Navigation method, apparatus, and vehicle

Abstract

A navigation method, apparatus, and vehicle are disclosed. The method includes: acquiring first-type guidance information, which indicates the accessibility of lanes in each of at least one road, wherein the at least one road is the route the vehicle must take to travel from its current location to a target location; the first-type guidance information is generated based on historical road topology; acquiring second-type guidance information, which indicates the accessibility of lanes in a target road perceived by the vehicle; and controlling the vehicle to travel to the target location via a first lane in the target road, based on the first-type guidance information and / or the second-type guidance information. This solution can be applied to the field of intelligent driving for electric vehicles, new energy vehicles, and other vehicles. It can reduce the number of invalid lane changes and / or the probability of deviation during the vehicle's journey to its destination, thereby improving driving efficiency without relying on high-precision maps.

Description

Technical Field

[0001] This application relates to the field of intelligent driving, and more specifically, to a navigation method, device, and vehicle. Background Technology

[0002] With the rapid development of the automotive industry, many driver assistance and autonomous driving technologies have emerged, which can reduce driving stress and improve safety and traffic efficiency. Currently, most autonomous driving technologies rely on high-precision maps for navigation. However, high-precision maps have drawbacks such as high collection and production costs, long processing times, insufficient coverage, and difficulty in ensuring data freshness, making it difficult to promote autonomous driving technologies that rely on high-precision maps nationwide or globally.

[0003] Intelligent vehicles use sensors such as cameras and LiDAR to detect static elements (such as lanes, roads, traffic lights, and signs) within a certain range around the vehicle in real time. This allows them to model the surrounding environment, enabling the vehicle to navigate in complex and ever-changing traffic conditions without relying on high-precision maps. However, the current vehicle sensors have limited sensing range, and the complex and ever-changing navigation rules during actual driving can lead to problems such as multiple lane changes, invalid lane changes, and even deviations from the intended path.

[0004] Therefore, a navigation scheme that can reduce unnecessary lane changes and veergence by vehicles is urgently needed. Summary of the Invention

[0005] This application provides a navigation method, device, and vehicle that can reduce the number of invalid lane changes and / or the probability of veergence during the vehicle's journey to a target location without relying on high-precision maps, thereby improving the vehicle's driving efficiency.

[0006] Firstly, a navigation method is provided that can be executed by a vehicle, for example, by the vehicle's computing platform, or by a chip or circuitry used in the vehicle.

[0007] The method includes: acquiring a first type of guidance information, which indicates the accessibility of each lane in at least one road, wherein the at least one road is the route that the vehicle needs to take from the current location to the target location, and the first type of guidance information is generated based on historical road topology; acquiring a second type of guidance information, which indicates the accessibility of each lane in the target road perceived by the vehicle; and controlling the vehicle to travel to the target location via a first lane in the target road according to the first type of guidance information and / or the second type of guidance information.

[0008] It is understandable that at least one road includes the target road, which is either the current road the vehicle is on or the road the vehicle needs to travel to the target location.

[0009] In the above technical solution, the first type of guidance information can provide guidance information beyond line of sight (i.e., beyond the vehicle's perception range). Thus, even without high-precision map guidance, the vehicle can acquire information considering far-field road conditions, enabling it to determine whether and / or when to change lanes in complex road conditions, reducing the probability of invalid lane changes. However, in actual implementation, some road sections may lack the first type of guidance information, or the guidance information on some road sections may not match the first type of guidance information due to temporary road repairs, road closures, etc. This technical solution can verify the validity of the first type of guidance information. When there is a significant difference between the first type of guidance information and the actual passable road information, navigation can be switched to the second type of guidance information or other guidance information, which helps improve the robustness and reliability of the vehicle's intelligent driving system, thereby reducing the probability of the vehicle deviating from its course while traveling to the target location and improving driving efficiency.

[0010] In conjunction with the first aspect, in certain implementations of the first aspect, controlling a vehicle to travel to a target location via a first lane in a target road according to a first type of guidance information and / or a second type of guidance information includes: during the process of controlling the vehicle to travel according to the first type of guidance information, when the offset between the boundary of the target road indicated by the first type of guidance information and the boundary of the target road perceived by the vehicle is greater than or equal to a distance threshold, switching to controlling the vehicle to travel to the target location according to the second type of guidance information; or, when the number of drivable lanes of the target road indicated by the first type of guidance information is inconsistent with the number of drivable lanes of the target road perceived by the vehicle, switching to controlling the vehicle to travel to the target location according to the second type of guidance information.

[0011] In the above technical solution, when the road boundary or number of drivable lanes indicated by the guidance information based on historical road topology does not match the road boundary or number of drivable lanes perceived in real time by the vehicle's perception system, switching to guidance information generated based on the vehicle's real-time perception information to control the vehicle's driving can reduce the probability of deviation caused by changes in road passability.

[0012] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: during the process of controlling the vehicle to drive according to the second type of guidance information, when the offset between the boundary of the target road indicated by the first type of guidance information and the boundary of the target road perceived by the vehicle is less than a distance threshold, or when the number of drivable lanes of the target road indicated by the first type of guidance information is consistent with the number of drivable lanes of the target road perceived by the vehicle, switching to controlling the vehicle to drive towards the target location according to the first type of guidance information.

[0013] In the aforementioned technical solution, because the first type of guidance information can provide information that takes far-field road conditions into account, the probability of invalid lane changes in the near field can be reduced when the vehicle navigates based on this guidance information. For example, when a vehicle is traveling towards a target location and needs to turn left at the target intersection, the navigation information instructs the vehicle to move right first and then left, allowing the vehicle to travel along the middle lane without making multiple invalid lane changes. Therefore, when the first type of guidance information matches the real road topology, switching to navigation based on the first type of guidance information can improve the vehicle's human-likeness and driving efficiency.

[0014] It should be noted that the far field involved in this application can be understood as a position that is far away from the vehicle and cannot be perceived by the vehicle's own perception system; the near field involved in this application can be understood as a position that is close to the vehicle and can be perceived by the vehicle's perception system.

[0015] In conjunction with the first aspect, in some implementations of the first aspect, the second type of guidance information is generated by the navigation neural network processing navigation guidance information and road element information. The navigation guidance information indicates the R roads that the vehicle needs to pass through from the current position to the target position and the direction of travel. The R roads include the target road. The road element information indicates at least one lane-level element perceived by the vehicle at the current position. The at least one lane-level element indicates the lane boundary and / or lane centerline of each of the S lanes. The S lanes are lanes in the R roads, and S and R are both positive integers.

[0016] In the above technical solution, processing various types of information through a navigation neural network helps to improve the efficiency and real-time performance of acquiring the second type of guidance information.

[0017] In conjunction with the first aspect, in certain implementations of the first aspect, controlling a vehicle to travel towards a target location via a first lane of a target road based on first-type guidance information and / or second-type guidance information includes: when the navigation directions corresponding to multiple lanes in the target road indicated by the first-type guidance information are opposite to the navigation directions corresponding to multiple lanes indicated by the second-type guidance information, controlling the vehicle to travel towards the target location via the first lane based on the second-type guidance information or based on third-type guidance information; wherein, the third-type guidance information indicates the passability of the lanes included in the target road, and the third-type guidance information is generated differently from the first-type guidance information and the second-type guidance information.

[0018] In the above technical solution, the accuracy of the first type of guidance information is verified by the second type of guidance information. When it is determined that the first type of guidance information differs significantly from the actual guidance trend, the vehicle is controlled to drive according to other guidance information, which helps to reduce the probability of vehicle deviation and / or invalid lane change.

[0019] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: during the process of controlling the vehicle's movement according to the third type of guidance information, when the navigation directions corresponding to the multiple lanes indicated by the third type of guidance information are opposite to the navigation directions corresponding to the multiple lanes indicated by the second type of guidance information, the method controls the vehicle to move towards the target location via the first lane according to the second type of guidance information.

[0020] In conjunction with the first aspect, in some implementations of the first aspect, controlling a vehicle to travel towards a target location via a first lane among multiple lanes, based on a first type of guidance information and a second type of guidance information, includes: when the navigation directions corresponding to the multiple lanes in the target road indicated by the first type of guidance information match the navigation directions corresponding to the multiple lanes indicated by the second type of guidance information, fusing the first type of guidance information and the second type of guidance information to obtain a fusion result; and controlling the vehicle to travel towards the target location via the first lane based on the fusion result.

[0021] In the above technical solution, using the second type of guidance information to correct the first type of guidance information can improve the matching degree between the first type of guidance information and the current road conditions, thereby reducing the probability of vehicle deviation and / or invalid lane change.

[0022] In conjunction with the first aspect, in some implementations of the first aspect, the navigation directions corresponding to multiple lanes indicated by the first type of guidance information are matched with the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, including: lanes whose passability level indicated by the first type of guidance information is greater than or equal to a threshold level include lanes whose passability level indicated by the second type of guidance information is greater than or equal to a threshold level.

[0023] In conjunction with the first aspect, in some implementations of the first aspect, the second type of guidance information is generated based on predefined rules.

[0024] In conjunction with the first aspect, in some implementations of the first aspect, the method further includes: determining N1 sampling results corresponding to the target road based on environmental perception information, each sampling result indicating at least one passable lane corresponding to one of the N1 sampling planes, and the distance between any two adjacent planes in the N1 sampling planes being a first distance; determining M1 intersection guidance information associated with the N1 sampling planes based on first navigation information, where M1 and N1 are both positive integers; wherein, the environmental perception information indicates the road boundary of the target road perceived by the vehicle, and / or the lane boundary associated with the target road, the first navigation information indicates at least one passable lane in the target road used by the vehicle to travel to the target location, and the position of at least one passable lane in the target road; determining M1*N1 combined results based on the N1 sampling results and M1 intersection guidance information, each combined result indicating a passable lane among multiple lanes and its position in the target road; and determining a second type of guidance information based on the M1*N1 combined results.

[0025] In a second aspect, a navigation device is provided, comprising an acquisition unit and a processing unit, wherein the acquisition unit is configured to: acquire first type of guidance information, the first type of guidance information indicating the passability of lanes contained in each of at least one road, the at least one road being the route that a vehicle must take from its current location to a target location, and the first type of guidance information being generated based on historical road topology; the acquisition unit is further configured to: acquire second type of guidance information, the second type of guidance information indicating the passability of lanes contained in a target road perceived by the vehicle; and the processing unit is configured to: control the vehicle to travel to the target location via a first lane in the target road according to the first type of guidance information and / or the second type of guidance information.

[0026] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit is configured to: switch to controlling the vehicle to drive towards the target location according to the second type of guidance information when the offset between the boundary of the target road indicated by the first type of guidance information and the boundary of the target road perceived by the vehicle is greater than or equal to a distance threshold during the process of controlling the vehicle to drive according to the first type of guidance information; or switch to controlling the vehicle to drive towards the target location according to the second type of guidance information when the number of drivable lanes of the target road indicated by the first type of guidance information is inconsistent with the number of drivable lanes of the target road perceived by the vehicle.

[0027] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit is further configured to: during the process of controlling the vehicle to drive according to the second type of guidance information, when the offset between the boundary of the target road indicated by the first type of guidance information and the boundary of the target road perceived by the vehicle is less than a distance threshold, or when the number of drivable lanes of the target road indicated by the first type of guidance information is consistent with the number of drivable lanes of the target road perceived by the vehicle, switch to controlling the vehicle to drive towards the target location according to the first type of guidance information.

[0028] In conjunction with the second aspect, in some implementations of the second aspect, the second type of guidance information is generated by the navigation neural network processing navigation guidance information and road element information. The navigation guidance information indicates the R roads that the vehicle needs to pass through from the current position to the target position and the direction of travel. The R roads include the target road. The road element information indicates at least one lane-level element perceived by the vehicle at the current position. The at least one lane-level element indicates the lane boundary and / or lane centerline of each of the S lanes. The S lanes are lanes in the R roads, and S and R are both positive integers.

[0029] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit is used to: when the navigation directions corresponding to multiple lanes in the target road indicated by the first type of guidance information are opposite to the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, control the vehicle to travel to the target location via the first lane according to the second type of guidance information or according to the third type of guidance information; wherein, the third type of guidance information indicates the passability of the lanes included in the target road, and the third type of guidance information is generated in a different way than the first type of guidance information and the second type of guidance information.

[0030] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit is further configured to: during the process of controlling the vehicle's movement according to the third type of guidance information, when the navigation directions corresponding to the multiple lanes indicated by the third type of guidance information are opposite to the navigation directions corresponding to the multiple lanes indicated by the second type of guidance information, control the vehicle to move towards the target location via the first lane according to the second type of guidance information.

[0031] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit is used to: when the navigation directions corresponding to multiple lanes in the target road indicated by the first type of guidance information match the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, fuse the first type of guidance information and the second type of guidance information to obtain a fusion result; and control the vehicle to travel to the target location via the first lane according to the fusion result.

[0032] In conjunction with the second aspect, in some implementations of the second aspect, the navigation directions corresponding to multiple lanes indicated by the first type of guidance information are matched with the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, including: lanes whose passability level indicated by the first type of guidance information is greater than or equal to the level threshold include lanes whose passability level indicated by the second type of guidance information is greater than or equal to the level threshold.

[0033] In conjunction with the second aspect, in some implementations of the second aspect, the second type of guidance information is generated based on predefined rules.

[0034] In conjunction with the second aspect, in some implementations of the second aspect, the processing unit is further configured to: determine N1 sampling results corresponding to the target road based on environmental perception information, each sampling result indicating at least one passable lane corresponding to one of the N1 sampling planes, the distance between any two adjacent planes in the N1 sampling planes being a first distance; determine M1 intersection guidance information associated with the N1 sampling planes based on first navigation information, where M1 and N1 are both positive integers; wherein, the environmental perception information indicates the road boundary of the target road perceived by the vehicle, and / or the lane boundary associated with the target road, the first navigation information indicates at least one passable lane in the target road used by the vehicle to travel to the target location, and the position of at least one passable lane in the target road; determine M1*N1 combined results based on the N1 sampling results and M1 intersection guidance information, each combined result indicating a passable lane among multiple lanes and its position in the target road; and determine a second type of guidance information based on the M1*N1 combined results.

[0035] Thirdly, a navigation device is provided, the device comprising: a processor for executing a computer program stored in the memory, such that the device performs the method in any possible implementation of the first aspect described above.

[0036] In conjunction with the third aspect, in some implementations of the third aspect, the device also includes a memory.

[0037] Fourthly, a computer program product is provided, comprising: computer program code, which, when executed on a computer or processor, causes the computer or processor to perform the method in any possible implementation of the first aspect.

[0038] It should be noted that the above computer program code can be stored in whole or in part on a storage medium, which can be packaged together with the processor or packaged separately from the processor.

[0039] Fifthly, a computer-readable storage medium is provided, the computer-readable medium storing instructions that, when executed by a processor, cause the processor to implement the method in any possible implementation of the first aspect.

[0040] In a sixth aspect, a chip is provided, the chip including circuitry for performing the method in any of the possible implementations of the first aspect described above.

[0041] In a seventh aspect, a vehicle is provided that includes means as in any possible implementation of the second or third aspect, or the vehicle includes a computer-readable storage medium as in any possible implementation of the fifth aspect, or the vehicle includes a chip as in any possible implementation of the sixth aspect, or the vehicle is loaded with a computer program product as in any possible implementation of the fourth aspect.

[0042] In conjunction with the seventh aspect, in some implementations of the seventh aspect, the vehicle is a vehicle in a broad sense, such as a means of transportation (e.g., commercial vehicles, passenger cars, motorcycles, flying cars, trains, etc.), industrial vehicles (e.g., forklifts, trailers, tractors, etc.), engineering vehicles (e.g., excavators, bulldozers, cranes, etc.), agricultural equipment (e.g., lawnmowers, harvesters, etc.), amusement equipment, toy vehicles, etc. In practical implementation, the vehicle can also be a road vehicle, a water vehicle, an air vehicle, industrial equipment, agricultural equipment, or other intelligent driving equipment such as entertainment equipment.

[0043] For the beneficial effects not described in detail in aspects two through seven, please refer to the description in aspect one, which will not be repeated here. Attached Figure Description

[0044] Figure 1 This is a functional block diagram of the vehicle provided in the embodiments of this application;

[0045] Figure 2 This is a schematic diagram of the autonomous driving system architecture provided in an embodiment of this application;

[0046] Figure 3 This is a schematic flowchart of a method for generating guidance information provided in an embodiment of this application;

[0047] Figure 4 This is a schematic diagram of the neural network for generating guidance information provided in an embodiment of this application;

[0048] Figure 5 This is a schematic diagram of the training process of the neural network provided in the embodiments of this application;

[0049] Figure 6 This is a schematic diagram of the guidance information provided in the embodiments of this application;

[0050] Figure 7 This is another illustrative flowchart of the method for generating guidance information provided in the embodiments of this application;

[0051] Figure 8 This is a schematic diagram illustrating the application scenarios involved in the embodiments of this application;

[0052] Figure 9 This is a schematic diagram illustrating the relationship between grid points and key points associated with the guidance information involved in the embodiments of this application;

[0053] Figure 10 This is a schematic diagram illustrating the relationship between grid points and query points associated with the guidance information involved in the embodiments of this application;

[0054] Figure 11 This is a schematic diagram illustrating the relationship between grid points, key points, and query points associated with the guidance information involved in the embodiments of this application;

[0055] Figure 12 This is another illustrative flowchart of the method for generating guidance information provided in the embodiments of this application;

[0056] Figure 13 This is yet another schematic diagram illustrating an application scenario provided in the embodiments of this application;

[0057] Figure 14 This is a schematic diagram of sampling matching and results provided in an embodiment of this application;

[0058] Figure 15 This is yet another schematic diagram of the sampling matching and results provided in the embodiments of this application;

[0059] Figure 16 This is another schematic diagram of the guidance information provided in the embodiments of this application;

[0060] Figure 17 This is a schematic diagram illustrating the result verification provided in an embodiment of this application;

[0061] Figure 18 This is a schematic flowchart illustrating the method for providing navigation information according to an embodiment of this application;

[0062] Figure 19 This is a schematic block diagram of the device provided in the embodiments of this application;

[0063] Figure 20 This is another schematic block diagram of the device provided in the embodiments of this application. Detailed Implementation

[0064] The technical solutions in this application will now be described with reference to the accompanying drawings.

[0065] Figure 1This is a functional block diagram of a vehicle provided in an embodiment of this application. For example... Figure 1 As shown, the vehicle 100 may include a perception system 120 and a computing platform 150. The perception system 120 may include several sensors for sensing information about the surrounding environment of the vehicle 100. For example, the perception system 120 may include a positioning system, which may be a Global Positioning System (GPS), a BeiDou system, or another positioning system. As another example, the perception system 120 may also include one or more of the following: an inertial measurement unit (IMU), lidar, millimeter-wave radar, ultrasonic radar, and a camera device.

[0066] Some or all of the functions of vehicle 100 can be controlled by computing platform 150. Computing platform 150 may include processors 151 to 15n. A processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction read and execute capabilities, such as a central processing unit (CPU), microprocessor, graphics processing unit (GPU) (which can be understood as a type of microprocessor), or digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. These logical relationships are fixed or reconfigurable. For example, the processor may be a hardware circuit implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as a field-programmable gate array (FPGA). In reconfigurable hardware circuits, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the process of the processor loading instructions to implement related functions. Furthermore, the processor can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (N1PU), tensor processing unit (TPU), deep learning processing unit (DPU), etc. In addition, the computing platform 150 may also include a memory for storing instructions. Some or all of the processors 151 to 15n can call the instructions in the memory to implement the corresponding functions.

[0067] The computing platform 150 can control the operation of the intelligent driving system, which may include an advanced driving assistance system (ADAS) and an autonomous driving system (ADS). The intelligent driving system utilizes various sensors on the vehicle (including but not limited to: LiDAR, millimeter-wave radar, cameras, ultrasonic sensors, GPS, and inertial measurement units) to acquire information from the vehicle's surroundings, and analyzes and processes this information to achieve functions such as obstacle perception, target recognition, vehicle localization, path planning, and driver monitoring / alerts, thereby improving the safety, automation, and comfort of driving.

[0068] At different levels of autonomous driving (or intelligent driving levels, ranging from L0 to L5, totaling six levels), intelligent driving systems can achieve different levels of automated driving assistance based on artificial intelligence algorithms and information acquired by multiple sensors. These levels of autonomous driving are based on the classification standards of the Society of Automotive Engineers (SAE). Specifically, L0 is no automation; L1 is driver assistance; L2 is partial automation; L3 is conditional automation; L4 is high automation; and L5 is full automation. At levels L1 to L3, the task of monitoring road conditions and reacting is jointly completed by the driver and the system, requiring the driver to take over dynamic driving tasks. Levels L4 and L5 allow the driver to completely transform into a passenger. Currently, the functions that intelligent driving systems can achieve mainly include, but are not limited to: adaptive cruise control, automatic emergency braking, automatic parking, blind spot monitoring, forward cross-traffic alert / braking, rear cross-traffic alert / braking, forward collision warning, lane departure warning, lane keeping assist, rear collision warning, traffic sign recognition, traffic jam assist, and highway assist. It should be understood that the above-mentioned functions can have specific modes at different levels of autonomous driving (L0-L5). The higher the level of autonomous driving, the more intelligent the corresponding mode.

[0069] The following is combined with Figure 2 The specific roles of the sensing system 120 and the computing platform 150 in this application are explained. Figure 2 A schematic block diagram of an autonomous driving system architecture provided in an embodiment of this application is shown. The system includes a perception module 210, a navigation information acquisition module 220, a map information acquisition module 230, and a traffic control module 240. The perception module 210 may include… Figure 1The sensing system 120 shown may include one or more camera devices, or may further include one or more radars; the navigation information acquisition module 220, map information acquisition module 230, and planning control module 240 may each include Figure 1 One or more processors in the computing platform 150 shown. Figure 2 The functions of each module in the system shown are as described in items (I) to (IV) below.

[0070] (a) Perception module 210: used to acquire environmental information around the vehicle and send it to the planning and control module 240. The environmental information may include the road boundary, lane boundary, etc. of the road the vehicle is currently traveling on.

[0071] (ii) Navigation information acquisition module 220: used to acquire navigation information, which indicates the road-level driving direction from the vehicle's current location to the target location. The road-level driving direction includes the vehicle's driving direction within a road, and may also include the vehicle's driving direction at an intersection between two roads.

[0072] For example, the navigation information can be SD map navigation information determined according to the standard definition (SD) map.

[0073] (iii) Map information acquisition module 230: used to acquire road topology information, which indicates the topological relationship between multiple roads within a certain area, and the number of lanes included in each road; or, the road topology information can also indicate the changes in the number of lanes in each road, and the location where the number of lanes changes.

[0074] For example, road topology information may include pre-made map data or information extracted from pre-made map data. The pre-made map data may be roadcode (RC) maps, electric horizon (EHP) data, etc.

[0075] It should be noted that there is a topological relationship between the two roads involved in this application, which can be understood as: the two roads are connected at a certain intersection, so that vehicles can enter the other road from one of the two roads through the intersection.

[0076] (iv) Guidance and Control Module 240: This module generates guidance information and controls the vehicle to travel to the target location based on the guidance information. Specifically, the guidance and control module 240 may include a Type A guidance information generation module 231, a Type B guidance information generation module 232, a verification module 233, and a path planning module 234. The Type A guidance information generation module 231 generates guidance information based on information from the perception module 210 and the navigation information acquisition module 220; the Type B guidance information generation module 232 generates guidance information based on information from the navigation information acquisition module 220 and the map information acquisition module 230. The verification module 233 verifies whether the road structure indicated by the guidance information (such as road boundaries, passable lanes, etc.) is consistent with the road structure perceived by the vehicle, thereby determining the guidance information used for navigation. In some implementations, the verification module 233 can also generate fallback guidance information based on the information from the perception module 210 and the navigation information acquisition module 220, and then compare the fallback guidance information with the guidance information from the A-type guidance information generation module 231 and the B-type guidance information generation module 232 to determine the guidance information used for navigation.

[0077] It should be understood that the above module is only an example, and in actual applications, it may be added or removed as needed. For example, Figure 2 In the system architecture shown, the navigation information acquisition module 220, the map information acquisition module 230, and the planning and control module 240 can be merged into one module.

[0078] The above describes the autonomous driving system architecture provided in the embodiments of this application. The following details the system architecture based on... Figure 2 The autonomous driving system shown implements the method for generating guidance information and the navigation method provided in the embodiments of this application.

[0079] Figure 3 A schematic flowchart illustrating a method for generating guidance information according to an embodiment of this application is shown. This method 300 can be applied to... Figure 1 In the vehicle shown, or the method can be derived from Figure 2 The system execution is shown. More specifically, method 300 can be executed by the planning and control module 240, and more specifically, by the Class A guidance information generation module 241. Method 300 may include:

[0080] S310, Obtain navigation guidance information, which indicates at least one road that the vehicle must take to travel from its current location to the target location, and the driving direction associated with each of the at least one road.

[0081] For example, navigation guidance information can be obtained by preprocessing navigation information.

[0082] In some implementations, the aforementioned target location can be the destination of the vehicle's current journey, or it can be a location the vehicle passes through while traveling to the destination.

[0083] In one example, the associated driving direction for each road can be understood as the vehicle's direction of travel and / or driving deviation within the road. The driving deviation indicates whether the vehicle is traveling closer to the left side of the road, closer to the right side, or in the center. For example, the vehicle's driving deviation within the road can be determined based on the driving direction at the intersection to which the road ends. For instance, if the driving direction at the intersection is straight, the vehicle's driving deviation within the road can be in the center; or, if the driving direction at the intersection is left turn, the vehicle's driving deviation within the road can be closer to the left side. When the road includes a destination, the vehicle's driving deviation within the road can be determined based on the destination's location within the road. For instance, if the destination is on the right side of the road, the vehicle's driving deviation within the road can be closer to the right side; or, if the destination is on the left side of the road, the vehicle's driving deviation within the road can be closer to the left side.

[0084] In another example, when at least one road includes two or more roads, and the vehicle will pass through at least one intersection on its way to the target location, the driving direction associated with each road can be understood as: the driving direction at the intersection that connects to the end of that road.

[0085] For example, the left or right side of the road can be determined relative to the vehicle's coordinate system. The positive direction of the Y-axis of the vehicle coordinate system points to the left side of the road, and the negative direction points to the right side of the road. The origin O of the vehicle coordinate system can be located at the projection point of the rear axle center of the vehicle onto the ground. The positive directions of the X-axis and Z-axis are the direction of the vehicle's front and the direction perpendicular to the vehicle's plane, respectively. Furthermore, the road includes two ends; as the vehicle travels, the end that gets closer to the vehicle is the end of the road, and the end that gets farther away from the vehicle is the beginning of the road.

[0086] For example, the driving direction at each intersection can be indicated by one or more of the following information: ① intersection guidance; ② intersection topology and the target road corresponding to this intersection, wherein the topology at each intersection can indicate the two or more roads connected to the intersection and the direction of each road, and the target road indicates the road to be entered after passing through the intersection; ③ lane information, which indicates the number of lanes included in the road, the guidance supported by each lane and the availability status of each lane guidance (i.e., whether the vehicle can travel in the lane based on the guidance when heading towards the target location).

[0087] S320, Obtain road element information. The road element information indicates at least one lane-level element perceived by the vehicle at the current position. The at least one lane-level element indicates the lane boundary and / or lane centerline of each of the M lanes. The M lanes are the lanes in the road that needs to be traversed from the current position to the target position, and M is a positive integer.

[0088] In some implementations, road element information can be determined based on vehicle perception information. For example, the perception information can be an image and / or a laser point cloud, in which case one or more processors extract one or more road elements from the perception information, including intersection boundaries, lane guide lines, lane center points (or lines), lane boundaries, road boundaries, and lane types. For example, the perception information can be directly processed as an image to extract road elements; alternatively, the perception information can be preprocessed in conjunction with navigation guidance information to improve the fit between the extracted road elements and their ground truth values.

[0089] For example, lane-level elements may include lane boundary lines, lane center lines, and other elements that can determine the lane position and / or boundary.

[0090] S330 inputs navigation guidance information and road element information into the navigation neural network to obtain guidance information 1, which indicates the passability of different positions of M lanes.

[0091] In some implementations, the navigation neural network includes an encoding network and a decoding network. S330 can be further refined as follows: inputting navigation guidance information and road element information into the encoding network to obtain fused environmental features, which indicate the relationship between any two of the M lanes and the driving direction of the vehicle in at least one of the M lanes; inputting the fused environmental features into the decoding network to obtain guidance information 1.

[0092] In practice, M lanes can be associated with one intersection, for example, M lanes are connected to one intersection; or, M lanes can be associated with multiple intersections, for example, some of the M lanes are connected to one intersection, and the remaining M lanes are connected to another intersection.

[0093] For example, the relationship between any two lanes may include: the two lanes being adjacent lanes or non-adjacent lanes on the same road; or, the two lanes being connected lanes on two roads connected by an intersection, meaning that a vehicle can enter the other lane from one of the two lanes.

[0094] In some implementations, the method further includes: determining the road trend line and road width of road 1 based on road element information, where road 1 is one of the roads that must be traversed from the current location to the target location, the road trend line indicates the direction of road 1, and the road width is the width of the portion of road 1 that is accessible to vehicles; generating N rows of points based on the road trend line and road width, where each row of points includes P points, and the distance between any two points in the P rows is 1, and each row of points is distributed along the perpendicular line of the road trend line, where N and P are both positive integers; determining query point features based on the N rows of points, where the query point features indicate multiple query points, and each query point indicates the position on the centerline of one of the M lanes; and inputting the fused environmental features and query point features into the decoding network to obtain guidance information 1.

[0095] For example, the aforementioned distance 1 can be 0.75m, or it can be any other value. The distance between two adjacent query points along the navigation path can be 1.5m, or it can be any other value. For example, this distance can be determined based on the vehicle speed, and it increases as the vehicle speed increases.

[0096] In practice, query point features can be obtained by processing the query point set; or, query point features can be extracted by the navigation neural network based on the aforementioned rules, road element information and / or road trend lines.

[0097] In some implementations, the encoding network includes a first model and a second model. Navigation guidance information and road element information are input into the encoding network to obtain fused environmental features. This can include: inputting road element information into the first model to obtain the processing result of the first model; and inputting the processing result of the first model and the navigation guidance information into the second model to obtain fused environmental features. It should be noted that inputting road element information into the first model can be: inputting the encoded road element information into the first model; and inputting the processing result of the first model and the navigation guidance information into the second model can be: inputting the processing result of the first model and the encoded navigation guidance information into the second model.

[0098] In some implementations, the road element information further indicates at least one road-level element, which includes elements indicating the boundary of each of at least one road, and the at least one road includes one or more lanes of M lanes. Inputting navigation guidance information and road element information into the coding network to obtain fused environment features may include: inputting information of at least one lane-level element into a first model to obtain lane feature information, which indicates the relationship between any two lanes of the M lanes; inputting the lane feature information into a second model, and using the navigation guidance information and information of at least one road-level element as input to the second model to obtain fused environment features. Specifically, at least one road-level element may include at least one of the following: a road boundary, an intersection boundary, or a lane guide line. The intersection boundary may indicate the boundary of each intersection the vehicle passes through during its journey to the target location.

[0099] For example, the navigation neural network provided in the embodiments of this application can be as follows: Figure 4 As shown, specifically, the encoding network of the navigation neural network may include encoding layers 1 to 4, a self-attention model, a cross-attention model 1, and a cross-attention model 2. The self-attention model can be considered an example of the aforementioned first model, and cross-attention models 1 and 2 can be considered examples of the aforementioned second model. In addition to processing the query point set to obtain query point features, encoding layer 1 can also process road element information to obtain query point features.

[0100] In some implementations, the encoding network can also include multiple self-attention models and cross-attention models, for example, such as Figure 4 As shown, the encoding network can include n model groups, each model group including a self-attention model, a cross-attention model 1, and a cross-attention model 2. The input of each subsequent model group can be the output of the previous group. The input of cross-attention model 1 in each model group is the output of the previous model and navigation guidance information, and the input of cross-attention model 2 in each model group is the output of the previous model and road-level element information. For example, n can be any value from 4 to 6, or it can be any other value. Furthermore, each model group can also include more self-attention models; for example, it can include two self-attention models, where the input of the first model is lane-level element information, and the input of the second model is the output of the first model.

[0101] In actual implementation, Figure 4The coding layer shown can be an artificial neural network (ANN), such as a multilayer perceptron (MLP); or, Figure 4 The encoding layer shown can be a convolutional neural network (CNN). Figure 4 The decoding layer shown can be an MLP.

[0102] Figure 5 A schematic diagram of the training process of the navigation neural network provided in this application embodiment is shown. Specifically, it includes the following steps (a) to (d):

[0103] (a) Obtaining a Supervised Training Set: This supervised training set includes navigation scenario data and guidance information ground truth. The navigation scenario data comprises multiple sets of data, each set including navigation data, topology data, and perception data. The navigation data indicates the driving direction associated with each of the multiple roads traversed from a starting point to a destination. The topology data indicates the topological relationships between the multiple roads traversed from a starting point to a destination. The perception data indicates the road information perceived by the vehicle's perception system (such as lane-level elements and road-level elements contained in each road). The guidance information ground truth indicates the drivability (such as remaining drivable distance) at different locations on the multiple roads associated with each set of data. For example, the navigation scenario data can be data obtained from virtual simulation or data collected from a real vehicle; the guidance information ground truth can be manually labeled ground truth or ground truth automatically labeled based on rules.

[0104] (b) Generating random sampling points: Random sampling points are generated for each set of data to enable the navigation neural network to learn the distribution of accessibility over a wider range. The area where each sampling point is located is also labeled, and higher learning weights are assigned to scenarios where intersections or areas near intersections have a greater impact on lane-changing decisions, thereby improving the navigation neural network's ability to output accurate accessibility information for these areas.

[0105] (c) The navigation scene data and random sampling points are input into the navigation neural network to obtain guidance information for the network's inference. Specifically, the navigation scene data is sequentially input into the encoding network to obtain scene features, and the scene features and random sampling points (or possibly query points) are input into the decoding network to obtain guidance information. In some implementations, a query point set can be generated for each set of data using the aforementioned method for generating query point sets, and the query point set can be input into the decoding network so that the navigation neural network learns the spatial variation trend of the guidance information.

[0106] (d) Based on the distance between the guidance information obtained from the inference and the true value of the guidance information determined by the loss function, the parameters of the neural network model are adjusted to make the loss function converge. For example, the loss function can be L1 loss, L2 loss, cross-entropy, Focal loss, etc. Taking the L1 loss function as an example, based on the L1 distance between the inferred or predicted accessibility at a random sampling point and the true accessibility value corresponding to that sampling point, the optimizer is used to adjust the parameters of the neural network model to make the loss function converge.

[0107] In some implementations, navigation scene data can be augmented. For example, random noise can be added to the data (such as navigation data, topology data, and perception data at least one), and road information such as lane lines and road boundaries can be detopologically encoded into isolated elements. The augmented data can then be input into the navigation neural network to simulate the input of the navigation neural network in a real driving scenario.

[0108] For example, Figure 6 A schematic diagram showing the visualization of the guidance information 1 obtained through method 300 is shown. (Example) Figure 6 As shown, different colored lookout points indicate the distance you can continue driving along their respective lanes from that location; the darker the color, the shorter the distance you can continue driving. In other words, Figure 6 The bottom lane has the longest remaining driving distance at each position.

[0109] In the method for generating guidance information provided in this application embodiment, the navigation guidance information can provide information on road topology and driving direction beyond line of sight (i.e., beyond the vehicle's perception range), and the road element information can provide information on the road structure near the vehicle. This enables the navigation neural network to infer the far-field information (such as the remaining driving distance) corresponding to each lane of the road where the vehicle is currently located. Processing the aforementioned information through the navigation neural network helps to improve the efficiency of obtaining guidance information 1, thereby improving the real-time performance of navigation.

[0110] Figure 7 This illustration shows another schematic flowchart of a method for generating guidance information provided in an embodiment of this application. This method 400 can be applied to... Figure 1 In the vehicle shown, or the method can be derived from Figure 2 The system execution is shown. More specifically, method 400 can be executed by the planning and control module 240, and more specifically, by the Class B guidance information generation module 242. Method 400 may include:

[0111] S410, obtain navigation information and road topology information. The navigation information indicates the road-level driving direction from the vehicle's current location to the target location. The road topology information indicates the topological relationship between multiple roads within area 1 associated with the current location, as well as the number of lanes contained in each of the multiple roads.

[0112] In some implementations, road topology information can be information indicating historical road topology. For example, road topology information can include pre-made map data or information extracted from pre-made map data. The pre-made map data can be roadcode (RC) maps, electric horizon (EHP) data, etc.

[0113] In some implementations, pre-built map data can be generated based on traffic flow data. Traffic flow data can be understood as data consisting of the trajectories formed by one or more vehicles traveling on the road. A set of traffic flow data can include multiple traffic flow points, each indicating a coordinate in a vehicle's trajectory, the time the vehicle arrived at that coordinate, and the vehicle's orientation and pose at that coordinate.

[0114] Pre-built map data can include at least one of road vectors, intersection vectors, and lane vectors. For example, a processor segments and clusters traffic flow data to obtain road vectors. Further, for multiple roads intersecting at the same intersection, based on traffic flow data and road vectors, the vector points connecting each road to the intersection are determined, and the vector points corresponding to multiple roads constitute the intersection vector. The road width is determined based on traffic flow data, and the intersections of multiple sets of traffic flow data with the perpendicular lines from the roads are clustered. The number of lanes is determined based on the clustering results, and then the lane vectors are determined based on the road width and the number of lanes. It can be understood that lane vectors, road vectors, and intersection vectors constitute a vectorized map. Specifically, road vectors indicate the location and direction of a road segment, intersection vectors indicate the location and boundaries of intersections, and lane vectors indicate the roadway for various vehicles to travel within the same width. Alternatively, lane vectors can also indicate the position of each lane within a road segment.

[0115] For example, the navigation information can be SD map navigation information determined according to the standard definition (SD) map.

[0116] In some implementations, region 1 can be an area that includes the current position, or it can be an area located at a certain distance from the current position (such as a distance between 10 and 30 meters). For example, region 1 can be a rectangular area with a length of n meters and a width of m meters, or it can be a circular area with a diameter of m meters, or it can be a region of other shapes. For instance, n can be a value between 500 and 1000, m can be a value between 10 and 20, or n and m can be other values.

[0117] In some implementations, road topology information includes multiple shape points indicating the positions of lanes within the road, with each shape point indicating a location on the lane centerline. When the number of lanes changes, the road topology information can also indicate the location of the lane change.

[0118] For example, Figure 8 This paper illustrates an example of an application scenario involving this application. If a vehicle needs to travel from its current location to its destination by passing through intersection a to intersection e, the navigation information at least indicates the vehicle's direction of travel at intersection a, and the road topology information at least indicates the road connected to intersection a and its direction.

[0119] It should be noted that the intersections involved in this application may include, but are not limited to, cross intersections, off-ramp connection intersections, on-ramp connection intersections, intersections leading to auxiliary roads, and intersections merging into main roads. Among them, a cross intersection can be an n-way intersection, where n is an integer greater than or equal to 3. That is, each intersection includes at least three boundaries, and each of the at least three boundaries connects to a road.

[0120] S420 determines topology guidance information based on navigation information and road topology information, which indicates the navigation trend of lanes contained in at least one road at intersections.

[0121] For example, at least one road is the route that a vehicle needs to take to travel to a target location, and any two adjacent roads in the at least one road are connected by an intersection.

[0122] In some implementations, based on navigation information and road topology information, the driving direction of the vehicle at each of the N2 target intersections and the number of lanes contained in the preceding road of that intersection are determined. These N2 target intersections can be some or all of the intersections the vehicle needs to pass through on its journey to the target location, where N2 is a positive integer. The preceding road of a given intersection refers to the nearest neighbor road the vehicle must pass through on its journey to that intersection. For example, if a vehicle travels from road 1 through intersection 1 to road 2, and road 1 and intersection 1 are connected, then road 1 is the preceding road of intersection 1.

[0123] In some implementations, navigation information and road topology information can also be used to determine the vehicle's deviation on a particular road. For example, for Figure 4 The navigation scenario shown indicates that the vehicle will travel on the right side of the road following intersection e to reach its destination. Here, a subsequent road at an intersection refers to the nearest road the vehicle enters after passing that intersection. For example, if a vehicle travels from road 1 through intersection 1 to road 2, and road 2 connects to intersection 1, then road 2 is the subsequent road of intersection 1.

[0124] Furthermore, based on the vehicle's direction of travel at each target intersection or the vehicle's deviation on the road, as well as the number of lanes contained in the preceding road of the target intersection, the initial navigation trend at each target intersection is determined.

[0125] For example, for each target intersection, determining its initial navigation trend may specifically include the following steps (a) and (b):

[0126] (a) Determine the accessibility of different sections of the preceding road at the target intersection.

[0127] In one example, when lane guidance (i.e., the drivable direction of vehicles supported by the lane at the intersection) cannot be determined, the portion of the preceding road connecting to the target intersection can be divided into M3 equal parts along the width direction of the preceding road, for example, into 60 equal parts. This M3 division can be further divided into six parts, each consisting of 10 equal parts; or, this M3 division can be further divided into ten parts, each consisting of 6 equal parts. Further, the passability of each part is determined based on the vehicle's drivable direction at the intersection. For example, if the navigation information determines that the vehicle's drivable direction at the intersection is turning towards a first side (such as left or right), the passability of the preceding road on the half closest to the first side at the intersection can be determined to be 100%, and the passability of the remaining half to be 0%. The portion with a passability of 100% can be considered a passable portion. For example, when navigation information indicates that the vehicle is going straight at an intersection, it can be determined that the passability of the preceding road in the middle 4 / 5 of the intersection is 100%, and the passability of the remaining 1 / 10 on each of the left and right sides is 0%.

[0128] In another example, if lane guidance can be determined, each lane can be divided into multiple parts of the preceding road to the target intersection based on the guidance of each lane. For example, the preceding road includes lanes 1 to 3, where lane 1 supports going straight and turning left at the nearest intersection, lane 2 only supports going straight at the nearest intersection, and lane 3 supports going straight, turning right, and making a U-turn at the nearest intersection. For example, lane 1 can be divided into two equal parts, each occupying 1 / 6 of the width of the preceding road at the target intersection, for vehicles to turn left and vehicles to go straight, respectively; since lane 2 only supports straight travel, lane 2 is not further divided, that is, lane 2 occupies 1 / 3 of the width of the preceding road at the target intersection; lane 3 can be divided into three equal parts, each occupying 1 / 9 of the width of the preceding road at the target intersection, for vehicles to go straight, turn right and make a U-turn, respectively; since lane 2 only supports straight travel, lane 2 is not further divided, that is, lane 2 occupies 1 / 3 of the width of the preceding road at the target intersection. Furthermore, when a vehicle needs to proceed straight through the target intersection, the passable portion of the preceding road includes 50% of lane 1, 100% of lane 2, and 33.3% of lane 3; when a vehicle needs to turn left at the target intersection, the passable portion of the preceding road includes 50% of lane 1; and when a vehicle needs to turn right or make a U-turn at the target intersection, the passable portion of the preceding road includes 33.3% or 66.7% of lane 3.

[0129] (b) Smooth the accessibility and obtain the navigation cost based on the smoothed result, which can characterize the navigation trend.

[0130] Taking an intersection where the preceding road has three lanes, its width is divided into 60 equal parts, and vehicles turn right at the intersection where the preceding road connects, as an example, the passability of the preceding road can be converted into the following form: [0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0 The expression is: .0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0, where each of the above terms represents the probability of passage of an equal division of the preceding road. A probability of 0.0 indicates that the road is impassable, and a probability of 1.0 indicates that the road is passable. From left to right, these terms represent the equal divisions of the preceding road from the left to the right.

[0131] In some implementations, the above passage probability can be smoothed, for example, by using the following formula (1):

[0132]

[0133] Where, σ 2 The variance σ is determined by the distance between each intersection and the vehicle's current location, based on the SD map navigation information. The greater the distance, the smaller the variance. For example, the variance σ in formula (1) can be determined using the following formula (2). 2 The specific value.

[0134]

[0135] Where dis represents the distance between the vehicle and the intersection of the preceding road indicated by the navigation information. σ0 is the baseline standard deviation, which can be 1.76 or other calibrated values; dis1 and dis2 are distance thresholds, which can be 0.5 km and 1 km respectively, or other calibrated values. For example, when driving on urban roads, dis1 and dis2 take the above values, and when driving on highways, dis1 and dis2 take 1 km and 2 km respectively; k, μ1, and μ2 are coefficients, which can be -2, 0.88, and -0.5 respectively, or other calibrated values.

[0136] For example, if the distance between the target intersection and the vehicle is 101 meters (i.e., 0.101 kilometers), the variance σ in formula (1) can be determined according to formula (2). 2 It is 2.5. Furthermore, inputting the above passage probability 1 into formula (1) yields the smoothed passage probability: [0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.02 0.05 0.11 0.19 0.30 0.43 1.14 1.14 1.14 1.14 1.14 1.14 ...

[0137] Furthermore, the smoothed passage probability is scaled to three lanes, resulting in passage probabilities of 0.0, 0.07, and 1.0 for the three lanes from left to right. Then, the passage probabilities of each lane are mapped using a pre-defined function (e.g., formula (3)) to obtain navigation costs of 4000, 2000, and 0 for the three lanes from left to right. The higher the navigation cost of a lane, the lower its recommended driving level.

[0138]

[0139] Where p is the actual passage probability, p0 and p1 are the passage probability thresholds, which can be 0.05 and 0.9 respectively, or other calibrated values. cost0 and cost1 represent the navigation cost, which can be 4000 and 0 respectively, or other calibrated values. a and b are coefficients, which can be 100 and 2000 respectively, or other calibrated values.

[0140] After determining the initial navigation trend at each of the N2 target intersections, the initial navigation trends of the N2 target intersections can be superimposed to obtain the final navigation trend for each target intersection. For example, starting from the initial navigation trend corresponding to the target intersection farthest from the current location of the vehicle, the trends of the target intersections closer to the current location are superimposed sequentially, and the navigation trends of at least one target intersection close to the current location are corrected in turn to obtain the final navigation trend.

[0141] In some implementations, topology guidance information can also indicate navigation trends for different positions of at least one lane in a road. In one example, after determining the navigation trends of N2 target intersections, the navigation trends for different positions of at least one lane in the road between any two adjacent intersections can be determined based on the navigation trends of those two adjacent intersections. In another example, during the overlay process of the initial navigation trends of N2 target intersections, the navigation trends for different positions of at least one lane in the road between any two adjacent intersections can be determined based on the navigation trends of those two adjacent intersections.

[0142] In actual implementation, some of the N2 target intersections may lack navigation information. In this case, the navigation information of the vehicle at the intersection can be deduced based on the vehicle's target location and the subsequent roads of the intersection with missing navigation information.

[0143] S430, based on the topology guidance information, generate guidance information 2, which includes multiple grids, each grid point indicating a position in the lane, and the value of the grid point indicating the navigation trend.

[0144] In some implementations, road topology information within a certain range can be obtained based on the vehicle's current location, and key points can then be generated based on the shape points included in the road topology information. This certain range can be within 1 to 1.5 kilometers of the vehicle, or it can be other ranges determined by the vehicle's speed; for example, this range expands as the vehicle's speed increases.

[0145] For example, road topology information may include a sparse set of key points. For each key point, it is extended laterally to the left and right by a certain distance (e.g., a value between 0.5 meters and 0.75 meters) to obtain two key points located to the left or right of that key point. Then, it is interpolated longitudinally to obtain a dense set of key points. Here, the longitudinal direction is parallel to the road boundary (or lane centerline), and the lateral direction is perpendicular to the road boundary (or lane centerline).

[0146] Furthermore, a raster matrix is ​​generated based on the location of the keypoints, and this raster matrix can cover the locations where the keypoints exist. For example, Figure 9 The dashed box on the left shows the key points, lane center positions, and the positional relationships between grid points in the grid matrix. Further, the navigation trend corresponding to each key point is determined, and values ​​are assigned to the grid points of the grid containing that key point based on the navigation trend. For example, when the topology guidance information indicates the navigation trend at an intersection, the corresponding navigation trend can be determined based on the distance between the key point and the intersection; or, when the topology guidance information also indicates the navigation trends at different positions within the lane, the navigation trend of that key point can be determined based on the navigation trends of its neighboring locations. After assigning values ​​to the grid points, the values ​​of grid points at different positions in the grid matrix are different, and the visualization of the grid matrix can be as follows: Figure 9 As shown in the dashed box on the left, the darker the color of the grid point, the higher the navigation cost at the corresponding location, and / or the shorter the remaining drivable distance at the corresponding location.

[0147] After assigning values ​​to the grid points of the raster containing key points, values ​​are then assigned to the grid points of the raster that does not contain key points, based on the already assigned grid points, to obtain guidance information 2. For example, the visualization of a portion of the area corresponding to guidance information 2 is as follows: Figure 9 As shown in the dashed box on the right. The darker the color of a grid point, the higher the navigation cost for that location, and correspondingly, it is less recommended to travel to that location.

[0148] In some implementations, grid points of a raster that does not contain keypoints can be assigned values ​​based on the following formula (4):

[0149]

[0150] Where f(x,y) represents the assigned values ​​of the grid points, u represents the values ​​of the grid points at different positions (i.e., the navigation trend), and x and y represent the horizontal and vertical directions, respectively. Let Ω be the boundary of region Ω. Indicates boundary conditions.

[0151] S440, acquire environmental perception information, which indicates the lane boundaries and / or lane centerline of the road where the vehicle is currently located.

[0152] For example, the environmental perception information can be an image including pixels of lane boundaries and / or lane centerlines, or it can be point cloud data including point clouds of lane boundaries and / or lane centerlines. Further, image processing can be performed on the environmental perception information to determine the lane boundary position of the vehicle's current location on the road.

[0153] S450 determines M2 query points based on environmental perception information, and each of the M2 query points corresponds to one of the M2 lanes of the target road.

[0154] For example, the target road may include the road where the vehicle is currently located, or the target road may also include the subsequent road of the nearest intersection that the vehicle needs to pass through during its journey to the target location.

[0155] For example, each of the M2 query points can indicate a position on the center line of the lane, and the distance between each of the M2 query points and the current position of the vehicle can be greater than or equal to distance 2 and less than or equal to distance 3. For example, distance 2 can be a value between 30 meters and 50 meters, and distance 3 can be a value between 50 meters and 100 meters, or distance 2 and distance 3 can be other values.

[0156] S460, based on M2 query points and guidance information 2, determine the navigation trend corresponding to each lane in M2 lanes.

[0157] For example, for each query point, its value is determined based on the values ​​of the four grid points of its surrounding grid. This value represents the navigation trend for the query point's location. For instance, the positional relationship between the query point and the grid can be as follows: Figure 10 As shown, where, Figure 10 The grid points within each dashed box shown are used to determine the value of the query point within the dashed box. For example, the value of the query point can be determined using bilinear interpolation based on the values ​​of the four grid points of the grid containing the query point.

[0158] It should be noted that in actual implementation, the guidance information 2 can also be generated in a different order than method 400. Specifically, the query point can be determined based on the perception information acquired by the vehicle; further, based on the query point and road topology information, multiple key points within a certain area corresponding to the query point can be determined, as shown in the following example. Figure 11 As shown in the diagram, each query point indicates a location on the centerline of a lane, and multiple key points corresponding to that query point indicate the location range of a segment of the lane centerline. Based on these key points, guidance information 2 is generated. For a more detailed implementation of generating query points, key points, and guidance information 2, please refer to the description in the preceding embodiments, which will not be repeated here.

[0159] In the method for generating guidance information provided in this application embodiment, road topology information can provide the topological relationship between roads beyond line of sight, and navigation information can provide information on the driving direction. In this way, even without high-precision map guidance, the vehicle can obtain information indicating far-field road conditions, enabling the vehicle to determine whether to change lanes and / or the timing of lane changes in complex road conditions, reducing the probability of invalid lane changes.

[0160] Figure 12 A further illustrative flowchart of the method for generating guidance information provided in an embodiment of this application is shown. This method 500 can be applied to... Figure 1 In the vehicle shown, or the method can be derived from Figure 2 The system execution is shown. More specifically, method 500 can be executed by the control module 240, and more specifically, by the verification module 243. Method 500 may include:

[0161] S510, acquire environmental perception information and first navigation information, wherein the environmental perception information indicates the road boundary of the target road perceived by the vehicle, and / or the lane boundary associated with the target road; the first navigation information indicates at least one passable lane in the target road for the vehicle to travel to the target location, and the position of the at least one passable lane in the target road.

[0162] The target road is either the current road the vehicle is on, or the road the vehicle needs to travel to reach the target location.

[0163] For example, the environmental perception information and the perception information in method 300 can be the same information; the first navigation information and the navigation information used to generate navigation guidance information in method 300 can be the same information.

[0164] S520, based on environmental perception information, determine N1 sampling results corresponding to the target road. Each sampling result indicates at least one passable lane corresponding to one of the N1 sampling planes. The distance between any two adjacent planes in the N1 sampling planes is the first distance.

[0165] For example, the projected lane of the road where the vehicle is currently located is determined based on environmental perception information. Here, the projected lane refers to an ordered sequence of lateral lanes selected within a preset distance (e.g., 200 meters, 250 meters, or other values) in the vehicle's direction of travel, based on perceived environmental information (such as an image including lane topology captured by a camera device). The ordered sequence of lateral lanes can be a sequence of lanes arranged sequentially from the left to the right of the vehicle, or it can be a sequence of lanes arranged sequentially from the right to the left of the vehicle.

[0166] Furthermore, the lanes on the projection plane are sampled at equal intervals along the longitudinal direction to obtain N1 sampling planes. Each sampling plane is a plane perpendicular to the vehicle's direction of travel. The first distance between any two adjacent sampling planes can be 15 meters to 30 meters, or it can be any other value. Then, the number of lanes contained in each sampling plane is determined. In some implementations, the position of each lane relative to the vehicle or the position of each lane in the road can also be determined.

[0167] In some implementations, each of the defined multiple lanes can be traversed, and a certain distance (e.g., 100 meters, 150 meters, or other distances) can be searched towards the target location. If a lane splits, the split lane replaces the original lane; if lanes merge, the merged lane replaces the original lane. This search is performed until a search threshold is reached, thus determining the final projected plane lanes.

[0168] In some other implementations, invalid lanes in the initially identified projection plane lanes can be deleted to obtain the final projection plane lanes. Invalid lanes can be, for example, emergency lanes, escape lanes, non-motorized vehicle lanes, etc., which are generally not allowed to be used by vehicles.

[0169] For example, the position of a vehicle in a lane can be determined by either of the following two methods, thereby determining the position of each lane in the road:

[0170] First, road boundaries can be determined based on environmental perception information, and then the number N of lanes to the left of the vehicle can be determined based on the road boundaries. left And the number of lanes N on the right. right And then according to N left and N right Determine the vehicle's lane position, where N left and N right All values ​​are integers greater than or equal to 0. It is understood that a road may include two road boundaries, and the area within these two road boundaries is the drivable area for vehicles. For example, these road boundaries may include, but are not limited to, shoulders, double yellow lines, and central medians.

[0171] Secondly, it can also be based on the number of lanes N to the left of the vehicle. left The number of lanes on the right, N right And the total number of lanes N indicated by the SD map navigation information. sd Determine the vehicle's current lane position. For example, determine the total number N of lanes on the current travel route based on environmental information. total =N left +N right +1. Furthermore, since the number of lanes N recorded in the SD map... sd There might be an error, or the vehicle's sensors might be misreading the number of lanes, leading to N. total With N sd If they are not equal, then it can be determined according to N. total With N sd Based on the comparison results, determine the lane the vehicle is in: ① In N total ≤N sd If the distance between the left lane line of the leftmost lane and the left road boundary is less than or equal to a preset distance threshold, then the vehicle is in the Nth lane from the left. left +1 lane; if the distance between the right lane line of the rightmost lane and the right road boundary is less than or equal to a preset distance threshold, then the vehicle is in the Nth lane from the left. sd -N right Lane 1. ② At N total >N sd If a lane split occurs ahead and a vehicle needs to pass through the right lane after the split, then the vehicle's lane will be the Nth lane from the left. total -N right There are 10 lanes; if a lane split occurs ahead and vehicles need to pass through the left lane after the split, then the vehicle's lane is the Nth lane from the left. left +1 lane; if a lane merge occurs ahead and a vehicle enters the merged lane from the right lane, the vehicle will then be in the Nth lane from the left. sd -N right There are 10 lanes; if a lane merge occurs ahead and a vehicle enters the merged lane from the left lane, the vehicle will be in the Nth lane from left to right. left +1 lane; otherwise, the vehicle is considered to be in the Nth lane from the left. left +1 lane.

[0172] S530, based on the first navigation information, determine the M1 intersection guidance information associated with N1 sampling planes.

[0173] Where M1 and N1 are both positive integers.

[0174] For example, the M1 intersection guidance information can be included in multiple intersection guidance information from the vehicle's current location to the target location. The M1 intersection guidance information can be the M1 intersection guidance information closest to the vehicle among the multiple intersection guidance information. The distance between each of the M1 intersection guidance information and the vehicle can be less than the aforementioned preset distance, or it can be greater than the aforementioned preset distance.

[0175] For example, Figure 13 The diagram illustrates the relative positional relationship between the projected lane plane determined based on environmental perception information and the vehicle's current location. Specifically, sampling planes 1 to 7 can be considered examples of N1 sampling planes, and navigation information sampling 1 and navigation information sampling 2 can be considered examples of M1 intersection guidance information. White arrows indicate lanes that are impassable, while gray arrows indicate lanes that are passable.

[0176] S540 determines M1*N1 combined results based on N1 sampling results and M1 intersection guidance information. Each combined result indicates the passable lanes among multiple lanes and their positions in the target road.

[0177] For example, one of the N1 sampling results is matched with one of the M1 intersection guidance information to obtain a combined result.

[0178] In one example, if the number of lanes indicated by the sampling results differs from the number of lanes indicated by the intersection guidance information, a sliding window matching is performed on both, and the results from multiple sliding window matchings are ORed to obtain a combined result. For example, such as... Figure 14 As shown, when matching sampling plane 1 and navigation information sampling 1, since sampling plane 1 includes three lanes and navigation information sampling 1 indicates the passability of four lanes, we can first match the three lanes with the three lanes on the left indicated by navigation information sampling 1 to determine matching result 1. This matching result 1 indicates that lane 1 is not passable, while lanes 2 and 3 are passable. Then, we match the three lanes with the three lanes on the right indicated by navigation information sampling 1 to determine matching result 2. This matching result 2 indicates that lanes 1 to 3 are all passable. Then, we take the OR of matching result 1 and matching result 2 to obtain combined result a, which indicates that lanes 1 to 3 are all passable. It can be understood that the combined results corresponding to sampling planes 2, 4, 6, and 7 are consistent with combined result a.

[0179] In another example, if the number of lanes indicated by the sampling result is related to the number of lanes indicated by the intersection guidance information, then the lanes included in both are matched one-to-one to obtain a combined result. For example, if sampling plane 3 and navigation information sampling 1 are matched, since sampling plane 3 includes four lanes and navigation information sampling 1 indicates the passability of the four lanes, the four lanes can be matched one-to-one with the four lanes indicated by navigation information sampling 1 to determine combined result b. Combined result b indicates that lanes 1 to 3 are all passable, and the lane to the left of lane 1 is not passable. Similarly, if the combined result c obtained by matching sampling plane 5 and navigation information sampling 1 indicates that lanes 2, lane 3, and the lane to the right of lane 3 are all passable, and lane 1 is not passable.

[0180] Figure 13 The seven sampling planes and navigation information sampling 2 are matched to obtain seven combined results, as shown below. Figure 15 As shown, specifically, the 7 combined results include 5 combined results d, 1 combined result e, and 1 combined result f. Among them, combined result d indicates that lane 1 and lane 2 are feasible, while lane 3 is not feasible; combined result e indicates that lane 1, lane 2, and the lane to the left of lane 1 are all feasible, while lane 3 is not feasible; combined result f indicates that lanes 1 to 3 are all feasible.

[0181] S550, based on the M1*N1 combination results, determine the guidance information 3.

[0182] For example, the M1*N1 combined results can be superimposed to obtain guidance information 3, which indicates the passability (or navigation recommendation) of each lane.

[0183] In one example, the M1*N1 combined results can be superimposed to obtain the navigation recommendation level for lanes 1 to 3. Figure 16 A schematic diagram of the navigation recommendation level obtained based on this example is shown, in which the longer the arrow corresponding to the lane, the higher the passability or navigation recommendation level of the lane, that is, the longer the remaining driving distance of the vehicle in the lane.

[0184] In another example, the M1*N1 combined results can be either ORed or summed to obtain the navigation recommendation level for lanes 1 to 3. It is understandable that after ORing and summing, lanes 1 to 3 will have the same navigation recommendation level, and all will be considered passable lanes.

[0185] It is understood that the OR operation of two or more results involved in the embodiments of this application is as follows: if one of the two or more results indicates that the lane is passable, then the combination of the two or more results indicates that the lane is passable.

[0186] In some implementations, the vehicle's current navigation information (hereinafter referred to as the original navigation information) is verified based on guidance information 3 to determine whether to switch to other navigation information for navigation. The original navigation information can be the aforementioned guidance information 1, guidance information 2, or other guidance information. More specifically, the relationship between the navigation direction indicated by guidance information 3 and the navigation direction indicated by the original navigation information can be used to determine whether to switch to other navigation information for navigation. The navigation direction refers to the passability trend of multiple lanes on the road indicated by the guidance information. For example, if the guidance information indicates that the right lane has the highest passability and the passability of lanes decreases sequentially to the left, then the navigation direction can be determined to recommend that the vehicle drive in the right lane.

[0187] In one example, when the navigation direction indicated by the original guidance information matches the navigation direction indicated by guidance information 3, the original guidance information and guidance information 3 are fused to obtain a fused result, and the vehicle is controlled to drive based on the fused result. Exemplarily, navigation direction matching includes: lanes with a passability level greater than or equal to a threshold indicated by the original guidance information, and lanes with a passability level greater than or equal to the threshold indicated by guidance information 3. For example, Figure 17 The scenarios shown in (1) and (2) both represent situations where the navigation direction indicated by the original guidance information matches the navigation direction indicated by guidance information 3. Each arrow indicates the passability of that lane; the longer the arrow, the higher the passability of the lane. In one example, fusing guidance information 3 and the original guidance information can include fusing the results for lanes where the passability indicated by guidance information 3 and the original guidance information differs. For example, for... Figure 17 Lane a shown in (1) and (2) is modified using guidance information 3 to make the passability of lane a consistent with the passability indicated by guidance information 3. In another example, fusing guidance information 3 and original guidance information may include: recalculating the passability of all lanes in the road for guidance information 3 and original guidance information respectively, and obtaining the fused result.

[0188] In another example, when the navigation direction indicated by the original guidance information is opposite to the navigation direction indicated by guidance information 3, the system switches to using other guidance information to control vehicle movement. For example, Figure 17 The scenario shown in (3) is the case where the navigation direction indicated by the original guidance information is opposite to the navigation direction indicated by guidance information 3. For example, if the original guidance information is guidance information 1, then the vehicle driving is controlled by guidance information 2, guidance information 3, or other guidance information.

[0189] In another example, when the navigation direction matches a lane whose passability level, indicated by guidance information 3, is greater than or equal to a threshold, and the original guidance information indicates a passability level greater than or equal to a threshold, the vehicle continues to be controlled based on guidance information 3. For example, Figure 17 The scenario shown in (4) is the scenario corresponding to this example.

[0190] It should be noted that, Figure 17 The example shown illustrates different traffic flow patterns for roads with two lanes. In practice, roads can include more lanes, and the corresponding traffic flow patterns for roads with more lanes can be referenced. Figure 17 The scene shown.

[0191] Figure 18 A schematic flowchart of a navigation method provided in an embodiment of this application is shown. This method 1000 can be applied to... Figure 1 In the vehicle shown, or the method can be derived from Figure 2 The system shown executes this method. More specifically, this method 1000 can be executed by the control module 240. This method 1000 may include:

[0192] S1010, Obtain first type of guidance information. The first type of guidance information indicates the passability of each lane in at least one road. At least one road is the road that the vehicle needs to travel from the current location to the target location. The first type of guidance information is generated based on historical road topology.

[0193] For example, the first type of guidance information may include guidance information 2 in method 400; or, the first type of guidance information may also include other guidance information generated based on historical road topology.

[0194] S1020, Obtain second type of guidance information, which indicates the passability of lanes in the target road perceived by the vehicle.

[0195] For example, the second type of guidance information may be generated based on a neural network. For instance, the second type of guidance information may include guidance information 1 in method 300; or, the second type of guidance information may be generated based on predefined rules. For instance, the second type of guidance information may include guidance information 3 in method 500; or, the second type of guidance information may also include other guidance information generated based on environmental information perceived by the vehicle.

[0196] S1030, based on the first type of guidance information and / or the second type of guidance information, control the vehicle to travel to the target location via the first lane in the target road.

[0197] In some implementations, S1030 can be further refined as follows: during the process of controlling the vehicle to drive according to the first type of guidance information, when the offset between the boundary of the target road indicated by the first type of guidance information and the boundary of the target road perceived by the vehicle is greater than or equal to a distance threshold, the system switches to controlling the vehicle to drive towards the target location according to the second type of guidance information; or, when the number of drivable lanes of the target road indicated by the first type of guidance information is inconsistent with the number of drivable lanes of the target road perceived by the vehicle, the system switches to controlling the vehicle to drive towards the target location according to the second type of guidance information.

[0198] For example, the distance threshold can be a value between 0.75 meters and 1 meter; or, the distance threshold can be other values.

[0199] For example, switching to control the vehicle to drive towards the target location based on the second type of guidance information may include: switching to control the vehicle to drive towards the target location based on guidance information 1.

[0200] In some implementations, the method further includes: when controlling the vehicle to drive according to the second type of guidance information, if the offset between the boundary of the target road indicated by the first type of guidance information and the boundary of the target road perceived by the vehicle is less than a distance threshold, or if the number of drivable lanes of the target road indicated by the first type of guidance information is consistent with the number of drivable lanes of the target road perceived by the vehicle, the method switches to controlling the vehicle to drive towards the target location according to the first type of guidance information.

[0201] In some implementations, while controlling the vehicle's movement based on the second type of guidance information, it is also possible to switch between multiple types of second type of guidance information, that is, to switch from one type of second type of guidance information to another type of second type of guidance information to control the vehicle to move towards the target location.

[0202] In some implementations, the second type of guidance information is generated by the navigation neural network through processing navigation guidance information and road element information. The navigation guidance information indicates the R roads that the vehicle needs to take from the current position to the target position and the direction of travel. The R roads include the target road. The road element information indicates at least one lane-level element perceived by the vehicle at the current position. The at least one lane-level element indicates the lane boundary and / or lane centerline of each of the S lanes. The S lanes are the lanes in the R roads, and S and R are both positive integers.

[0203] For example, the navigation neural network can be the navigation neural network in the foregoing embodiments. A more specific implementation of generating the second type of guidance information based on the navigation neural network can be found in the description in method 300, which will not be repeated here.

[0204] In some implementations, S1030 can be further refined as follows: when the navigation directions corresponding to multiple lanes in the target road indicated by the first type of guidance information are opposite to the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, the vehicle is controlled to travel to the target location via the first lane according to the second type of guidance information or according to the third type of guidance information; wherein, the third type of guidance information indicates the passability of the lanes included in the target road, and the third type of guidance information is generated in a different way than the first type of guidance information and the second type of guidance information.

[0205] For example, the second type of guidance information can be guidance information 3 in method 500, the third type of guidance information can be guidance information 1 in method 300, or the third type of guidance information can also be other guidance information generated based on the environmental information perceived by the vehicle.

[0206] In some implementations, the method further includes: when controlling the vehicle's movement according to the third type of guidance information, if the navigation directions corresponding to the multiple lanes indicated by the third type of guidance information are opposite to the navigation directions corresponding to the multiple lanes indicated by the second type of guidance information, the method further includes controlling the vehicle to move towards the target location via the first lane according to the second type of guidance information.

[0207] For example, during the process of controlling the vehicle's movement based on the third type of guidance information, when the navigation directions corresponding to the multiple lanes indicated by the third type of guidance information match the navigation directions corresponding to the multiple lanes indicated by the second type of guidance information, the third type of guidance information and the second type of guidance information are fused to obtain a fusion result; based on the fusion result, the vehicle is controlled to move towards the target location via the first lane.

[0208] For example, the specific implementation of fusing the second type of guidance information and the third type of guidance information can be referred to the description in method 500, and will not be repeated here.

[0209] In some implementations, S1030 can be further refined as follows: when the navigation directions corresponding to multiple lanes in the target road indicated by the first type of guidance information match the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, the first type of guidance information and the second type of guidance information are fused to obtain a fusion result; based on the fusion result, the vehicle is controlled to travel to the target location via the first lane.

[0210] For example, the specific implementation of fusing the first type of guidance information and the second type of guidance information can be referred to the description in method 500, and will not be repeated here.

[0211] In some implementations, the navigation directions corresponding to multiple lanes indicated by the first type of guidance information are matched with the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, including: lanes whose passability level indicated by the first type of guidance information is greater than or equal to a threshold level include lanes whose passability level indicated by the second type of guidance information is greater than or equal to a threshold level.

[0212] In practical implementation, some road sections may lack the first type of guidance information, or the guidance information on some road sections may not match the indication of the first type of guidance information due to temporary road repairs, road closures, or other reasons. The navigation method provided in this application embodiment can verify the validity of the first type of guidance information through the second type of guidance information. When there is a large difference between the first type of guidance information and the actual passability information of the road, navigation can be switched to the second type of guidance information or other guidance information. This helps to improve the robustness and reliability of the vehicle's intelligent driving system, thereby reducing the probability of the vehicle deviating from the target location and improving driving efficiency.

[0213] In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions between the various embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0214] The above text combines Figures 1 to 18 The methods provided in the embodiments of this application are described in detail below. Figure 19 and Figure 20 The apparatus provided in the embodiments of this application is described in detail. It should be understood that the description of the apparatus embodiments corresponds to the description of the method embodiments. Therefore, for content not described in detail, please refer to the method embodiments above. For the sake of brevity, it will not be repeated here.

[0215] Figure 19 A schematic block diagram of an apparatus 2000 provided in an embodiment of this application is shown. The apparatus 2000 may include units for executing the methods described in the foregoing embodiments. Furthermore, each unit in the apparatus 2000 implements a corresponding process of the above method embodiments. The apparatus 2000 includes an acquisition unit 2010, which can be used to implement corresponding data acquisition or transmission / reception functions. The apparatus 2000 also includes a processing unit 2020, which can be used to implement corresponding processing functions.

[0216] Optionally, the device 2000 further includes a storage unit, which can be used to store instructions and / or data. The processing unit 2020 can read the instructions and / or data in the storage unit so that the device can perform the relevant actions in the aforementioned method embodiments.

[0217] It should be understood that the specific process of each unit performing the above-mentioned corresponding steps has been described in detail in the above method embodiments, and will not be repeated here for the sake of brevity.

[0218] It should also be understood that the device 2000 described herein is embodied in the form of a functional unit. The terms “module” or “unit” may refer to application-specific ASICs, electronic circuits, processors (e.g., shared processors, proprietary processors, or group processors) and memory for executing one or more software or firmware programs, integrated logic circuits, and / or other suitable components that support the described functions.

[0219] The apparatuses described above have the function of implementing the corresponding steps in the methods described above. These functions can be implemented in hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the functions described above; for example, the acquisition unit 2010 can be replaced by a transceiver, and other units, such as the processing unit, can be replaced by a processor, used to execute the relevant processing operations in each method embodiment.

[0220] For example, the acquisition unit 2010 and the processing unit 2020 can be set in Figure 1 In the vehicle 100 shown, or it can also be set in Figure 2 In the illustrated system, more specifically, the acquisition unit 2010 and processing unit 2020 can be housed within the control module 240. Exemplarily, the operations performed by the acquisition unit 2010 and processing unit 2020 can be executed by a single processor, or by different processors. In specific implementations, one or more processors can be configured to be located within the control module 240. Figure 1 The processor in the vehicle 100 shown; or, the device 2000 described above may be a chip disposed in the vehicle 100.

[0221] In the specific implementation process, the units in the above device can be fully or partially integrated together, or they can be implemented independently. In one implementation, these units are integrated together and implemented in the form of a system-on-a-chip (SoC).

[0222] Figure 20 This is another schematic block diagram of the device provided in the embodiments of this application. Figure 20The illustrated device 2100 may include a processor 2110, a transceiver 2120, and a memory 2130. The processor 2110, transceiver 2120, and memory 2130 are connected via internal interconnects. The memory 2130 stores instructions, and the processor 2110 executes the instructions stored in the memory 2130 to implement the methods described in the above embodiments. Optionally, the memory 2130 may be coupled to the processor 2110 via an interface or integrated with the processor 2110.

[0223] It should be noted that the transceiver 2120 mentioned above may include, but is not limited to, transceiver devices such as input / output interfaces, to realize communication between device 2100 and other devices or communication networks.

[0224] Memory 2130 can be volatile memory and / or non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM). For example, RAM can be used as an external cache. By way of example and not limitation, RAM includes various forms such as: static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0225] Transceiver 2120 uses transceiver devices, such as but not limited to transceivers, to enable communication between device 2100 and other devices or communication networks to receive / send data / information for implementing the methods in the above embodiments.

[0226] This application also provides an intelligent driving device, which includes the device 2000 or device 2100 in the above embodiments.

[0227] This application also provides a computer program product, which includes computer program code. When the computer program code is run on a computer, it causes the computer to implement the methods described in the above embodiments of this application.

[0228] This application also provides a computer-readable storage medium storing computer instructions that, when executed on a computer, cause the computer to implement the methods described in the above embodiments of this application.

[0229] This application also provides a chip, including circuitry, for performing the methods described in the above embodiments of this application.

[0230] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0231] In the description of the embodiments of this application, unless otherwise stated, " / " means "or", for example, A / B can mean A or B; "and / or" in this document describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. In this application, "at least one" means one or more, and "more" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0232] The use of prefixes such as "first" and "second" in this application embodiment is solely for distinguishing different descriptive objects and does not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes to distinguish descriptive objects in this application embodiment does not constitute a limitation on the described objects. The description of the described objects is found in the claims or the context of the embodiments, and the use of such prefixes should not constitute unnecessary restrictions.

[0233] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0234] In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions between the various embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0235] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0236] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0237] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A navigation method, characterized in that, include: Obtain a first type of guidance information, which indicates the passability of each lane in at least one road, wherein the at least one road is the road that a vehicle needs to travel from its current location to its target location, and the first type of guidance information is generated based on historical road topology. Obtain a second type of guidance information, which indicates the passability of lanes in the target road perceived by the vehicle; Based on the first type of guidance information and / or the second type of guidance information, the vehicle is controlled to travel towards the target location via the first lane of the target road.

2. The method according to claim 1, characterized in that, The step of controlling the vehicle to travel towards the target location via the first lane of the target road based on the first type of guidance information and / or the second type of guidance information includes: During the process of controlling the vehicle's movement based on the first type of guidance information. When the offset between the boundary of the target road indicated by the first type of guidance information and the boundary of the target road perceived by the vehicle is greater than or equal to a distance threshold, the system switches to controlling the vehicle to travel towards the target location based on the second type of guidance information; or... When the number of drivable lanes on the target road indicated by the first type of guidance information is inconsistent with the number of drivable lanes on the target road perceived by the vehicle, the system switches to controlling the vehicle to drive towards the target location based on the second type of guidance information.

3. The method according to claim 2, characterized in that, The method further includes: During the process of controlling the vehicle's movement based on the second type of guidance information When the offset between the boundary of the target road indicated by the first type of guidance information and the boundary of the target road perceived by the vehicle is less than the distance threshold, or when the number of drivable lanes of the target road indicated by the first type of guidance information is consistent with the number of drivable lanes of the target road perceived by the vehicle, the system switches to controlling the vehicle to drive towards the target location according to the first type of guidance information.

4. The method according to claim 2 or 3, characterized in that, The second type of guidance information is generated by the navigation neural network through processing navigation guidance information and road element information. The navigation guidance information indicates the R roads that the vehicle needs to take from the current position to the target position and the direction of travel. The R roads include the target road. The road element information indicates at least one lane-level element perceived by the vehicle at the current location. The at least one lane-level element indicates the lane boundary and / or lane centerline of each of the S lanes, where the S lanes are lanes in the R roads, and S and R are both positive integers.

5. The method according to claim 1, characterized in that, The step of controlling the vehicle to travel towards the target location via the first lane of the target road based on the first type of guidance information and / or the second type of guidance information includes: When the navigation directions corresponding to multiple lanes in the target road indicated by the first type of guidance information are opposite to the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, the vehicle is controlled to travel towards the target location via the first lane according to the second type of guidance information or according to the third type of guidance information. The third type of guidance information indicates the passability of the lanes included in the target road, and the third type of guidance information is generated in a different way than the first type of guidance information and the second type of guidance information.

6. The method according to claim 5, characterized in that, The method further includes: During the process of controlling the vehicle's movement based on the third type of guidance information. When the navigation direction corresponding to the multiple lanes indicated by the third type of guidance information is opposite to the navigation direction corresponding to the multiple lanes indicated by the second type of guidance information, the vehicle is controlled to travel towards the target location via the first lane according to the second type of guidance information.

7. The method according to claim 1, characterized in that, The step of controlling the vehicle to travel towards the target location via the first lane of the multiple lanes based on the first type of guidance information and the second type of guidance information includes: When the navigation directions corresponding to multiple lanes in the target road indicated by the first type of guidance information match the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, the first type of guidance information and the second type of guidance information are fused to obtain a fusion result; Based on the fusion result, the vehicle is controlled to travel towards the target location via the first lane.

8. The method according to claim 7, characterized in that, The navigation directions corresponding to the multiple lanes indicated by the first type of guidance information are matched with the navigation directions corresponding to the multiple lanes indicated by the second type of guidance information, including: lanes whose passability level indicated by the first type of guidance information is greater than or equal to a certain threshold, including lanes whose passability level indicated by the second type of guidance information is greater than or equal to the certain threshold.

9. The method according to any one of claims 5 to 8, characterized in that, The second type of guidance information is generated based on predefined rules.

10. The method according to claim 9, characterized in that, The method further includes: Based on environmental perception information, N1 sampling results corresponding to the target road are determined. Each sampling result indicates at least one passable lane corresponding to one of the N1 sampling planes. The distance between any two adjacent planes in the N1 sampling planes is the first distance. Based on the first navigation information, determine the M1 intersection guidance information associated with the N1 sampling planes, where M1 and N1 are both positive integers; Wherein, the environmental perception information indicates the road boundary of the target road perceived by the vehicle, and / or the lane boundary associated with the target road, and the first navigation information indicates at least one passable lane in the target road for the vehicle to travel to the target location, and the position of the at least one passable lane in the target road; Based on the N1 sampling results and the M1 intersection guidance information, M1*N1 combined results are determined, and each combined result indicates the passable lane among the multiple lanes and its position in the target road; Based on the M1*N1 combination results, the second type of guidance information is determined.

11. A navigation device, characterized in that, include: The acquisition unit is used to acquire a first type of guidance information, which indicates the passability of each lane in at least one road, wherein the at least one road is the road that a vehicle needs to travel from the current location to the target location, and the first type of guidance information is generated based on historical road topology. The acquisition unit is further configured to: acquire a second type of guidance information, wherein the second type of guidance information indicates the passability of lanes in the target road perceived by the vehicle; The processing unit is configured to control the vehicle to travel towards the target location via a first lane in the target road, based on the first type of guidance information and / or the second type of guidance information.

12. The apparatus according to claim 11, characterized in that, The processing unit is used for: During the process of controlling the vehicle's movement based on the first type of guidance information. When the offset between the boundary of the target road indicated by the first type of guidance information and the boundary of the target road perceived by the vehicle is greater than or equal to a distance threshold, the system switches to controlling the vehicle to travel towards the target location based on the second type of guidance information; or... When the number of drivable lanes on the target road indicated by the first type of guidance information is inconsistent with the number of drivable lanes on the target road perceived by the vehicle, the system switches to controlling the vehicle to drive towards the target location based on the second type of guidance information.

13. The apparatus according to claim 12, characterized in that, The processing unit is also used for: During the process of controlling the vehicle's movement based on the second type of guidance information When the offset between the boundary of the target road indicated by the first type of guidance information and the boundary of the target road perceived by the vehicle is less than the distance threshold, or when the number of drivable lanes of the target road indicated by the first type of guidance information is consistent with the number of drivable lanes of the target road perceived by the vehicle, the system switches to controlling the vehicle to drive towards the target location according to the first type of guidance information.

14. The apparatus according to claim 12 or 13, characterized in that, The second type of guidance information is generated by the navigation neural network through processing navigation guidance information and road element information. The navigation guidance information indicates the R roads that the vehicle needs to take from the current position to the target position and the direction of travel. The R roads include the target road. The road element information indicates at least one lane-level element perceived by the vehicle at the current location. The at least one lane-level element indicates the lane boundary and / or lane centerline of each of the S lanes, where the S lanes are lanes in the R roads, and S and R are both positive integers.

15. The apparatus according to claim 11, characterized in that, The processing unit is used for: When the navigation directions corresponding to multiple lanes in the target road indicated by the first type of guidance information are opposite to the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, the vehicle is controlled to travel towards the target location via the first lane according to the second type of guidance information or according to the third type of guidance information. The third type of guidance information indicates the passability of the lanes included in the target road, and the third type of guidance information is generated in a different way than the first type of guidance information and the second type of guidance information.

16. The apparatus according to claim 15, characterized in that, The processing unit is also used for: During the process of controlling the vehicle's movement based on the third type of guidance information. When the navigation direction corresponding to the multiple lanes indicated by the third type of guidance information is opposite to the navigation direction corresponding to the multiple lanes indicated by the second type of guidance information, the vehicle is controlled to travel towards the target location via the first lane according to the second type of guidance information.

17. The apparatus according to claim 11, characterized in that, The processing unit is used for: When the navigation directions corresponding to multiple lanes in the target road indicated by the first type of guidance information match the navigation directions corresponding to multiple lanes indicated by the second type of guidance information, the first type of guidance information and the second type of guidance information are fused to obtain a fusion result; Based on the fusion result, the vehicle is controlled to travel towards the target location via the first lane.

18. The apparatus according to claim 17, characterized in that, The navigation directions corresponding to the multiple lanes indicated by the first type of guidance information are matched with the navigation directions corresponding to the multiple lanes indicated by the second type of guidance information, including: lanes whose passability level indicated by the first type of guidance information is greater than or equal to a certain threshold, including lanes whose passability level indicated by the second type of guidance information is greater than or equal to the certain threshold.

19. The apparatus according to any one of claims 15 to 18, characterized in that, The second type of guidance information is generated based on predefined rules.

20. The apparatus according to claim 19, characterized in that, The processing unit is also used for: Based on environmental perception information, N1 sampling results corresponding to the target road are determined. Each sampling result indicates at least one passable lane corresponding to one of the N1 sampling planes. The distance between any two adjacent planes in the N1 sampling planes is the first distance. Based on the first navigation information, determine the M1 intersection guidance information associated with the N1 sampling planes, where M1 and N1 are both positive integers; Wherein, the environmental perception information indicates the road boundary of the target road perceived by the vehicle, and / or the lane boundary associated with the target road, and the first navigation information indicates at least one passable lane in the target road for the vehicle to travel to the target location, and the position of the at least one passable lane in the target road; Based on the N1 sampling results and the M1 intersection guidance information, M1*N1 combined results are determined, and each combined result indicates the passable lane among the multiple lanes and its position in the target road; Based on the M1*N1 combination results, the second type of guidance information is determined.

21. A navigation device, characterized in that, include: A processor for executing a computer program stored in memory to cause the apparatus to perform the method as described in any one of claims 1 to 10.

22. A computer-readable storage medium, characterized in that, It stores instructions that, when executed by a processor, implement the method as described in any one of claims 1 to 10.

23. A chip, characterized in that, The chip includes circuitry for performing the method as described in any one of claims 1 to 10.

24. A computer program product, characterized in that, The computer program product includes: computer program code, which, when executed by a processor, implements the method as described in any one of claims 1 to 10.

25. A vehicle, characterized in that, Includes the apparatus as described in any one of claims 11 to 21, or the computer-readable storage medium as described in claim 22, or the chip as described in claim 23, or the vehicle is equipped with the computer program product as described in claim 24.

Images