algorithm) to search for passable areas within the occupancy grid, finding a search path from the current location to the target location (e.g., search path 50). Search path 51 can represent an impassable search path.

[0062] It should be understood that the passable area scanning module can send passable area data to the lane scanning module.

[0063] In some embodiments of this application, obtaining lane line position data of multiple lanes based on lane scanning may include: scanning lane lines and static obstacles in the current driving road using a lane scanning module to obtain lane line position data in the current driving road. The lane line position data may include lane line position, type and color, the position and type of static obstacles, etc. The lane line position may be a set of discrete points of lane lines and roadside lines.

[0064] It is understood that the lane scanning module can receive traffic flow data from multiple lanes sent by the traffic flow scanning module and passable area data sent by the passable area scanning module. It can then generate a lane scanning topology map based on lane line position data, traffic flow data from multiple lanes, and passable area data. This involves projecting lane line position data (a set of discrete points), traffic flow data from multiple lanes (at least one set of curves composed of multiple discrete points), and passable area data (a set of discrete points) onto a specific set of scan lines, resulting in a lane scanning topology map that includes lane line position data, traffic flow data, and passable area data. This topology map can include the lane midpoints, lane boundary points, and the connections between these points. Based on this lane scanning topology map, the traffic conditions of each lane in the current road can be assessed.

[0065] For example, Figure 4A A lane scanning topology map is shown, based on Figure 4A This allows us to determine the lane markings within the road where the vehicle is traveling. (Reference) Figure 4A The lane scan topology map covers the area in front of the vehicle. Figure 4A The sensor employs scan lines arranged at equal intervals from top to bottom, with boundary lines 20 and 21 serving as the left and right scanning interval boundaries. Each scan line represents a single scan by the sensor at a different height. As the scan line passes over points on the lane lines, the sensor records information about these points, including the position, color, and type of the lane line corresponding to that point. Each scan line horizontally links the information scanned in the same row (the same scan line) and vertically links corresponding points on different scan lines to construct a topology map containing all scan points and their interrelationships. Furthermore, centerline points within the same lane can be connected. By connecting the centerline points of the same lane, the lane centerlines (e.g., lane centerlines) in the lane scan topology map can be obtained. Figure 4A Lane center lines (e.g., lane center lines 23), and can identify lane lines corresponding to lane center lines (e.g., lane center lines 23). Figure 4A Lane 22, lane 24, etc. Figure 4A The lane scan topology map shown can help the navigation system understand the structure and layout of the current road, thereby better determining the navigation decision scheme to guide the vehicle.

[0066] It is understandable that the vehicle's current travel path can be as follows: Figure 4B The road diagram shown is for reference. Figure 4B The road in which the vehicle is currently traveling may include lane centerline 30, lane boundary line 31 and lane boundary line 32, and traffic flow ( Figure 4B (represented by white squares) and passable areas ( Figure 4B (The area in the middle is represented as gray).

[0067] Understandably, in traditional architectures, perceived data is first used to reconstruct the entire environment model. This process may lose detailed information because the goal of environment reconstruction is to build a macroscopic view of the environment, rather than focusing on every detail. However, in the embodiments of this application… Figure 1 The traffic flow scanning module, lane scanning module, and passable area scanning module in the environmental scanning layer can directly extract information from the perceived data, rather than performing complex environmental reconstruction first. In other words, the environmental scanning layer in this application can directly focus on traffic flow data, lane line location data, and passable areas, which are crucial components of subsequent decision-making. For example, the traffic flow scanning module can directly obtain vehicle speed and density information, the lane scanning module can directly obtain lane line boundaries and types, and the passable area scanning module can directly determine the areas where vehicles can travel, reducing the amount of environmental information that may be lost during environmental reconstruction.

[0068] Continue to refer to Figure 1 The environmental decision layer may include a navigation guidance decision module, an intersection route decision module, a route search decision module, a traffic light passage decision module, and a speed limit decision module.

[0069] In some embodiments of this application, the navigation guidance decision module can make navigation decisions based on the following multi-source information: lane scan topology map, navigation path, and map (SD map). The navigation guidance decision module applies a map matching algorithm, which uses the lane scan topology map as a reference to perform road-level matching with the SD map and integrates the global intent provided by the navigation path to generate navigation guidance instructions. Specifically, this map matching algorithm can achieve precise lane-level positioning of the vehicle and correlation analysis of its planned path by deeply fusing multi-source information, thereby generating instructions including lane change commands, navigation speed limits, and lane markings allowing free overtaking (without deviating from the navigation direction). This provides the vehicle with an accurate and safe navigation guidance solution.

[0070] It is understandable that the navigation guidance decision-making module can deeply fuse multi-source information through map matching algorithms. Regarding multi-source information: On one hand, multi-source information can include vehicle sensor data, such as high-precision Global Positioning System (GPS) information, acceleration, and angular velocity, which allows the map matching algorithm to understand the vehicle's real-time motion state in physical space. On the other hand, multi-source information can include the position, curvature, shape, color, and type of lane lines in the lane scan topology map, which allows the map matching algorithm to accurately understand the lane line feature information of the road the vehicle is currently traveling on. Furthermore, multi-source information can also include the road network topology (such as road connections, intersections, and branching points) and road types (such as highways and urban roads) provided by the map, enabling the map matching algorithm to understand the overall layout and macrostructure of the roads.

[0071] Using the aforementioned multi-source information, the navigation guidance decision module can not only determine the vehicle's precise location through map matching algorithms, but also comprehensively analyze parameters such as the consistency between the vehicle's heading and the current driving road direction, the vehicle's lateral deviation relative to the lane centerline, and its relative positional relationship with adjacent lanes. This provides a comprehensive and accurate environmental understanding for generating reasonable and safe navigation decision schemes.

[0072] In some embodiments of this application, the vehicle's current lane is different from the target lane, and the navigation decision scheme includes a lane-changing instruction to switch from the current lane to the target lane. The navigation guidance decision module can determine the navigation decision scheme instructing the vehicle to travel based on the navigation path and road information. Specifically, the navigation guidance decision module can determine the target driving direction from the current lane to the target lane based on the navigation path, and determine the lane-changing direction in the lane-changing instruction based on the target driving direction; it can also determine the target lane-changing time for the vehicle to switch from the current lane to the target lane based on a dynamic path planning algorithm, and determine the lane-changing time in the lane-changing instruction based on the target lane-changing time.

[0073] For example, Figure 5 A schematic diagram of a vehicle changing lanes is shown.

[0074] refer to Figure 5 ,exist Figure 5 In the vehicle merging scenario shown, the navigation guidance decision module can dynamically generate navigation decision schemes containing different lane-changing instructions and overtaking lane markings based on the different distances between the vehicle and the merging point (macro-decision zone and micro-decision zone). Specifically: the macro-decision zone can refer to the area far from the merging point (e.g., distance ≥ 200m). The micro-decision zone can refer to the area close to the merging point (e.g., distance < 200m).

[0075] Based on this, at distances greater than 500m (macro decision-making zone), the navigation guidance decision-making module can instruct vehicles to overtake from the current lane ( Figure 5 At 500m (left lane to the left of the center), the vehicle can switch from its current lane (right lane between 500m and 200m) to the target lane (left lane in the same area). This target lane can be marked as an overtaking lane. At 500m (still within the macro decision-making zone), the navigation guidance decision module can instruct the vehicle to switch from its current lane (right lane between 500m and 200m) to the target lane (left lane in this section). At 200m (entering the micro decision-making zone), the road surface arrows indicate a right lane change, and the navigation guidance decision module can instruct the vehicle to switch from its current lane (left lane between 200m and the merging point) to the target lane (right lane in this section). Near the merging point (-100m), considering that vehicles are about to merge, the navigation guidance decision module can instruct the vehicle to maintain its current lane and not to change lanes to the right.

[0076] In some embodiments of this application, the intersection path decision module can receive traffic flow scanning data from the traffic flow scanning module, lane scanning topology map output by the lane scanning module, and raw intersection element data acquired by a camera. The intersection elements of the first intersection may include stop lines, zebra crossings, guide lines, and / or traffic lights, etc. The intersection path decision module can process the raw intersection element data as follows: cluster the intersection elements of the first intersection to obtain at least one intersection element set, each intersection element set corresponding to a type of intersection element. Filter the at least one intersection element set to remove element points that do not meet preset stability conditions (such as isolated noise points caused by perception errors), thereby obtaining stable intersection information for the first intersection, including the accurate location, type, and topological relationships of each intersection element.

[0077] In some embodiments of this application, after obtaining the intersection information of the first intersection, the intersection path decision module can determine at least one candidate intersection driving path using the Bezier curve algorithm based on the intersection information, traffic flow data, and lane scan topology map. Then, based on a multi-dimensional cost evaluation mechanism, it selects the candidate intersection driving path with the highest score from the at least one candidate intersection driving path as the target intersection driving path for the first intersection. During the determination of the target intersection driving path, the multi-dimensional cost evaluation mechanism can be simultaneously used to determine whether a lane change is required within the intersection when the vehicle passes through the first intersection along the target intersection driving path; if a lane change is required, a corresponding lane change instruction within the intersection can be generated. Therefore, the intersection path decision module can output the target intersection driving path and / or the lane change instruction within the intersection, constituting a navigation decision scheme instructing the vehicle to pass through the first intersection.

[0078] It should be understood that a multi-dimensional cost assessment mechanism may include the following dimensions: traffic rule compliance, such as whether the route conforms to lane guidance and avoids non-motorized vehicle lanes and pedestrian areas; route smoothness, such as the continuity of route curvature and driving comfort; and traffic safety, such as judging whether there is a risk of conflict between the route and the trajectories of other traffic participants based on traffic flow data, etc. This application does not specifically limit these dimensions.

[0079] For example, Figure 6 A schematic diagram of a traffic scenario at an intersection is shown.

[0080] refer to Figure 6 The first intersection P1 may include intersection elements such as entrance line L1, exit line L2, guide line L3, stop line, zebra crossing, and traffic lights (not shown). The intersection path decision module can evaluate candidate intersection paths A1, A2, and A3 based on the intersection elements in the first intersection P1, traffic flow scan data, and lane scan topology map, and determine A2 as the target intersection path. Furthermore, the intersection path decision module can send the target intersection path corresponding to the first intersection as intersection information to the path search decision module and the traffic light passage decision module.

[0081] In some embodiments of this application, the path search decision module can receive intersection information of the first intersection and the target intersection driving route from the intersection path decision module. Furthermore, the path search decision module can also receive a lane scan topology map from the lane scan module, a lane change instruction from the navigation guidance decision module, and a navigation route. Based on this, the path search decision module can determine a target connecting path matching the target intersection driving route based on the aforementioned multiple inputs, and ultimately generate a continuous and smooth path.

[0082] Specifically, the path search decision module can determine the vehicle's current scan line position based on the lane scan topology map, and use a depth-first search to obtain the path identifier of the vehicle's current location from this starting point. It then expands left and right on the lane scan topology map sequentially, using a recursive algorithm to obtain two paths to the left and two paths to the right of the current vehicle's path, thus identifying multiple candidate connecting paths. Combining the lane-change command output by the navigation guidance decision module, it identifies the target path identifier from these candidate connecting paths; the path corresponding to this target path identifier is the target connecting path. The target connecting path and the target intersection travel path are then smoothly fitted using a B-spline curve to generate a continuous and smooth path indicating the vehicle's journey from the current road through the first intersection to the next road.

[0083] For example, Figure 7 According to an embodiment of this application, a schematic diagram of a path search and connection scenario is shown.

[0084] refer toFigure 7 , Figure 7 It contains multiple equally spaced horizontally arranged scan lines, on which discrete points (such as discrete point set 701) representing lane lines or lane centers are distributed. These discrete points and their connections together constitute the lane scan topology map. Figure 7 In the area near the first intersection, the discrete point set of the current driving road (representing the vehicle's location and adjacent lanes, such as discrete point set 701) is spatially adjacent to another set of discrete points from inside the first intersection (representing the driving path to the target intersection, such as discrete point set 702).

[0085] The path search decision module, based on depth-first search and recursive algorithms, determines the vehicle's current position and lane on the scan line (e.g., point P1 marked as "vehicle position" in the figure). It then expands left and right to identify multiple candidate lanes (vertical sequences of discrete points, such as lane sequences 703 and 704). Based on the lane-changing instruction, a target lane (e.g., lane sequence 703) is selected, and the target connecting path corresponding to this target lane is smoothly connected to the path point sequence within the first intersection (e.g., the path represented by lane sequence 705). This is illustrated in the figure by a smooth curve C1, indicating the continuous smooth path from the current driving road through the first intersection to the next driving road.

[0086] In some embodiments of this application, the traffic light passage decision module (or traffic signal light passage decision module) can generate longitudinal control commands to control vehicles to pass through the intersection based on the traffic signal light status at the intersection. The traffic light passage decision module can receive the following inputs: intersection information of the first intersection, the target intersection travel path from the intersection path decision module, the lane scanning topology map from the lane scanning module, and the navigation path. Based on this, the traffic light passage decision module can determine the passage status of a vehicle entering the first intersection from its current travel road and generate corresponding passage control commands in the following manner.

[0087] Specifically, the traffic light decision-making module can associate traffic lights with other intersection elements (such as zebra crossings, stop lines, etc.) at the first intersection using an association algorithm. During this process, the module can use the coarse traffic light information provided in the navigation path to assist in achieving precise association, thereby determining the correspondence between the traffic lights at the first intersection and other intersection elements. Based on the vehicle's travel direction (which can be indicated by the navigation path) and the target intersection's travel path, and combined with spatial projection matching using the lane scan topology map, the module identifies the target traffic lights the vehicle needs to focus on. For example, if the target intersection's travel path indicates going straight, then the straight-ahead light is focused on. If the target intersection's travel path indicates turning left or right, then the left-turn light or right-turn light is focused on, respectively. Based on the projection of the target traffic light onto the lane scan topology map, its status (red, yellow, or green) and precise location are determined, thereby generating corresponding longitudinal control commands. For example, when the target traffic light is red or yellow, a passage instruction to decelerate to a stop is generated; when the target traffic light changes from red to green, a passage instruction to start or accelerate can be generated to control the passage status of vehicles entering the first intersection.

[0088] For example, Figure 8 A schematic diagram illustrating the association between intersection path decision-making and traffic lights is shown.

[0089] refer to Figure 8 When a vehicle reaches the first intersection (Junction_1), the target intersection travel path is straight (S). The traffic light decision module determines the target traffic light as a straight-ahead light based on the vehicle's direction of travel and the target intersection travel path. The traffic light decision module obtains the current lane's positional relationship with the traffic light. If the straight-ahead signal is detected as red, it generates a deceleration to stop command, controlling the vehicle to wait at the stop line. When the straight-ahead signal switches to green, it generates a start-through command, controlling the vehicle to pass through the first intersection in the straight-ahead direction.

[0090] Continue to refer to Figure 8 When the vehicle continues to the second intersection (Junction_2), the target intersection's travel path is updated to a left turn (L). Based on the updated travel direction and the target intersection's travel path, the traffic light decision module determines the target traffic light to be a left turn light. If the left turn light is red, a deceleration to a complete stop command is generated. If the left turn light is green, start-up and steering control commands are generated to guide the vehicle to complete the left turn.

[0091] In some embodiments of this application, the speed limit decision module can be used to determine the target speed limit that the vehicle should follow on the current road. The speed limit decision module can receive the following inputs: road information (such as speed limit sign information of the current road perceived by a camera) and / or road speed limit information provided in the navigation path. Based on this, the speed limit decision module generates the target speed limit included in the navigation decision scheme.

[0092] Specifically, when the road information includes speed limit signs for the currently traveling road, the speed limit decision module can process the speed limit sign information using an association algorithm to obtain reference speed limit sign information. The association algorithm can be used to verify the validity of the speed limit sign information. For example, by spatiotemporally associating the speed limit value corresponding to the perceived speed limit sign information with the speed limit corresponding to the navigation path, misidentification or outliers can be filtered out, thus obtaining reliable reference speed limit sign information. Then, based on the reference speed limit sign information, the speed limit corresponding to the navigation path is calibrated to obtain the target speed limit, where the target speed limit is the same as the speed limit corresponding to the reference speed limit sign information or the error is within a preset range. When the road information does not include speed limit signs for the currently traveling road, the speed limit decision module uses the speed limit corresponding to the navigation path as the target speed limit.

[0093] Based on this, the speed limit decision module can determine the speed limit of the calibrated navigation route or the speed limit corresponding to the navigation route as the target speed limit based on whether speed limit sign information is detected.

[0094] pass Figure 1 The illustrated environmental model allows onboard equipment to utilize modules in the environmental scanning layer to perceive real-time road information, such as traffic flow data, lane position data, and passable area data. This enables the environmental decision layer to determine a navigation decision scheme based on the real-time perceived road information, navigation path, and low-precision map (SD Map). This eliminates the need for high-precision map (HD Map), reducing navigation misjudgments caused by untimely HD Map updates and improving driving safety. Furthermore, the elimination of HD Maps reduces map update costs.

[0095] It should be understood that in traditional architectures, perceived data is first used to reconstruct the entire environment model. This process may lose detailed information because the purpose of environment reconstruction is to build a macroscopic view of the environment, rather than focusing on every detail. However, in the embodiments of this application… Figure 1The traffic flow scanning module, lane scanning module, and passable area scanning module in the environmental scanning layer can directly extract information from the perceived data (such as traffic flow data, lane line position data, and passable area data), instead of performing complex environmental reconstruction first. In other words, the environmental scanning layer in this application can directly focus on traffic flow data, lane line positions, and passable area locations, which are crucial components of subsequent navigation decisions. For example, the traffic flow scanning module directly extracts vehicle speed and density information, the lane scanning module directly extracts lane boundaries and types, and the passable area scanning module directly determines the areas where vehicles can travel. This directness reduces the amount of information that may be lost during environmental reconstruction.

[0096] Furthermore, in traditional architectures, the decision-making module relies on multi-layered environmental models, which may not accurately reflect complex perception results. However, in the embodiments of this application... Figure 1 Each decision-making module in the environmental decision layer can directly determine the navigation decision scheme to instruct the vehicle to drive based on the perception data in the environmental decision layer. It does not need to rely on the simplified and potentially distorted environmental model in the traditional architecture.

[0097] In summary, this embodiment of the application directly extracts key information such as traffic flow data, lane position data, and passable area data from the perception data, avoiding information loss in the traditional environmental reconstruction process and ensuring the integrity and accuracy of the navigation decision-making basis. At the same time, the environmental decision layer makes navigation decisions directly based on real-time perception data without relying on high-precision maps, reducing the risk of misjudgment caused by untimely updates of high-precision maps, reducing map update costs, and improving the system's flexibility and driving safety.

[0098] Combination Figure 1 , Figure 9 A flowchart of a vehicle control method is shown.

[0099] S101: Obtain road information for the vehicle on the current driving road, wherein the current driving road includes multiple lanes, and the road information includes at least lane line position data for the multiple lanes.

[0100] In some embodiments of this application, obtaining road information of the vehicle on the current driving road includes any one or more of the following: obtaining traffic flow data of multiple lanes based on traffic flow scanning; obtaining lane line position data of multiple lanes based on lane scanning; obtaining passable area data of multiple lanes based on passable area scanning.

[0101] It should be understood that road information can be based on Figure 1The traffic flow scanning module, lane scanning module, and passable area scanning module in the middle environmental perception layer sense the data. The process by which these modules acquire traffic flow data, lane position data, and passable area data can be found in [reference needed]. Figure 2 The specific descriptions in the text will not be repeated here.

[0102] S102: Obtain a navigation path, which is at least used to indicate the next road to travel on the current road.

[0103] In some embodiments of this application, the in-vehicle device may include a map, through which a navigation path can be obtained. The navigation path marks the start and end points, i.e., the vehicle's current location and the target location. Furthermore, the navigation path can determine the next route the vehicle is currently traveling on.

[0104] S103: Based on the navigation path and road information, determine a navigation decision scheme to instruct the vehicle to travel. The navigation decision scheme is used to instruct the vehicle to travel from its current location to the next road via a target lane, wherein the target lane is a lane determined from multiple lanes.

[0105] In some embodiments of this application, the vehicle-mounted device can be based on Figure 1 Each module in the environmental decision layer shown in the diagram determines a navigation decision scheme to instruct the vehicle to travel based on the navigation path and road information. The navigation decision scheme is used to instruct the vehicle to travel from its current location to the next travel road via the target lane.

[0106] Specifically, the in-vehicle device can determine the navigation decision scheme instructing the vehicle to travel using a navigation guidance decision module. This navigation decision scheme can include a lane-changing instruction to switch from the current lane to a target lane within the current road. Determining the navigation decision scheme based on the navigation path and road information can specifically include: the navigation guidance decision module determining the target travel direction from the current lane to the target lane using the navigation path, and determining the lane-changing direction based on the target travel direction. For example, if the current lane is a straight-ahead lane and the target lane is a right-turn lane, then the target travel direction is a right-turn direction, and the lane-changing direction is also a right-turn direction. Secondly, the navigation guidance decision module can determine the lane-changing time from the current lane to the target lane based on a dynamic path planning algorithm. Thus, the navigation guidance decision module can determine the lane-changing instruction based on the lane-changing direction and time.

[0107] In other embodiments of this application, the vehicle-mounted device can determine a navigation decision scheme to instruct the vehicle to travel through the intersection route decision module.

[0108] Specifically, the intersection path decision module can receive traffic flow scanning data from the traffic flow scanning module, lane scanning topology map output by the lane scanning module, and raw intersection element data acquired by the camera. The intersection elements of the first intersection may include stop lines, zebra crossings, guide lines, and / or traffic lights. The intersection path decision module can process the raw intersection element data as follows: cluster the intersection elements of the first intersection to obtain at least one set of intersection elements, each set corresponding to a type of intersection element. Filter the at least one set of intersection elements to remove element points that do not meet preset stability conditions (such as isolated noise points caused by perception errors), thereby obtaining stable intersection information for the first intersection, including the accurate location, type, and topological relationships of each intersection element.

[0109] After obtaining the intersection information of the first intersection, the intersection path decision module can determine at least one candidate intersection driving path using the Bezier curve algorithm based on the intersection information, traffic flow data, and lane scan topology map. Then, based on a multi-dimensional cost evaluation mechanism, it selects the candidate intersection driving path with the highest score from these at least one candidate intersection driving path as the target intersection driving path for the first intersection. During the process of determining the target intersection driving path, the multi-dimensional cost evaluation mechanism can simultaneously be used to determine whether a lane change is required within the intersection when the vehicle passes through the first intersection along the target intersection driving path; if a lane change is required, a corresponding lane change instruction within the intersection can be generated. Therefore, the intersection path decision module can output the target intersection driving path and / or the lane change instruction within the intersection, forming a navigation decision scheme instructing the vehicle to pass through the first intersection.

[0110] It should be understood that a multi-dimensional cost assessment mechanism can consider the following dimensions: traffic rule compliance, such as whether the route conforms to lane guidance and avoids non-motorized vehicle lanes and pedestrian areas; route smoothness, such as the continuity of route curvature and driving comfort; and traffic safety, such as judging whether there is a risk of conflict between the route and the trajectories of other traffic participants based on traffic flow data.

[0111] In other embodiments of this application, the on-board device can determine a navigation decision scheme to instruct the vehicle to travel via a path search decision module. The path search decision module can receive intersection information from a first intersection and the target intersection travel path from an intersection path decision module. Furthermore, the path search decision module can also receive a lane scan topology map from a lane scan module, a lane change instruction from a navigation guidance decision module, and a navigation path. Based on this, the path search decision module can determine a target connection path matching the target intersection travel path based on the aforementioned multiple inputs, and ultimately generate a continuous and smooth path.

[0112] Specifically, the path search decision module can determine the vehicle's current scan line position based on the lane scan topology map, and use a depth-first search to obtain the path identifier of the vehicle's current location from this starting point. Then, it expands left and right on the lane scan topology map sequentially, using a recursive algorithm to obtain the two paths to the left and two paths to the right of the current vehicle's path, thus identifying multiple candidate connecting paths. Then, combined with the lane-change command output by the navigation guidance decision module, it identifies the target path identifier from the multiple candidate connecting paths; the path corresponding to this target path identifier is the target connecting path. The target connecting path and the target intersection travel path are smoothly fitted using a B-spline curve to generate a continuous and smooth path indicating the vehicle's journey from the current road through the first intersection to the next road.

[0113] In other embodiments of this application, the on-board device can determine the navigation decision scheme instructing the vehicle to travel through the traffic light decision module. The traffic light passage decision module can receive the following inputs: intersection information of the first intersection, the target intersection travel path from the intersection path decision module, the lane scan topology map from the lane scan module, and the navigation path. The traffic light passage decision module can determine the passage status of the vehicle entering the first intersection from the current travel road and generate the corresponding passage control command in the following manner.

[0114] Specifically, the traffic light decision-making module can use an association algorithm to associate traffic lights with other intersection elements (such as zebra crossings, stop lines, etc.) at the first intersection. During this process, the module can use coarse traffic light information provided in the navigation path to assist in achieving precise association, thereby determining the correspondence between the traffic lights at the first intersection and other intersection elements. Based on the vehicle's travel direction (which can be indicated by the navigation path) and the target intersection's travel path, and combined with spatial projection matching using the lane scan topology map, the module identifies the target traffic lights the vehicle needs to focus on. For example, if the target intersection's travel path indicates going straight, then the straight-ahead light is focused; if the target intersection's travel path indicates turning left or right, then the left-turn light or right-turn light is focused respectively. Based on the projection of the target traffic light onto the lane scan topology map, its status (red, yellow, or green) and precise location are determined, thereby generating corresponding longitudinal control commands. For example, when the target traffic light is red or yellow, a passage instruction to decelerate to a stop is generated; when the target traffic light changes from red to green, a passage instruction to start or accelerate can be generated to control the passage status of vehicles entering the first intersection.

[0115] In other embodiments of this application, the on-board device can determine a navigation decision scheme instructing the vehicle to travel using a speed limit decision module. The speed limit decision module can be used to determine the target speed limit that the vehicle should follow on the current road. The speed limit decision module can receive the following inputs: road information (such as speed limit sign information of the current road perceived by a camera) and / or road speed limit information provided in the navigation path. The speed limit decision module generates the target speed limit included in the navigation decision scheme.

[0116] Specifically, the speed limit decision module can determine a navigation decision scheme to instruct the vehicle's travel based on the navigation path and road information. This navigation decision scheme includes the target speed limit for the vehicle on the current road. Determining the navigation decision scheme based on the navigation path and road information specifically includes: when the road information includes speed limit sign information for the current road, the speed limit decision module can process the speed limit sign information using an association algorithm to obtain reference speed limit sign information. The association algorithm can be used to verify the validity of the speed limit sign information, for example, by spatiotemporally associating the speed limit value corresponding to the perceived speed limit sign information with the speed limit corresponding to the navigation path, to filter out misidentifications or outliers, thereby obtaining reliable reference speed limit sign information. Then, the speed limit corresponding to the navigation path is calibrated based on the reference speed limit sign information to obtain the target speed limit, where the target speed limit is consistent with or has an error within a preset range compared to the speed limit corresponding to the reference speed limit sign information. When the road information does not include speed limit sign information for the current road, the speed limit decision module uses the speed limit corresponding to the navigation path as the target speed limit.

[0117] S104: Control vehicle movement based on navigation decision-making scheme.

[0118] In some embodiments of this application, the vehicle-mounted device can control the vehicle's movement based on the navigation decision scheme determined in S103.

[0119] based on Figure 9 The vehicle control method shown allows the onboard equipment to... Figure 1 The example environmental perception model uses an environmental scanning layer to perceive road information of the vehicle's current route in real time. Then, based on the various modules of the environmental decision layer, a navigation decision scheme is determined to guide the vehicle's movement based on the road information and navigation path. This method can provide accurate navigation decision schemes for the vehicle without relying on high-resolution maps (HD Maps), enabling the vehicle to safely and efficiently reach its destination, thus improving the user's driving experience. Furthermore, by not relying on HD Maps, the cost of map updates is reduced.

[0120] Based on the method provided in this application, navigation-assisted driving functions can be achieved without relying on high-precision maps or LiDAR, requiring only a camera, standard navigation path, and standard map, resulting in low cost. Furthermore, compared to traditional multi-level information processing flows, this application adopts a simplified "perception → decision → execution" process, which reduces information loss during transmission and improves the accuracy and timeliness of navigation decisions compared to the traditional "perception → environment reconstruction → map fusion → environment model → decision → planning" process. Moreover, the environmental perception module in this application processes raw perception data directly, without relying on complex structured fitting results, enabling more flexible adaptation to complex road scenarios and improving navigation reliability.

[0121] Based on the vehicle control method provided in the embodiments of this application, this application also provides a vehicle control device 800.

[0122] For example, Figure 10 A schematic diagram of a vehicle control device 800 is shown.

[0123] refer to Figure 10 The vehicle control device 800 includes a perception module 801, a decision module 802, and a control module 803. The perception module 801 acquires road information of the vehicle's current driving route, wherein the current driving route includes multiple lanes, and the road information includes at least lane line position data for the multiple lanes. The perception module 801 acquires a navigation path, which at least indicates the next driving route from the current driving route. The decision module 802 determines a navigation decision scheme based on the navigation path and road information, which at least instructs the vehicle to travel from its current position to the next driving route via a target lane, wherein the target lane is a lane determined from the multiple lanes. The control module 803 controls the vehicle's movement according to the navigation decision scheme.

[0124] It is understood that the methods mentioned in the embodiments of this application can be applied to in-vehicle equipment in vehicles. Figure 11 This is a schematic diagram of a possible functional framework of a vehicle 100 provided in an embodiment of this application.

[0125] like Figure 11 As shown, the functional framework of vehicle 100 may include various subsystems, such as Figure 11The diagram shows a sensor system 110, a control system 120, one or more peripheral devices 130 (one is shown as an example), a power supply 140, and an onboard device 150. Optionally, the vehicle 100 may also include other functional systems, such as an engine system that powers the vehicle 100, etc., which are not limited herein. Optionally, the vehicle 100 may include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and component of the vehicle 100 can be interconnected via wired or wireless means.

[0126] The sensor system 110 may include several detection devices that can sense the measured information and convert the sensed information into electrical signals or other desired forms of information output according to a certain rule. Figure 11 As shown, these detection devices may include a global positioning system (GPS), a vehicle speed sensor (112), an inertial measurement unit (IMU), etc., and this application does not limit them.

[0127] The Global Positioning System 111 is a system that uses GPS positioning satellites to perform real-time positioning and navigation globally. In this application, the Global Positioning System 111 can be used to achieve real-time positioning of the vehicle 100 and provide the geographical location information of the vehicle 100. The vehicle speed sensor 112 is used to detect the vehicle speed of the vehicle 100. The inertial measurement unit 113 may include a combination of an accelerometer and a gyroscope, and is a device for measuring the angular rate and acceleration of the vehicle 100.

[0128] The control system 120 may include a steering unit 121 and a braking unit 122, etc.

[0129] Steering unit 121 can represent a system for adjusting the direction of travel of vehicle 100, and may include, but is not limited to, a steering wheel or other structural device for adjusting or controlling the direction of travel of vehicle 100. Braking unit 122 can represent a system for slowing down the speed of vehicle 100, and may also be referred to as a vehicle braking system. It may include, but is not limited to, a brake controller, a reducer, or other structural device for slowing down vehicle 100. In practical applications, braking unit 122 can use friction to slow down the tires of vehicle 100, thereby slowing down the speed of vehicle 100.

[0130] Peripheral device 130 may include several components, such as Figure 11 The illustrated components include a traffic access system 131, a touchscreen 132, a user interface 133, etc. The traffic access system 131 is used to enable network access between the vehicle 100 and other devices besides the vehicle.

[0131] In practical applications, the access system 131 can employ wireless or wired access technologies to enable network access between the vehicle 100 and other devices. Wired access technology can refer to access between the vehicle 100 and other devices via network cables or fiber optic cables. Wireless access technologies include, but are not limited to, Global System for Mobile Communications (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Time Division Code Division Multiple Access (TD-SCDMA), Long Term Evolution (LTE), Wireless Local Area Networks (WLAN) (such as Wireless Fidelity (Wi-Fi) networks), Bluetooth (BT), Global Navigation Satellite System (GNSS), Frequency Modulation (FM), Near Field Communication (NFC), and Infrared (IR) technologies, among others.

[0132] The touchscreen 132 can be used to detect operation commands on the touchscreen 132. For example, the user can perform touch operations on the content data displayed on the touchscreen 132 according to actual needs to achieve the corresponding function, such as playing music, video, or other multimedia files. The user interface 133 can specifically be a touch panel, used to detect operation commands on the touch panel. The user interface 133 can also be a physical button or a mouse. The user interface 133 can also be a display screen, used to output data and display images or data. Optionally, the user interface 133 can also be at least one device belonging to the category of peripheral devices, such as a touchscreen, microphone, and speaker.

[0133] Several functions of vehicle 100 are controlled and implemented by on-board equipment 150. On-board equipment 150 may include multiple processors such as processor 151, chassis domain controller (CDC) 152, mobility domain controller (MDC) 153, telematics box (T-BOX) 154, as well as memory 155 (also referred to as storage device) and gateway 156. In practical applications, the memory 155 may be located inside or outside the on-board equipment 150, for example, as a cache in vehicle 100, etc., and this application does not limit this.

[0134] Among them, processor 151, CDC 152, MDC 153, and T-BOX 154 can be used to run relevant programs or instructions corresponding to programs stored in memory 155 to realize the corresponding functions of vehicle 100, such as the function of calling vehicle camera.

[0135] Memory 155 may include volatile memory, such as RAM; it may also include non-volatile memory, such as ROM, flash memory, HDD, or SSD; or it may include a combination of the above types of memory. Memory 155 can be used to store a set of program code or instructions corresponding to program code, so that processor 151 can call the program code or instructions stored in memory 155 to implement the corresponding functions of vehicle 100. This function includes, but is not limited to, […]. Figure 11 The vehicle functional framework diagram shown includes some or all of the functions.

[0136] Optionally, in addition to storing program code or instructions, the memory 155 may also store information such as road maps, driving routes, and sensor data. The on-board device 150 can be combined with other components in the vehicle functional framework diagram, such as sensors in the sensor system and GPS, to realize the relevant functions of the vehicle 100. For example, the on-board device 150 can control the driving direction or speed of the vehicle 100 based on data input from the sensor system 110; this application does not impose limitations on this.

[0137] This application also provides a vehicle that may include the above-described vehicle-mounted equipment.

[0138] This application also provides a readable storage medium storing one or more programs, which, when executed by an in-vehicle device, enable the in-vehicle device to implement the methods provided in the foregoing embodiments.

[0139] This application also provides a program product that stores instructions that, when executed on an in-vehicle device, enable the in-vehicle device to implement the methods provided in the foregoing embodiments.

[0140] Various embodiments of the mechanisms disclosed in this application can be implemented in hardware, software, firmware, or combinations of these implementation methods. Embodiments of this application can be implemented as computer programs or program code executable on a programmable system, the programmable system including at least one processor, a storage system (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device.

[0141] Program code can be applied to input instructions to execute the functions described in this application and generate output information. The output information can be applied to one or more output devices in a known manner. For the purposes of this application, the processing system includes any system having a processor such as, for example, a digital signal processor, a microcontroller, an application-specific integrated circuit, or a microprocessor.

[0142] The program code can be implemented using a high-level procedural language or an object-oriented programming language to ensure compatibility with the processing system. Assembly language or machine language can also be used when necessary. In fact, the mechanisms described in this application are not limited to any particular programming language. In either case, the language can be a compiled language or an interpreted language.

[0143] In some cases, the disclosed embodiments may be implemented in hardware, firmware, software, or any combination thereof. The disclosed embodiments may also be implemented as instructions carried on or stored thereon by one or more transient or non-transitory machine-readable (e.g., computer-readable) storage media, which may be read and executed by one or more processors. For example, the instructions may be distributed via a network or through other computer-readable media. Therefore, machine-readable media can include any mechanism for storing or transmitting information in a machine-readable (e.g., computer-readable) form, including but not limited to floppy disks, optical disks, CD-ROMs, compact disc-read-only memory (CD-ROMs), magneto-optical disks, read-only memory (ROM), random-access memory (RAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic cards or optical cards, flash memory, or tangible machine-readable storage for transmitting information (e.g., carrier waves, infrared signals, digital signals, etc.) using the Internet in the form of electrical, optical, acoustic, or other forms of propagation signals. Therefore, machine-readable media includes any type of machine-readable medium suitable for storing or transmitting electronic instructions or information in a machine-readable (e.g., computer-readable) form.

[0144] In the accompanying drawings, some structural or methodological features may be shown in a specific arrangement and / or order. However, it should be understood that such a specific arrangement and / or order may not be necessary. Rather, in some embodiments, these features may be arranged in a manner and / or order different from that shown in the illustrative drawings. Furthermore, the inclusion of structural or methodological features in a particular figure does not imply that such features are required in all embodiments, and in some embodiments, these features may be omitted or may be combined with other features.

Claims

1. A vehicle control method, characterized in that, The method includes: Obtain road information of the vehicle on the current driving road, wherein the current driving road includes multiple lanes, and the road information includes at least lane line position data of the multiple lanes; Obtain a navigation path, which is at least used to indicate the next driving road of the current driving road; Based on the navigation path and the road information, a navigation decision scheme is determined to instruct the vehicle to travel. The navigation decision scheme is at least used to instruct the vehicle to travel from its current location to the next travel road via a target lane, wherein the target lane is a lane determined from the plurality of lanes. The vehicle is controlled to move according to the navigation decision scheme.

2. The method according to claim 1, characterized in that, Obtaining road information for the vehicle's current route includes one or more of the following: Traffic flow data for the multiple lanes is obtained based on traffic flow scanning; Lane line position data of the multiple lanes are obtained based on lane scanning; Accessible area data for the multiple lanes is obtained based on accessible area scanning.

3. The method according to claim 2, characterized in that, The acquisition of traffic flow data for the multiple lanes based on traffic flow scanning includes: The traffic flow scanning module clusters and fits the dynamic obstacle information of each lane perceived by the vehicle's sensors to obtain traffic flow data for each lane. The traffic flow data for each lane includes the identification, average speed, and orientation of at least one dynamic obstacle.

4. The method according to claim 3, characterized in that, The process of obtaining passable area data for the multiple lanes based on passable area scanning includes: The passable area scanning module projects the spatial occupancy information of the current driving road sensed by the sensor onto a preset scanning line. Based on the scan lines, the boundary of the passable area of ​​the current driving road and the passable area defined by the boundary are determined.

5. The method according to claim 4, characterized in that, The vehicle's current lane is different from the target lane, and the navigation decision scheme includes a lane-changing instruction to switch from the current lane to the target lane; The step of determining a navigation decision scheme to instruct the vehicle to travel based on the navigation path and the road information includes: Based on the navigation path, the target driving direction from the current lane to the target lane is determined, and based on the target driving direction, the lane change direction in the lane change instruction is determined; The target lane-changing time for the vehicle to switch from the current lane to the target lane is determined based on the dynamic path planning algorithm, and the lane-changing time in the lane-changing instruction is determined based on the target lane-changing time.

6. The method according to claim 4, characterized in that, The navigation path is used to indicate the first intersection encountered when traveling from the current road to the next road; The road information also includes the intersection information of the first intersection, which includes the type of the first intersection, stop line location data, zebra crossing location data, guide line location data and / or traffic light status; The navigation decision-making scheme also includes the target intersection driving route that passes through the first intersection; The step of determining a navigation decision scheme to instruct the vehicle to travel based on the navigation path and the road information includes: Based on the intersection information of the first intersection, the driving path to the target intersection is determined using the Bézier curve algorithm.

7. The method according to claim 6, characterized in that, The intersection elements of the first intersection include stop lines, zebra crossings, guide lines, and / or traffic lights; The methods for determining the intersection information of the first intersection include: Clustering the intersection features of the first intersection yields at least one set of intersection features, wherein each set of intersection features corresponds to a type of intersection feature; The at least one set of intersection features is filtered to remove intersection feature points that do not meet the preset stability conditions, thereby obtaining the intersection information of the first intersection.

8. The method according to claim 6, characterized in that, The step of determining the driving path at the target intersection using the Bézier curve algorithm includes: Based on the intersection information of the first intersection, at least one candidate intersection driving path is determined using the Bézier curve algorithm, wherein the at least one candidate intersection driving path is used to indicate that the vehicle passes through the first intersection. The driving paths of each candidate intersection are evaluated based on traffic flow data and passable area data at least for each candidate intersection, and the driving path of the candidate intersection with the highest score among the driving paths of each candidate intersection is determined to obtain the driving path of the target intersection.

9. The method according to claim 6, characterized in that, The navigation decision scheme also includes a continuous and smooth path from the current driving road through the first intersection to the next driving road; The step of determining a navigation decision scheme to instruct the vehicle to travel based on the navigation path and the road information further includes: Based on the lane change command, a target connecting route that matches the driving path at the target intersection is determined. The target connecting path is smoothly fitted to the target intersection driving path to generate a continuous smooth path from the current driving road through the first intersection to the next driving road.

10. The method according to claim 9, characterized in that, The step of determining a target connecting route that matches the driving path at the target intersection based on the lane change instruction includes: A recursive algorithm is used to determine at least one candidate connecting path based on the road information, and the target connecting path is determined from the at least one candidate connecting path according to the lane change instruction.

11. The method according to claim 6, characterized in that, The navigation decision scheme also includes the vehicle's passage status from the current driving road to the first intersection; The step of determining a navigation decision scheme to instruct the vehicle to travel based on the navigation path and the road information further includes: By using an association algorithm, the traffic lights at the first intersection are associated with at least one intersection element in the intersection information to determine the driving direction of the vehicle, and the target traffic light is determined based on the driving direction of the vehicle and the driving path to the target intersection. Based on the state of the target traffic light, generate traffic control commands corresponding to acceleration, constant speed, deceleration, or stopping.

12. The method according to claim 1, characterized in that, The navigation decision-making scheme also includes the target speed limit of the current driving road; When the road information includes speed limit sign information for the currently traveling road, determining the navigation decision scheme to instruct the vehicle to travel based on the navigation route and the road information includes: The speed limit sign information is processed based on the association algorithm to obtain reference speed limit sign information; Based on the reference speed limit sign information, the speed limit corresponding to the navigation path is calibrated to obtain the target speed limit, wherein the target speed limit is the same as the speed limit corresponding to the reference speed limit sign information or the error is within a preset range.

13. The method according to claim 12, characterized in that, If the road information does not include speed limit sign information for the currently traveling road, the step of determining a navigation decision scheme to instruct the vehicle to travel based on the navigation route and the road information further includes: The speed limit corresponding to the navigation path is taken as the target speed limit.

14. A vehicle control device, characterized in that, It includes a perception module, a decision-making module, and a control module; The perception module is used to acquire road information of the vehicle on the current driving road, wherein the current driving road includes multiple lanes, and the road information includes at least lane line position data of the multiple lanes; The perception module is used to obtain a navigation path, and the navigation path is at least used to indicate the next driving road of the current driving road; The decision module is used to determine a navigation decision scheme to instruct the vehicle to travel based on the navigation path and the road information. The navigation decision scheme is at least used to instruct the vehicle to travel from the current position to the next travel road via a target lane, wherein the target lane is a lane determined from the plurality of lanes. The control module is used to control the vehicle's movement according to the navigation decision scheme.

15. A vehicle-mounted device, characterized in that, include: One or more processors and one or more memories, the memories being coupled to the processors; the one or more memories storing one or more programs, which, when executed by the one or more processors, cause the vehicle-mounted device to perform the method of any one of claims 1 to 13.

16. A vehicle, characterized in that, The vehicle includes the on-board equipment as described in claim 15.

17. A readable storage medium, characterized in that, The readable storage medium stores instructions that, when executed on the vehicle-mounted device, cause the vehicle-mounted device to perform the method of any one of claims 1 to 13.

18. A program product, characterized in that, When the program product is run on an in-vehicle device, it causes the in-vehicle device to perform the method of any one of claims 1 to 13.

Images