Autonomous driving method and apparatus, and intelligent driving device

Abstract

An autonomous driving method and apparatus, and an intelligent driving device. The method comprises: S710, acquiring guidance information, wherein the guidance information indicates a direction and position for an intelligent driving device to travel out of a first intersection, the guidance information is determined on the basis of SD map navigation information, and the SD map navigation information indicates a navigation route from the current position of the intelligent driving device to a target position; and S720, on the basis of the guidance information, controlling the speed and / or pose of the intelligent driving device when same travels to the first intersection or passes through the first intersection. In the method, when a sensing system of a vehicle cannot recognize a road boundary and / or a clear lane line of an intersection position, on the basis of guidance information, the vehicle can be controlled to travel at a relatively low speed, and thus the capability and efficiency of the vehicle in passing through complex topological scenarios such as an intersection and a ramp can be improved without depending on a high-precision map, thereby improving the autonomous driving capability of the vehicle.

Description

Automatic driving method, device and intelligent driving equipment

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of China on March 22, 2024, with application number 202410345295.X and invention name “Autonomous driving method, device and intelligent driving equipment”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of intelligent driving, and more specifically, to an automatic driving method, apparatus, and intelligent driving device. Background Art

[0003] The rapid development of the automotive industry has led to the emergence of numerous assisted driving and autonomous driving technologies, which can reduce driving stress, improve safety, and enhance traffic efficiency. Currently, most autonomous driving technologies rely on high-precision maps for navigation. However, these maps suffer from drawbacks such as high acquisition and production costs, time-consuming processes, insufficient coverage, and difficulty ensuring data freshness. This makes them difficult to promote nationwide or global.

[0004] In view of this, an autonomous driving solution that is independent of high-precision maps needs to be developed urgently. Summary of the Invention

[0005] The present application provides an autonomous driving method, apparatus, and intelligent driving device that can improve the ability and efficiency of intelligent driving devices through complex topological scenarios such as intersections and ramps without relying on high-precision maps.

[0006] In a first aspect, an autonomous driving method is provided. The method can be executed by an intelligent driving device, for example, by a computing platform of the intelligent driving device, or by a chip or circuit used for the intelligent driving device; alternatively, the method can also be executed by a cloud server associated with the intelligent driving device, which is not specifically limited in this application.

[0007] The method includes: obtaining traction information, the traction information indicating the direction and position of the intelligent driving device exiting a first intersection, the traction information being determined based on standard definition (SD) map navigation information, the SD map navigation information indicating a navigation route from the current position of the intelligent driving device to a target position; and controlling the speed and / or posture of the intelligent driving device toward or through the first intersection based on the traction information.

[0008] In some implementations, the first intersection may include at least two boundaries, such as a boundary on the side where the intelligent driving device enters and a boundary on the side where the intelligent driving device exits. Alternatively, the first intersection may include other boundaries in addition to the two boundaries. For example, the first intersection may be an n-fork intersection, where n is an integer greater than or equal to 3. It should be understood that each of the at least two boundaries of the first intersection connects to a road.

[0009] In some implementations, the traction information may include a traction road, that is, a road boundary connected to the boundary of the intelligent driving device after exiting the first intersection. The traction road may include at least one lane, and the traction road may be a one-way road.

[0010] In the above technical solution, when the intelligent driving device is in autonomous driving mode and is too far from an intersection to obtain intersection-related perception information (such as road boundaries or lane lines used for autonomous driving navigation) through the perception device, or when the obtained perception information is not accurate enough, the direction of exiting the intersection can be determined in advance based on the traction information, thereby adjusting the heading of the traction vehicle body to prevent the intelligent driving device from getting stuck near the intersection, improving the pass rate and success rate of the intelligent driving device at the intersection in the absence of a high-precision map, and reducing the chance of the intelligent driving device entering the opposite lane or colliding with the road boundary when exiting the intersection. In addition, adjusting the speed of the intelligent driving device based on the traction information helps improve the driving safety of the intelligent driving device.

[0011] In combination with the first aspect, in certain implementations of the first aspect, controlling the speed and / or posture of the intelligent driving device toward or through the first intersection based on traction information includes: controlling the intelligent driving device to drive toward or through the first intersection at a first speed based on the traction information, where the first speed is less than or equal to a speed threshold.

[0012] For example, the speed threshold may be 70% or 80% of the maximum speed limit at the first intersection, or another speed determined based on the maximum speed limit at the first intersection. The maximum speed limit at the first intersection may be the maximum speed limit of the road connected to the first intersection, for example, the maximum speed limit of the road the intelligent driving device is currently traveling on (e.g., the road entering the first intersection), or the maximum speed limit of the target road the intelligent driving device is traveling on (e.g., the road exiting the first intersection).

[0013] In the above technical solution, the intelligent driving device is controlled to travel at a slower speed based on the traction information, so that when the intelligent driving device approaches a specific boundary (that is, the boundary where the intelligent driving device exits the first intersection), it has sufficient time to obtain road information at the specific boundary, thereby increasing the probability of obtaining clear road information (such as clear lane lines), and further improving the regulation and control capabilities of the intelligent driving device during the automatic driving process.

[0014] In combination with the first aspect, in certain implementations of the first aspect, the first boundary of the first intersection is the target exit boundary of the intelligent driving device, and the first boundary is connected to the first road. The method also includes: obtaining information of the first road perceived by the intelligent driving device; and controlling the speed and / or posture of the intelligent driving device toward or through the first intersection based on the traction information, including: controlling the speed and / or posture of the intelligent driving device based on the information of the first road and the traction information.

[0015] The information of the first road perceived by the intelligent driving device may be understood as information of the first road acquired by a perception system (such as a camera device, a radar, etc.) of the intelligent driving device.

[0016] It is understood that when the first boundary is far from the intelligent driving device and beyond the intelligent driving device's perception range, the intelligent driving device cannot obtain the first road information, or the reliability and / or accuracy of the first road information obtained by the intelligent driving device are insufficient. As the intelligent driving device approaches the first boundary, the reliability and accuracy of the first road information obtained by the intelligent driving device gradually improve. Because the traction information is only information indicating the position and direction upon exiting the first intersection, generated based on the SD map navigation information, it may differ from the actual position and boundary of the first road. Therefore, when the intelligent driving device is sufficiently close to the first boundary, controlling the intelligent driving device's posture based on the first road information is more reliable than controlling the intelligent driving device's posture based on the traction information. The above "more reliable" can be understood as being able to avoid the intelligent driving device from crossing lane lines or road boundaries. Therefore, when navigating solely based on the SD map, controlling the intelligent driving device's speed and / or posture based on the first road information and the traction road during the process of approaching and passing the first intersection helps improve the intelligent driving device's pass rate through complex intersections and the robustness of the intelligent driving device's autonomous driving system.

[0017] In combination with the first aspect, in certain implementations of the first aspect, the information of the first road indicates the boundary of the first road and / or the position of the lane line in the first road; controlling the speed and / or posture of the intelligent driving device based on the information of the first road and the traction information includes: when the reliability of the information of the first road is less than or equal to a preset threshold, controlling the intelligent driving device to travel at a first speed based on the traction information; or, when the reliability of the information of the first road is greater than the preset threshold, controlling the intelligent driving device to switch from the first speed to the second speed based on the information of the first road; wherein the first speed is less than or equal to the second speed.

[0018] In the above technical solution, when the reliability of the first road information is greater than a preset threshold, the intelligent driving device is controlled to travel at a higher speed, which can improve the speed and efficiency of the intelligent driving device passing the first intersection.

[0019] In combination with the first aspect, in certain implementations of the first aspect, the traction information includes a traction road, and obtaining the traction information includes: obtaining boundary information of the first intersection; determining the direction of exiting the first intersection based on the SD map navigation information; determining the position of exiting the first intersection based on the boundary information and the direction of exiting the first intersection; and generating a traction road based on the position of exiting the first intersection and the direction of exiting the first intersection.

[0020] It should be noted that this application does not strictly distinguish between traction lanes and traction roads. If information about the number of lanes of the road (such as the first road described above) connected to the location of the intelligent driving device after exiting the first intersection is not obtained, the traction road generated based on the location of the first intersection and the direction of exiting the first intersection may include only one traction lane. If information about the number of lanes is n, the traction road generated based on the location of the first intersection and the direction of exiting the first intersection may include n traction lanes.

[0021] In the above technical solution, for roads beyond the perception range of the intelligent driving device, a traction road is generated based on the SD map navigation information, which can enable the intelligent driving device to adjust and advance in the target driving direction in advance, reducing the probability of the intelligent driving device entering the opposite lane or deviating from the road, and helping to improve the automatic driving capability and driving safety of the intelligent driving device in scenarios without high-precision maps.

[0022] In combination with the first aspect, in certain implementations of the first aspect, generating a traction road based on the position and direction of exiting the first intersection includes: generating an initial traction road based on the position and direction of exiting the first intersection; and correcting the initial traction road based on the position of the first road connected to the boundary of the intelligent driving device when exiting the first intersection to obtain the traction road.

[0023] Exemplarily, the location of the first road may be perceived by a perception system of an intelligent driving device.

[0024] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: controlling a display device to display a navigation interface, the navigation interface including an image of a traction road and a first icon, the first icon indicating a position of the intelligent driving device relative to the traction road.

[0025] Exemplarily, when the speed and / or posture of the intelligent driving device is controlled by traction information, the navigation interface is controlled to display an image of the traction road. The first icon may be an icon representing the intelligent driving device. The first icon and the image of the traction road indicate the relative positional relationship between the intelligent driving device and the traction road.

[0026] In the above technical solution, by displaying the image of the towing road, the user of the intelligent driving device can clearly understand the data source currently used for navigation, and can also warn the driver of the intelligent driving device to take over the intelligent driving device when necessary.

[0027] In combination with the first aspect, in some implementations of the first aspect, the navigation interface further includes an image of a boundary of the first road.

[0028] For example, the image of the boundary of the first road may be obtained by processing information about the first road perceived by a perception system of the intelligent driving device.

[0029] In some implementations, when the image of the traction road indicates a boundary of the first road, the navigation interface may be controlled to switch from displaying the image of the traction road to displaying the image of the boundary of the first road.

[0030] In yet other implementations, when the image of the traction road indicates the location of a lane line in the first road, the navigation interface may be controlled to superimpose and display the image of the traction road and an image of a boundary of the first road.

[0031] In the above technical solution, by displaying an image of the boundary of the first road, the user can clearly understand the difference between the towing road and the actual first road, as well as the perception result of the intelligent driving device on the current driving road, so that the driver can take over the intelligent driving device when necessary.

[0032] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: when the lane line information of the first road is acquired, controlling the navigation interface to switch from the image of the traction road to the image of the lane line of the first road.

[0033] Exemplarily, when the speed and / or posture of the intelligent driving device is controlled by the lane lines of the first road, the navigation interface is controlled to switch from displaying an image of the traction road to displaying an image of the lane lines of the first road.

[0034] The "switching" of the content of the navigation interface of this application can be understood as the result of refreshing the navigation interface in real time.

[0035] In the above technical solution, by switching images, the user can determine the improvement of the intelligent driving device's perception ability of the first road, thereby increasing confidence in the automatic driving function of the intelligent driving device.

[0036] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: obtaining second road information within a first range of the intelligent driving device, the second road information indicating multiple perceived routes passing through the first intersection; determining a main channel from the multiple perceived routes based on SD map navigation information; controlling the speed and / or posture of the intelligent driving device toward the first intersection or passing through the first intersection, including: controlling the speed and / or posture of the intelligent driving device when traveling along the main channel.

[0037] In the above technical solution, when controlling the intelligent driving device to drive towards the first intersection and passing through the first intersection, it is necessary to control the intelligent driving device to drive along the main channel to avoid the intelligent driving device from yaw and improve the robustness of the automatic driving function of the intelligent driving device.

[0038] In a second aspect, an automatic driving device is provided, which includes an acquisition unit and a processing unit, wherein the acquisition unit is used to: acquire traction information, the traction information indicating the direction and position of the intelligent driving device exiting the first intersection, the traction information is determined based on SD map navigation information, and the SD map navigation information indicates the navigation route from the current position of the intelligent driving device to the target position; the processing unit is used to: control the speed and / or posture of the intelligent driving device toward the first intersection or through the first intersection based on the traction information.

[0039] In combination with the second aspect, in some implementations of the second aspect, the processing unit is used to: control the intelligent driving device to travel at a first speed based on the traction information, where the first speed is less than or equal to a speed threshold.

[0040] In combination with the second aspect, in certain implementations of the second aspect, the first boundary of the first intersection is the target exit boundary of the intelligent driving device, and the first boundary is connected to the first road. The acquisition unit is also used to: obtain information of the first road perceived by the intelligent driving device; the processing unit is used to: control the speed and / or posture of the intelligent driving device based on the information of the first road and the traction information.

[0041] In combination with the second aspect, in certain implementations of the second aspect, the information of the first road indicates the boundary of the first road and / or the position of the lane line in the first road; the processing unit is used to: when the reliability of the information of the first road is less than or equal to a preset threshold, control the intelligent driving device to travel at a first speed according to the traction information; or, when the reliability of the information of the first road is greater than the preset threshold, control the intelligent driving device to switch from the first speed to the second speed according to the information of the first road; wherein the first speed is less than or equal to the second speed.

[0042] In combination with the second aspect, in certain implementations of the second aspect, the traction information includes a traction road, and the acquisition unit is further used to: obtain boundary information of the first intersection; the processing unit is further used to: determine the direction of exiting the first intersection based on the SD map navigation information; determine the position of exiting the first intersection based on the boundary information and the direction of exiting the first intersection; and generate a traction road based on the position of exiting the first intersection and the direction of exiting the first intersection.

[0043] In combination with the second aspect, in certain implementations of the second aspect, the processing unit is used to: generate an initial traction road based on the position and direction of exiting the first intersection; and correct the initial traction road based on the position of the first road connected at the boundary of the intelligent driving device exiting the first intersection to obtain the traction road.

[0044] In combination with the second aspect, in certain implementations of the second aspect, the processing unit is further used to: control the display device to display a navigation interface, the navigation interface including an image of the traction road and a first icon, the first icon indicating the position of the intelligent driving device relative to the traction road.

[0045] In combination with the second aspect, in some implementations of the second aspect, the navigation interface further includes an image of a boundary of the first road.

[0046] In combination with the second aspect, in some implementations of the second aspect, the processing unit is further used to: when the lane line information of the first road is obtained, control the navigation interface to switch from the image of the traction road to the image of the lane line of the first road.

[0047] In combination with the second aspect, in certain implementations of the second aspect, the acquisition unit is further used to: obtain second road information within the first range of the intelligent driving device, the second road information indicating multiple perceived routes passing through the first intersection; the processing unit is further used to: determine the main channel from the multiple perceived routes based on the SD map navigation information; and control the speed and / or posture of the intelligent driving device when traveling along the main channel.

[0048] In a third aspect, an automatic driving device is provided, comprising: a processor for executing a computer program stored in the memory so that the device performs the method in any possible implementation of the first aspect above.

[0049] In combination with the third aspect, in certain implementations of the third aspect, the automatic driving device also includes a memory.

[0050] In a fourth aspect, an intelligent driving device is provided, which includes the device in any possible implementation of the second to third aspects.

[0051] In combination with the fourth aspect, in some implementations of the fourth aspect, the intelligent driving device is a vehicle.

[0052] In a fifth aspect, a computer program product is provided, comprising: a computer program code, which, when executed on a computer or a processor, enables the computer or the processor to execute the method in any possible implementation of the first aspect.

[0053] It should be noted that the above-mentioned computer program code may be stored in whole or in part on a storage medium, wherein the storage medium may be packaged together with the processor or separately from the processor.

[0054] In a sixth aspect, a computer-readable medium is provided, wherein the computer-readable medium stores instructions. When the instructions are executed by a processor, the processor implements the method in any possible implementation of the first aspect.

[0055] In a seventh aspect, a chip is provided, which includes a circuit for executing the method in any possible implementation of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] FIG1 is a functional schematic block diagram of an intelligent driving device provided in an embodiment of the present application;

[0057] FIG2 is a schematic diagram of an autonomous driving system architecture provided in an embodiment of the present application;

[0058] FIG3 is a schematic flowchart of an autonomous driving method provided in an embodiment of the present application;

[0059] FIG4 is a schematic diagram of an application scenario of the autonomous driving method provided in an embodiment of the present application;

[0060] FIG5 is a schematic diagram of a matching result between a perception route and an SD map navigation route according to an embodiment of the present application;

[0061] FIG6 is another schematic flowchart of the autonomous driving method provided in an embodiment of the present application;

[0062] FIG7 is a schematic diagram of a result of determining an exit location according to an embodiment of the present application;

[0063] FIG8 is another schematic flowchart of the autonomous driving method provided in an embodiment of the present application;

[0064] FIG9 is a schematic diagram of a traction road provided in an embodiment of the present application;

[0065] FIG10 is another schematic flowchart of the autonomous driving method provided in an embodiment of the present application;

[0066] FIG11 is another schematic flowchart of the autonomous driving method provided in an embodiment of the present application;

[0067] FIG12 is a schematic block diagram of an automatic driving device provided in an embodiment of the present application;

[0068] FIG13 is another schematic block diagram of the automatic driving device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0069] Before introducing the solution of this application, we first introduce the relevant concepts involved in this application:

[0070] 1. SD maps: These maps generally have meter-level accuracy and low richness. They primarily include road information and point of interest (POI) information. POIs are point-type data in electronic maps and contain at least four attributes: name, address, coordinates, and category.

[0071] 2. High-definition (HD) maps: HD maps offer higher accuracy and richness than SD maps, with both absolute and relative accuracy at the centimeter level. In terms of richness, HD maps provide a model of the environment in which autonomous vehicles operate, including static HD maps and other dynamic information. Static HD maps include lane models, road components, and road attributes. Lane models include road details such as lane lines, lane centerlines, and changes in lane attributes. Other dynamic information includes all dynamic information within the intelligent network system, including map dynamics, sensor information, driving behavior, and traffic dynamic information management.

[0072] As mentioned above, most current autonomous driving technologies rely on high-precision maps for navigation. However, high-precision maps have drawbacks such as high acquisition and production costs, long production times, insufficient coverage, and difficulty ensuring data freshness. This makes autonomous driving technologies that rely on high-precision maps difficult to promote nationwide or globally. Without relying on high-precision maps, navigation relies solely on environmental information perceived by vehicle-side sensors. Due to the limited perception range of vehicle-side sensors, they cannot obtain information beyond their perception range, thus limiting the vehicle's autonomous driving capabilities. For example, when a vehicle needs to pass through a large or irregular intersection, or needs to turn left or right at an intersection, the vehicle may not be able to obtain information such as the lane position and road boundary at the exit location, thereby affecting the vehicle's autonomous driving planning capabilities. For example, the vehicle cannot determine the direction of deviation and direction of travel, which in turn leads to low vehicle traffic efficiency in the above scenarios.

[0073] In view of this, the embodiments of the present application provide an automatic driving method, apparatus, and intelligent driving device. When the road ahead of the intelligent driving device includes an intersection, traction information for the intelligent driving device in the direction of exiting the intersection is generated based on the road ahead information (including information about the intersection boundary) obtained by the intelligent driving device and the navigation route generated based on the SD map. The traction information indicates the direction and position of the intelligent driving device exiting the intersection. The intelligent driving device controls the posture and speed of the intelligent driving device through the intersection based on the traction information. When the perception system of the intelligent driving device fails to identify the road boundary and / or clear lane line at the exit location, the intelligent driving device can be controlled to travel at a lower speed based on the traction information. As the intelligent driving device approaches the exit location, when the perception system of the intelligent driving device can identify the road boundary and / or clear lane line at the exit location, the intelligent driving device can be controlled to travel at a higher speed based on the road boundary and / or clear lane line. This improves the traffic efficiency of the intelligent driving device through the intersection in the absence of a high-precision map, thereby improving the automatic driving capability of the intelligent driving device.

[0074] The technical solution in this application will be described below with reference to the accompanying drawings.

[0075] Figure 1 is a functional block diagram of an intelligent driving device provided in an embodiment of the present application. As shown in Figure 1, the intelligent driving device 100 may include a perception system 120, a display device 130 and a computing platform 150, wherein the perception system 120 may include several sensors for sensing information about the environment surrounding the intelligent driving device 100. For example, the perception system 120 may include a positioning system, and the positioning system may be a global positioning system (GPS), a Beidou system or other positioning systems. For another example, the perception system 120 may also include one or more of an inertial measurement unit (IMU), a laser radar, a millimeter wave radar, an ultrasonic radar and a camera device.

[0076] The display device 130 is mainly divided into two categories: the first category is the vehicle-mounted display screen; the second category is a projection display screen, such as a head-up display (HUD). The vehicle-mounted display screen is a physical display screen and an important component of the vehicle infotainment system. The vehicle-mounted display screen may include a human-machine interface (HMI). A head-up display, also known as a head-up display system, is mainly used to display driving information such as speed and navigation on a display device in front of the user (such as a windshield) to reduce the user's line of sight diversion time, avoid pupil changes caused by the user's line of sight diversion, and improve driving safety and comfort.

[0077] Some or all functions of the intelligent driving device 100 can be controlled by a computing platform 150. The 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 the ability to read and execute instructions, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), or a digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. The logical relationships of the hardware circuits are fixed or reconfigurable. For example, the processor is a hardware circuit implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as a field programmable gate array (FPGA). In a reconfigurable hardware circuit, the process of the processor loading a configuration file to implement the hardware circuit configuration can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units. In addition, the processor may also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), etc. In addition, the computing platform 150 may also include a memory for storing instructions, and some or all of the processors 151 to 15n may call the instructions in the memory to implement corresponding functions.

[0078] The intelligent driving device 100 may include ADAS, which uses a variety of sensors on the intelligent driving device (including but not limited to: lidar, millimeter wave radar, camera device, ultrasonic sensor, global positioning system, inertial measurement unit) to obtain information from the surrounding of the intelligent driving device, and analyze and process the obtained information to achieve functions such as obstacle perception, target recognition, intelligent driving device positioning, path planning, driver monitoring / reminder, etc., thereby improving the safety, automation and comfort of driving the intelligent driving device.

[0079] From a logical function perspective, ADAS systems generally include three main functional modules: perception module, decision module and execution module. The perception module perceives the surrounding environment of the vehicle body through sensors and inputs corresponding real-time data to the decision-making layer processing center. The perception module mainly includes on-board cameras / ultrasonic radars / millimeter-wave radars / lidars, etc.; the decision module uses computing devices and algorithms to make corresponding decisions based on the information obtained by the perception module; the execution module takes corresponding actions after receiving the decision signal from the decision module, such as driving, changing lanes, steering, braking, warnings, etc.

[0080] ADAS can provide varying degrees of automated driving assistance at different levels of automation (L0-L5), based on artificial intelligence algorithms and information from multiple sensors. These levels are based on the Society of Automotive Engineers (SAE) grading standards. L0 is no automation; L1 is driving assistance; L2 is partial automation; L3 is conditional automation; L4 is high automation; and L5 is full automation. At L1-L3, monitoring and responding to road conditions are performed jointly by the driver and the system, with the driver taking over dynamic driving tasks. At L4 and L5, the driver transitions completely to the role of passenger. Currently, ADAS features include, but are not limited to, adaptive cruise control, automatic emergency braking, automated parking, blind spot monitoring, front cross-traffic alert / braking, rear cross-traffic alert / braking, forward collision warning, lane departure warning, lane keep assist, rear collision warning, traffic sign recognition, traffic jam assistance, and highway assistance. It should be understood that the various functions described above may have specific modes at different autonomous driving levels (L0-L5). The higher the autonomous driving level, the smarter the corresponding mode.

[0081] In an embodiment of the present application, the computing platform 150 may generate traction information based on the SD map and the surrounding environment information of the intelligent driving device obtained by the perception system 120, or the computing platform 150 may also control the display device 130 to display the traction information.

[0082] FIG2 shows a schematic diagram of the autonomous driving system architecture provided by an embodiment of the present application. As shown in FIG2 , the system includes a perception module 210, a map information acquisition module 220, a road information generation module 230, a regulation and control module 240, and a display module 250. The perception module 210 may include one or more cameras in the perception system 120 shown in FIG1 , or may also include one or more radars in the perception system 120; the map information acquisition module 220, the road information generation module 230, and the regulation and control module 240 may each include one or more processors in the computing platform 150 shown in FIG1 ; and the display module 250 may include one or more display devices 130, such as an HMI.

[0083] The perception module 210 is used to obtain environmental information surrounding the intelligent driving device and transmit it to the road information generation module 230. This environmental information includes road information, such as the road boundaries and lane markings of the road currently being traveled by the intelligent driving device. For example, the perception module 210 may also include one or more processors to process the collected perception information to obtain environmental information. For example, if the perception information is an image, the one or more processors in the perception module 210 may extract environmental information such as road boundaries, intersection boundaries, and lane markings from the acquired image.

[0084] The map information acquisition module 220 is used to determine a navigation route between the current position of the intelligent driving device and the target position based on the SD map, and then send the navigation route to the road information generation module 230.

[0085] The road information generation module 230 includes a main channel determination module 231, an exit information determination module 232, and a guidance information generation module 233. The main channel determination module determines the main channel from multiple perceived routes based on the degree of match between the perceived routes identified by the perception module 210 and the navigation route. The exit information determination module 232 determines the exit location and direction based on the intersection boundaries of the main channel identified by the perception module 210. The guidance information generation module 233 generates a virtual towing road based on the exit location and direction. When the perception module 210 determines the boundaries of the real road corresponding to the towing road based on the perception information, the guidance information generation module 233 can also correct the position of the towing road based on the boundaries of the real road. Furthermore, the road information generation module 230 sends the towing road to the regulation and control module 240.

[0086] The regulation and control module 240 can also obtain the boundary of the real road corresponding to the towing road determined by the perception module 210, as well as the clear lane lines of the real road corresponding to the towing road collected by the perception module 210. Furthermore, the regulation and control module 240 can control the speed of the intelligent driving device based on the perception results of the perception module 210. The regulation and control module 240 can also control the display module 250 to display different navigation elements, such as one or more of the towing road, the real road boundary, and the clear lane lines.

[0087] It should be understood that the above modules are merely examples; in actual applications, these modules may be added or removed based on actual needs. For example, in the system architecture shown in Figure 2, the map information acquisition module 220 and the road information generation module 230 can be combined into a single module; alternatively, the road information generation module 230 and the regulation and control module 240 can be combined into a single module. For another example, the processor in the perception module 210 can be a separate processing module, which can be located in the intelligent driving device or in a cloud server that communicates with the intelligent driving device.

[0088] The above introduces the autonomous driving system architecture provided in the embodiment of the present application. The following details the process of implementing the autonomous driving method provided in the embodiment of the present application based on the autonomous driving system shown in Figure 2.

[0089] The intelligent driving devices involved in the embodiments of the present application may include road vehicles, water vehicles, air vehicles, industrial equipment, agricultural equipment, or entertainment equipment, etc. For example, the intelligent driving device can be a vehicle, which is a vehicle in a broad sense, and can be a vehicle (such as a commercial vehicle, a passenger car, a motorcycle, a flying car, a train, etc.), an industrial vehicle (such as a forklift, a trailer, a tractor, etc.), an engineering vehicle (such as an excavator, a bulldozer, a crane, etc.), agricultural equipment (such as a mower, a harvester, etc.), amusement equipment, a toy vehicle, etc. The embodiments of the present application do not specifically limit the type of vehicle. For ease of understanding, the following description is based on the intelligent driving device being a vehicle as an example.

[0090] FIG3 shows a schematic flow chart of an autonomous driving method provided by an embodiment of the present application. This method 300 can be applied to the intelligent driving device shown in FIG1 , or the method can be executed by the system shown in FIG2 . More specifically, this method 300 can be executed by the main course determination module 231 , and the method 300 can include:

[0091] S301: Determine a plurality of perception routes based on environmental information acquired by a perception system, where each of the plurality of perception routes is a route connecting one road to another road via an intersection.

[0092] In some implementations, the environmental information may include an image of a preset range in front of the vehicle. Elements such as roads and intersections are extracted from the image and then stitched together to generate multiple perceived routes. For example, the preset range may be the maximum range that the vehicle's camera or radar can perceive, or it may be within 150 meters, 200 meters, or any other range in front of the vehicle.

[0093] For example, as shown in (a) in Figure 4, when the vehicle is traveling from Road A to Intersection a, an image containing Intersection a and parts of Roads AD is acquired. Then, based on image processing, elements such as Road A, Road B, Road C, Road D, and Intersection a are extracted. Then, based on the topological relationship of the roads, the following three perception routes are determined: a route from Road A to Road B via Intersection a, a route from Road A to Road C via Intersection a, and a route from Road A to Road D via Intersection a.

[0094] S302: Determine a main channel from multiple perceived routes based on the SD map navigation route.

[0095] Among them, the SD map navigation route is a partial or complete navigation route from the current position of the vehicle to the target position generated based on the SD map. For example, the SD map navigation route can be a navigation route within a certain range of the current position of the vehicle intercepted from all navigation routes. The SD map navigation route is composed of multiple points, and each point indicates a geographical space location.

[0096] For example, the current position of the vehicle can be determined based on GPS data, and then an SD navigation route matching the current position of the vehicle can be determined, wherein the starting point of the SD map navigation route can be the current position of the vehicle, or the starting point of the SD map navigation route can also be within a preset distance after the current position of the vehicle, for example, the preset distance can be 20 meters, or 30 meters, or other distances; the total length of the SD map navigation route can be 100 meters, or 120 meters, or any other length.

[0097] For example, when the vehicle is traveling to the target location, it is necessary to change from Road A shown in (a) in Figure 4 to Road B via Intersection a. The SD map navigation route can be as shown in (b) in Figure 4. Furthermore, the SD map navigation route can be geometrically matched with multiple perception routes respectively, and the degree of overlap between multiple shape points of the SD navigation route and each perception route is calculated, and the perception route with the highest overlap is selected as the main channel. (a) to (c) in Figure 5 show the results of geometric matching of the SD map navigation route with the three perception routes respectively. It can be seen that the degree of overlap between the SD map navigation route and the route from Road A to Road B via Intersection a is the highest. Therefore, it can be determined that the route from Road A to Road B via Intersection a is the main channel.

[0098] The autonomous driving method provided in the embodiment of the present application can determine the main channel for the vehicle to travel to the target location based on the SD map navigation route and the road information perceived by the vehicle, so that in the absence of a high-precision map, the vehicle can be controlled to travel along the main channel to the target location.

[0099] FIG6 shows another schematic flow chart of an autonomous driving method provided by an embodiment of the present application. This method 400 can be applied to the intelligent driving device shown in FIG1 , or the method can be executed by the system shown in FIG2 . More specifically, the method 400 can be executed by the exit information determination module 232 , and the method 400 can include:

[0100] S401: Determine the intersection boundaries of the intersections that the main channel passes through based on the environmental information obtained by the perception system.

[0101] For example, the intersections that the main channel passes through may include intersection a in method 300 .

[0102] In some implementations, the environmental information may include an image of a preset range in front of the vehicle. An image of the intersection area through which the main channel passes may be extracted from the image, and a perceived intersection boundary may be determined based on the image of the intersection area. The perceived intersection boundary may be a boundary of the intersection area directly extracted from the image. In some scenarios, because the vehicle is far away from the intersection, the image acquired by the vehicle's perception system only includes an image of a portion of the intersection area. For example, as shown in (a) of FIG7 , the perceived intersection boundary is a boundary determined based on an image acquired by the vehicle that includes a portion of the intersection area. The initial intersection boundary may be adjusted based on the length of the boundary on the side entering the intersection (e.g., boundary a) to obtain a predicted intersection boundary, so that the shape and position of the predicted intersection boundary are as close as possible to the shape and position of the actual intersection boundary, and the predicted intersection boundary is used as the intersection boundary of the intersection through which the main channel passes.

[0103] As the vehicle moves forward, the area of ​​the intersection region included in the image acquired by the vehicle becomes larger and larger, and the intersection region becomes more and more complete. When the area of ​​the intersection region extracted through the image (i.e., the area enclosed by the initial intersection boundary) is greater than or equal to the area enclosed by the predicted intersection boundary, or when the farthest end of the intersection region (such as boundary b) is included in the image acquired by the vehicle and the tracking is stable, it can be determined that the perceived intersection boundary is the boundary of the complete (or real) intersection area. Further, it can be determined that the perceived intersection boundary is the intersection boundary of the intersection passed by the main channel.

[0104] It should be noted that the intersection boundary involved in this application refers to the area where the road and the intersection meet. For example, the stop line at a real intersection can be regarded as the boundary of the intersection.

[0105] S402, determining the exit direction according to the SD map navigation route.

[0106] For example, the SD map navigation route can be divided into an entrance route and an exit route based on the SD map navigation route and the intersection location. The entrance route indicates the direction of the intersection through which the vehicle enters the main channel, and the exit route indicates the direction of the intersection through which the vehicle exits the main channel. Furthermore, a certain number of points can be selected on the exit route, and the exit direction can be determined based on the selected points. The certain number can be 3, 5, or other values.

[0107] For example, as shown in (b) in FIG7 , three shape points may be taken at ①, ②, and ③ respectively, and fitting may be performed according to the coordinates indicated by the three shape points to obtain the exit direction as shown in (c) in FIG7 .

[0108] S403: Determine the exit location based on the intersection boundary and the exit azimuth.

[0109] For example, the intersection boundary can be a predicted intersection boundary, or it can be a perceived intersection boundary. As the vehicle travels in the main lane toward the intersection, the specific type of the intersection boundary can be switched. The specific switching rules can be referred to the description in S401 and will not be repeated here.

[0110] In some implementations, the geometric center of the intersection is determined based on the intersection boundary, and the intersection of the ray from the geometric center of the intersection along the exit direction and the intersection boundary (such as boundary c) is the initial exit position. Furthermore, based on the initial exit position, a first distance can be offset toward the vehicle's travel side to avoid the oncoming lane to obtain the final exit position. The first distance can be a preset distance, such as 3 meters or 2 meters, or the preset distance can also be another distance determined based on the number of lanes on the road at the exit. The number of lanes on the road at the exit can be the number of lanes of the electronic horizon provider (EHP).

[0111] It should be noted that when the traffic rules are based on the principle of right-hand traffic, the side on which the vehicle travels is the right side, that is, based on the initial exit position, the final exit position is obtained by offsetting the first distance to the right; when the traffic rules are based on the principle of left-hand traffic, the side on which the vehicle travels is the left side, that is, based on the initial exit position, the final exit position is obtained by offsetting the first distance to the left.

[0112] The autonomous driving method provided in the embodiment of the present application can determine the position and direction of the vehicle exiting the intersection along the main channel based on the SD map navigation route and the intersection boundary perceived by the vehicle.

[0113] FIG8 shows another schematic flow chart of an autonomous driving method provided by an embodiment of the present application. This method 500 can be applied to the intelligent driving device shown in FIG1 , or the method can be executed by the system shown in FIG2 . More specifically, the method 500 can be executed by the guidance information generation module 233 , and the method 500 can include:

[0114] S501: Generate a traction lane / road for exiting the intersection according to the exit position and exit direction.

[0115] For example, as shown in FIG9 , a traction road can be generated with a straight line passing through the exit and along the exit as the road centerline. The width of the traction road can be a first preset width. For example, the first preset width can be 5 meters, 6 meters, or other values.

[0116] In some implementations, when information about the number of lanes on the road at the exit is obtained, the width of the traction road can be the product of the number of lanes and a second preset width. For example, the second preset width can be 3.5 meters, 3.6 meters, or other values. For example, if the number of lanes is 2 and the second preset width is 3.5 meters, the width of the traction road is 2 * 3.5 = 7 meters. For another example, if the number of lanes is 1 and the second preset width is 3.5 meters, the width of the traction road is 3.5 meters. Furthermore, multiple traction lanes can be generated on the traction road based on the number of lanes.

[0117] In some implementations, the traction lane / road is a one-way lane / road. Hereinafter, "traction lane" refers to "traction lane / road." Unless otherwise specified, the traction lane may refer to either a traction road or a traction lane.

[0118] S502: Correct the position of the traction road.

[0119] In some implementations, the position of the traction road can be dynamically corrected based on an inertial filtering method. For example, the position of the traction road can satisfy the following relationship: Q(xi,yi)=AQ f (xi,yi)+BQ f-1 (xi,yi);

[0120] Among them, Q(xi,yi) represents the coordinates of point i on the boundary of the corrected traction road, Q f (xi,yi) represents the coordinates of point i on the boundary of the traction road determined in the current frame, Q f-1 (xi, yi) are the coordinates of point i on the traction lane boundary determined in the previous frame. A and B represent the weights of the results from the current frame and the previous frame, respectively, with A:B = 1:9. Alternatively, the ratio of A to B can be other values. For example, running methods 400 and 500 once can yield one frame of traction lane results. That is, the current frame represents the coordinates of the traction lane obtained in the current run of methods 400 and 500, and the previous frame represents the coordinates of the traction lane obtained in the immediately preceding run of methods 400 and 500.

[0121] In some other implementations, the position of the traction road can also be corrected based on the position of the fuzzy road, wherein the fuzzy road indicates the road boundary determined based on the environment obtained by the perception system, and the fuzzy road is a one-way road. As the vehicle travels along the main channel toward the exit, the vehicle can obtain environmental information near the exit and determine the boundary of the road at the exit, i.e., the fuzzy road, based on the environmental information near the exit. The position of the fuzzy road may deviate from the position of the traction road generated in S501 to a certain extent, and the position of the traction road can be adjusted based on the position of the fuzzy road. Exemplarily, the position of the traction road in the current frame can satisfy the following relationship: Q f (xi,yi)=CQ f’ (xi,yi)+DQ s (xi,yi);

[0122] Among them, Q f’ (xi,yi) is the coordinate of point i on the boundary of the traction road generated in the current frame, Q s(xi,yi) is the coordinate of point i on the boundary of the fuzzy road, C and D represent Q f’ (xi,yi) and Q s The weights of (xi,yi) are C:D=8:2, or the ratio of C to D can be other values.

[0123] The autonomous driving method provided in the embodiment of the present application can generate a traction lane based on the position and direction of exiting the intersection determined in method 400, so as to control the posture and / or speed of the vehicle when passing through the intersection.

[0124] FIG10 shows another schematic flow chart of the autonomous driving method provided by an embodiment of the present application. The method 600 can be applied to the intelligent driving device shown in FIG1 , or the method can be performed by the system shown in FIG2 . More specifically, the method 600 can include S601 and S602 . S601 and S602 can be performed by the regulation and control module 240 ; alternatively, S601 can be performed by the perception module 210 , and S602 can be performed by the regulation and control module 240 .

[0125] S601: Determine the navigation information source based on the perception result of the road information at the exit.

[0126] Exemplarily, a navigation information source refers to information used to control the vehicle's position and / or speed to guide the vehicle toward an exit. The navigation information source may include one or more of a towpath, a fuzzy path, and a clear lane. The fuzzy path indicates the boundary of the road at the exit, determined based on environmental information perceived by the vehicle, while the clear lane indicates the position of the lane markings on the road at the exit, determined based on environmental information perceived by the vehicle.

[0127] It is understandable that when the vehicle is far away from the exit, it may not be able to obtain effective environmental information at the exit. The above-mentioned "effective environmental information at the exit" refers to the perception information (such as images and other information) that can extract the road boundary and / or lane line. As the vehicle gradually moves closer to the exit, the vehicle can obtain effective environmental information at the exit. For example, as the distance between the vehicle and the exit gets closer, the fuzzy road and clear lane can be extracted from the environmental information at the exit in turn. For example, when the vehicle is 50 meters away from the exit, it may be possible to extract a fuzzy road from the environmental information at the exit, but due to the limitation of perception accuracy, it may not be possible to extract a clear lane from the environmental information at the exit; when the vehicle is 20 meters away from the exit, it may be possible to extract a clear lane from the environmental information at the exit.

[0128] If the vehicle fails to obtain information about fuzzy roads and clear lanes, it can determine the towing road as the navigation information source. If the vehicle obtains information about fuzzy roads or clear lanes, it can switch to the fuzzy roads or clear lanes as the navigation information source. For example, the navigation information source can be determined and / or switched based on the information shown in Table 1.

[0129] Table 1

[0130] In some implementations, before switching the navigation information source, a determination can be made based on the reliability (or confidence) of the switching target. When the reliability of the switching target is greater than or equal to a reliability threshold, the navigation information source is switched. The switching target refers to the navigation information source after switching. For example, when switching from a traction road to an obscure road, the obscure road is the switching target. When information about the obscure road is obtained but information about a clear lane is not, and the reliability of the obscure road is greater than or equal to a reliability threshold, the navigation information source is switched from the traction road to the obscure road.

[0131] Exemplarily, the reliability of the switching target can be determined according to the formula P = aP1 + bP2, where P represents reliability, P1 represents reliability determined based on statistical data, and P2 represents reliability determined based on a machine learning algorithm. a and b represent the weights of P1 and P2, respectively, with a and b both taking values ​​between 0 and 1, and the sum of a and b being 1. More specifically, the statistical data can be the result of a statistical analysis of multiple frames of blurred road (or clear lane) position information. For example, the variance of samples comprising multiple frames of position information can be determined, and P1 can be determined based on the variance. It is understood that a larger variance indicates a greater deviation between position information in different frames, and thus, the collected results can be considered unstable, resulting in a smaller P1. The model used by the machine learning algorithm can be trained based on manually annotated road boundary (or lane line) positions. By inputting the road boundary (or lane line) positions collected by the vehicle's perception system into the aforementioned model, the deviation between the data collected by the vehicle's perception system and the actual road (or lane) position can be obtained, and P2 can be determined based on this deviation. It can be understood that the larger the deviation, the less accurate the result collected by the vehicle, and the smaller P2.

[0132] Exemplarily, when the navigation information source is a traction road and the reliability of the fuzzy road is greater than reliability threshold 1, the vehicle switches from the traction road to the fuzzy road. When the navigation information source is a traction road or a fuzzy road and the reliability of the clear lane is greater than reliability threshold 2, the vehicle switches from the traction road or fuzzy road to the clear lane. Reliability threshold 2 is greater than reliability threshold 1. Exemplarily, reliability threshold 2 can be a value between 0.6 and 1, and reliability threshold 1 can be a value between 0.3 and 0.5. Alternatively, when different methods are used to determine the reliability thresholds, reliability thresholds 1 and 2 can also be other values.

[0133] In some implementations, the next navigation information source switch is performed after a certain time period after the previous navigation information source switch is performed. The certain time period may be 15 seconds, or 20 seconds, or other time periods.

[0134] S602: Control the position and / or speed of the vehicle in the main channel according to the navigation information source.

[0135] In some implementations, when the navigation information source is a towpath, the vehicle's position upon entering the intersection is controlled based on the location and direction of the exit indicated by the towpath. This allows the vehicle to travel along the main lane in the direction indicated by the towpath, avoiding collisions with the road boundary at the exit or conflicts with traffic in the oncoming lane. Furthermore, when the navigation information source is a towpath, the vehicle is controlled to travel at speed 1. Exemplarily, speed 1 can be any speed between 5 kilometers per hour (kph) and 15 kph. This allows the vehicle sufficient time to acquire road information (e.g., road boundaries, lane markings, etc.) at the intersection through the perception system.

[0136] When the navigation information source indicates an fuzzy road, the vehicle can be controlled to travel at speed 2; when the navigation information source indicates a clear lane, the vehicle can be controlled to travel at speed 3. Speed ​​2 can be any speed between 20 kph and 30 kph, and speed 3 can be any speed between 30 kph and 60 kph. It should be understood that speeds 1, 2, and 3 are merely exemplary. In actual implementation, speeds 1, 2, and 3 can also be determined based on the maximum speed limit at the intersection. For example, speed 1 can be less than or equal to 70% or 80% of the maximum speed limit at the intersection; speed 2 can be less than or equal to 80% or 90% of the maximum speed limit at the intersection; and speed 3 can be less than or equal to the maximum speed limit at the intersection. The maximum speed limit at the intersection can be the maximum speed limit at the intersection, or it can be the maximum speed limit of a road connected to the intersection, for example, the maximum speed limit of the road the vehicle is currently traveling on (e.g., the road entering the intersection), or it can be the maximum speed limit of the road the vehicle is currently traveling on (e.g., the road exiting the intersection). Alternatively, speed 1, speed 2, and speed 3 may also be other values ​​that satisfy the following conditions: speed 1 is less than or equal to speed 2, and speed 2 is less than speed 3.

[0137] In some implementations, there is a certain deviation between the position of the traction road and the position of the fuzzy road or clear lane indication. In this case, when the navigation information source is a fuzzy road or a clear lane, the vehicle's posture can be further adjusted according to the fuzzy road and the clear lane to improve navigation accuracy.

[0138] The automatic driving method provided in the embodiment of the present application can control the vehicle's position and / or speed based on the vehicle's perception of the road at the exit. When the vehicle cannot obtain road information at the boundary of the exit, the vehicle's position is controlled based on the traction information, and the vehicle is controlled to travel at a slower speed, so that the vehicle has sufficient time to obtain and process the road information at the boundary of the exit, thereby improving the probability of obtaining clear road information (such as clear road boundaries and / or lane lines). As the vehicle approaches the boundary of the exit, the reliability and accuracy of the road information at the boundary of the exit perceived by the vehicle gradually increase. At this time, controlling the vehicle's position based on the fuzzy road and / or clear lane helps to improve navigation accuracy. Controlling the vehicle to travel at a faster speed based on the fuzzy road and / or clear lane helps to improve the efficiency of the vehicle passing through the intersection.

[0139] It should be noted that the above-mentioned "fuzzy road" and "clear lane" are only used to distinguish the types of perception information. "Fuzzy road" indicates the road boundary perceived by the vehicle, and "clear lane" indicates the lane boundary (such as lane line) perceived by the vehicle. "Fuzzy" and "clear" should not be understood as restrictions on the clarity of the perception results.

[0140] Figure 11 shows another schematic flow chart of the autonomous driving method provided in an embodiment of the present application. This method 700 can be applied to the intelligent driving device shown in Figure 1, or the method can be executed by the system shown in Figure 2. More specifically, this method 700 may include S710 and S720.

[0141] S710, obtaining traction information, the traction information indicating the direction and position of the intelligent driving device exiting the first intersection, the traction information being determined based on SD map navigation information, the SD map navigation information indicating a navigation route from the current position of the intelligent driving device to the target position.

[0142] Exemplarily, the intelligent driving device may include the vehicle in the above embodiment, or it may be other intelligent driving devices; the first intersection may include intersection a in the above embodiment; the traction information may include the position and direction of the exit determined by method 400, or the traction information may also include the traction road determined by method 500; the SD map navigation information may include the SD map navigation route in the above embodiment.

[0143] S720: Control the speed and / or posture of the intelligent driving device when it drives toward or passes through the first intersection based on the traction information.

[0144] In some implementations, second road information within a first range of the intelligent driving device is obtained, the second road information indicating multiple perceived routes passing through a first intersection; a main channel is determined from the multiple perceived routes based on SD map navigation information; and the speed and / or posture of the intelligent driving device when driving toward or through the first intersection is controlled, including: controlling the speed and / or posture of the intelligent driving device when driving along the main channel.

[0145] For example, the first range may include the preset range in method 300. When the first intersection is intersection a, the multiple perceived routes may include: a route connecting road A to road B via intersection a, a route connecting road A to road C via intersection a, and a route connecting road A to road D via intersection a. The specific implementation of determining the multiple perceived routes can be found in the description of method 300 and will not be repeated here.

[0146] In some implementations, controlling the speed and / or posture of the intelligent driving device toward or through the first intersection based on the traction information includes: controlling the intelligent driving device toward or through the first intersection at a first speed based on the traction information. For example, the first speed may include speed 1 in method 600.

[0147] In some implementations, the first boundary of the first intersection is the target exit boundary of the intelligent driving device, and the first boundary is connected to the first road. The method also includes: obtaining information of the first road perceived by the intelligent driving device; controlling the speed and / or posture of the intelligent driving device toward or through the first intersection based on the traction information, including: controlling the speed and / or posture of the intelligent driving device based on the information of the first road and the traction information.

[0148] Exemplarily, when the first intersection is intersection a, the first boundary may be boundary c shown in FIG7 (c), the first road connected to the first boundary is road B, and the information of the first road includes the above-mentioned blurred road and / or clear lane information.

[0149] In some implementations, the information of the first road indicates the boundary of the first road and / or the position of the lane line in the first road; and the speed and / or posture of the intelligent driving device are controlled according to the information of the first road and the traction information, including: when the reliability of the information of the first road is less than or equal to a preset threshold, the intelligent driving device is controlled to travel at a first speed according to the traction information; or, when the reliability of the information of the first road is greater than the preset threshold, the intelligent driving device is controlled to switch from the first speed to the second speed according to the information of the first road; wherein the first speed is less than or equal to the second speed.

[0150] For example, when the first road information indicates the boundary of the first road (e.g., information about a blurred road), the second speed may include speed 2. When the first road information indicates the location of a lane line within the first road (e.g., information about a clear lane), the second speed may include speed 3. For more specific implementations of controlling the speed of the intelligent driving device based on the first road information and traction information, please refer to the description of method 600 and will not be repeated here.

[0151] In some implementations, the traction information includes a traction road, and the method further includes: obtaining boundary information of the first intersection; determining a direction of exiting the first intersection based on SD map navigation information; determining a position of exiting the first intersection based on the boundary information and the direction of exiting the first intersection; and generating a traction road based on the position of exiting the first intersection and the direction of exiting the first intersection.

[0152] For example, the boundary information of the first intersection may include the intersection boundary described in method 400. The specific implementation of determining the direction of exiting the first intersection and the location of exiting the first intersection can be referenced to the description of method 400 and will not be repeated here. Furthermore, the method for generating a traction road can be referenced to the description of method 500 and will not be repeated here.

[0153] In some implementations, generating a traction road based on a position and a direction of exiting a first intersection includes: generating an initial traction road based on the position and the direction of exiting the first intersection; and correcting the initial traction road based on a position of a first road connected to a boundary of the intelligent driving device when exiting the first intersection to obtain the traction road.

[0154] For example, the position of the first road may be indicated by information of the first road. For a method of correcting the initial traction road according to the position of the first road, reference may be made to the description in S502 and will not be repeated here.

[0155] In some implementations, the method further includes: controlling a display device to display a navigation interface, the navigation interface including an image of the traction road and a first icon, the first icon indicating a position of the intelligent driving device relative to the traction road.

[0156] In some implementations, the navigation interface further includes an image of a boundary of the first road. In actual implementation, the image of the boundary of the first road may gradually change, for example, gradually become clearer, as the intelligent driving device approaches the first boundary of the first intersection.

[0157] In some implementations, when the lane line information of the first road is acquired, the navigation interface is controlled to switch from the image of the traction road to the image of the lane line of the first road.

[0158] In some implementations, when the lane line information of the first road is acquired, the navigation interface may also switch from displaying an image of the boundary of the first road to displaying an image of the lane line of the first road.

[0159] The autonomous driving method provided by the embodiments of the present application can improve the autonomous driving capabilities of intelligent driving devices without relying on high-precision maps. Specifically, when the intelligent driving device is in an autonomous driving state and the intelligent driving device is far away from an intersection and cannot obtain road information at the intersection (such as road boundaries or lane lines, etc. used for autonomous driving navigation) through a sensing device, or when the obtained sensing information is not accurate enough, the exit direction of the intersection can be determined in advance based on the traction information, thereby adjusting the heading of the traction vehicle body to avoid the intelligent driving device from getting stuck near the intersection, improving the pass rate and success rate of the intelligent driving device at the intersection in the absence of a high-precision map, and at the same time reducing the probability of the intelligent driving device entering the opposite lane or scratching the road boundary when exiting the intersection. In addition, the intelligent driving device is controlled to travel at a lower speed based on the traction information, so that the intelligent driving device has sufficient time to obtain road information at the first boundary of the first intersection when it approaches the first boundary, thereby increasing the probability of obtaining clear road information (such as clear lane lines), which helps to improve the control capability and driving safety of the intelligent driving device during autonomous driving. As the intelligent driving device approaches the first boundary, the reliability and accuracy of the first road information obtained by the intelligent driving device gradually improves. Controlling the position of the intelligent driving device based on the first road information obtained by the vehicle perception system (such as the first road information) is more reliable than controlling the position of the intelligent driving device based on traction information. This helps prevent the intelligent driving device from running over lane markings or road boundaries while driving on the first road. In addition, controlling the intelligent driving device to travel at a higher speed based on the first road information can also improve the speed and efficiency of the intelligent driving device in passing the first intersection.

[0160] In the various embodiments of the present application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between the various embodiments are consistent and can be referenced by each other. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0161] The autonomous driving method provided by the embodiment of the present application is described in detail above with reference to Figures 1 to 11 . The apparatus provided by the embodiment of the present application will be described in detail below with reference to Figures 12 and 13 . It should be understood that the description of the apparatus embodiment corresponds to the description of the method embodiment. Therefore, for matters not described in detail, reference can be made to the method embodiment above, and for the sake of brevity, no further description will be given here.

[0162] Figure 12 shows a schematic block diagram of an autonomous driving device 2000 provided in an embodiment of the present application. The device 2000 may include units for executing methods 300, 400, 500, 600, and 700. Furthermore, each unit in the device 2000 is configured to implement the corresponding processes of the aforementioned method embodiments. The device 2000 includes an acquisition unit 2010, which can be used to implement corresponding data acquisition or transceiver functions. The device 2000 also includes a processing unit 2020, which can be used to implement corresponding processing functions.

[0163] Optionally, the device 2000 also 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 implements the relevant actions in the aforementioned method embodiments.

[0164] It should be understood that the specific process of each unit executing the above corresponding steps has been described in detail in the above method embodiment, and for the sake of brevity, it will not be repeated here.

[0165] It should also be understood that the apparatus 2000 herein is embodied in the form of functional units. The term "module" or "unit" herein may refer to an application-specific ASIC, electronic circuitry, a processor (e.g., a shared processor, a dedicated processor, or a group of processors, etc.) and memory for executing one or more software or firmware programs, combined logic circuitry, and / or other suitable components that support the described functionality.

[0166] The apparatuses of each of the above-described solutions have the functionality to implement the corresponding steps performed by the computing platform 150 in the above-described methods. These functions can be implemented in hardware, or by hardware executing corresponding software implementations. The hardware or software includes one or more modules corresponding to the above-described functions; 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 to perform the relevant processing operations in each method embodiment.

[0167] Exemplarily, the acquisition unit 2010 and the processing unit 2020 may be provided in the intelligent driving device 100 shown in FIG1 , or may also be provided in the system shown in FIG2 . More specifically, the acquisition unit 2010 and the processing unit 2020 may be provided in the regulation and control module 240 , or may also be provided in the road information generation module 230 . Exemplarily, the operations performed by the acquisition unit 2010 and the processing unit 2020 may be performed by a single processor, or may also be performed by different processors. In a specific implementation, the one or more processors may be provided in the intelligent driving device 100 shown in FIG1 ; alternatively, the apparatus 2000 may be provided in a chip in the intelligent driving device 100 .

[0168] In a specific implementation process, the various units in the above apparatus may be fully or partially integrated together, or may also be implemented independently. In one implementation, these units are integrated together and implemented in the form of a system-on-a-chip (SoC).

[0169] Figure 13 is another schematic block diagram of an autonomous driving device provided in an embodiment of the present application. The autonomous driving device 2100 shown in Figure 13 may include: a processor 2110, a transceiver 2120, and a memory 2130. The processor 2110, the transceiver 2120, and the memory 2130 are connected via an internal connection path. The memory 2130 is used to store instructions, and the processor 2110 is used to execute the instructions stored in the memory 2130 to implement the methods in the above embodiments. Optionally, the memory 2130 can be coupled to the processor 2110 through an interface or integrated with the processor 2110.

[0170] It should be noted that the transceiver 2120 may include but is not limited to a transceiver device such as an input / output interface to implement communication between the device 2100 and other devices or a communication network.

[0171] Memory 2130 may be a volatile memory and / or a non-volatile memory. The non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM). For example, RAM may be used as an external cache. By way of example and not limitation, RAM includes the following forms: 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 link dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0172] The transceiver 2120 uses a transceiver device such as but not limited to a transceiver to implement communication between the device 2100 and other devices or communication networks to receive / send data / information used to implement the methods in the above embodiments.

[0173] An embodiment of the present application also provides an intelligent driving device, which includes the automatic driving device 2000 or the automatic driving device 2100 in the above embodiment.

[0174] An embodiment of the present application further provides a computer program product, which includes computer program code. When the computer program code runs on a computer, the computer implements the methods in the above embodiments of the present application.

[0175] An embodiment of the present application further provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed on a computer, the computer implements the methods in the above embodiments of the present application.

[0176] An embodiment of the present application also provides a chip, including a circuit, for executing the methods in the above embodiments of the present application.

[0177] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0178] In the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in this article is a kind of association relationship that describes associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In this application, "at least one" refers to one or more, and "more than one" refers to two or more. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0179] In the embodiments of this application, prefixes such as "first" and "second" are used only to distinguish different description objects and have no limiting effect on the position, order, priority, quantity, or content of the described objects. The use of prefixes such as ordinal numbers in the embodiments of this application to distinguish description objects does not constitute a limitation on the described objects. For a statement of the described objects, please refer to the description in the context of the claims or embodiments, and the use of such prefixes should not constitute an unnecessary limitation.

[0180] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0181] In the various embodiments of the present application, unless otherwise specified or there is a logical conflict, the terms and / or descriptions between the various embodiments are consistent and can be referenced by each other. The technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0182] The units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0183] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0184] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An automatic driving method, characterized in that: include: Obtaining traction information, the traction information indicating a direction and position of the intelligent driving device exiting the first intersection, the traction information being determined based on standard SD map navigation information indicating a navigation route from a current position of the intelligent driving device to a target position; According to the traction information, the speed and / or posture of the intelligent driving device when driving towards the first intersection or passing through the first intersection is controlled.

2. The method according to claim 1, characterized in that The controlling, based on the traction information, the speed and / or posture of the intelligent driving device driving toward or passing through the first intersection includes: According to the traction information, the intelligent driving device is controlled to drive toward or pass through the first intersection at a first speed, where the first speed is less than or equal to a speed threshold.

3. The method according to claim 1 or 2, characterized in that The first boundary of the first intersection is a target exit boundary of the intelligent driving device, and the first boundary is connected to the first road. The method further includes: Acquiring information about the first road perceived by the intelligent driving device; The controlling, based on the traction information, the speed and / or posture of the intelligent driving device driving toward or passing through the first intersection includes: The speed and / or posture of the intelligent driving device is controlled according to the information of the first road and the traction information.

4. The method according to claim 3, characterized in that The information of the first road indicates a boundary of the first road and / or a position of a lane line in the first road; The controlling the speed and / or posture of the intelligent driving device according to the first road information and the traction information includes: When the reliability of the information of the first road is less than or equal to a preset threshold, controlling the intelligent driving device to travel at a first speed according to the traction information; or When the reliability of the information about the first road is greater than the preset threshold, controlling the intelligent driving device to switch from the first speed to a second speed according to the information about the first road; Wherein, the first speed is less than or equal to the second speed.

5. The method according to any one of claims 1 to 4, characterized in that The traction information includes a traction road, and obtaining the traction information includes: Obtaining boundary information of the first intersection; determining a direction to exit the first intersection according to the SD map navigation information; determining a position for exiting the first intersection based on the boundary information and the direction of exiting the first intersection; The traction road is generated according to the position of exiting the first intersection and the direction of exiting the first intersection.

6. The method according to claim 5, characterized in that Generating the traction road according to the exit position and the exit direction of the first intersection includes: generating an initial traction road according to the exit position and the exit direction of the first intersection; The initial traction road is corrected according to a position of a first road connected to a boundary where the intelligent driving device exits the first intersection to obtain the traction road.

7. The method according to claim 5 or 6, characterized in that The method further comprises: The display device is controlled to display a navigation interface, where the navigation interface includes an image of the towing road and a first icon, where the first icon indicates a position of the intelligent driving device relative to the towing road.

8. The method according to claim 7, characterized in that The navigation interface further includes an image of a boundary of a first road, where the first road is a road connected to a target exit boundary of the intelligent driving device at the first intersection.

9. The method according to claim 7 or 8, characterized in that The method further comprises: When the lane line information of the first road is obtained, the navigation interface is controlled to switch from the image of the traction road to the image of the lane line of the first road, where the first road is connected to the target exit boundary of the intelligent driving device in the first intersection.

10. The method according to any one of claims 1 to 9, characterized in that The method further comprises: obtaining second road information within the first range of the intelligent driving device, where the second road information indicates a plurality of sensed routes passing through the first intersection; determining a main course from the plurality of perceived routes according to the SD map navigation information; The controlling the speed and / or posture of the intelligent driving device toward or through the first intersection includes: Control the speed and / or posture of the intelligent driving device when traveling along the main channel.

11. An automatic driving device, characterized in that: include: The acquiring unit is configured to acquire traction information, the traction information indicating a direction and a position of the intelligent driving device exiting the first intersection, the traction information being determined based on standard SD map navigation information, the SD map navigation information indicating a navigation route from a current position of the intelligent driving device to a target position; The processing unit is used to: control the speed and / or posture of the intelligent driving device when it drives towards the first intersection or passes through the first intersection according to the traction information.

12. The device according to claim 11, characterized in that The processing unit is used for: According to the traction information, the intelligent driving device is controlled to travel at a first speed, where the first speed is less than or equal to a speed threshold.

13. The device according to claim 11 or 12, characterized in that The first boundary of the first intersection is a target exit boundary of the intelligent driving device, and the first boundary is connected to the first road. The acquisition unit is further configured to: Acquiring information about the first road perceived by the intelligent driving device; The processing unit is used for: The speed and / or posture of the intelligent driving device is controlled according to the information of the first road and the traction information.

14. The device according to claim 13, characterized in that The information of the first road indicates a boundary of the first road and / or a position of a lane line in the first road; The processing unit is used for: When the reliability of the information of the first road is less than or equal to a preset threshold, controlling the intelligent driving device to travel at a first speed according to the traction information; or When the reliability of the information about the first road is greater than the preset threshold, controlling the intelligent driving device to switch from the first speed to a second speed according to the information about the first road; Wherein, the first speed is less than or equal to the second speed.

15. The device according to any one of claims 11 to 14, characterized in that The traction information includes a traction road, and the acquiring unit is further configured to: Obtaining boundary information of the first intersection; The processing unit is further configured to: determining a direction to exit the first intersection according to the SD map navigation information; determining a position for exiting the first intersection based on the boundary information and the direction of exiting the first intersection; The traction road is generated according to the position of exiting the first intersection and the direction of exiting the first intersection.

16. The device according to claim 15, characterized in that The processing unit is used for: generating an initial traction road according to the exit position and the exit direction of the first intersection; The initial traction road is corrected according to a position of a first road connected to a boundary where the intelligent driving device exits the first intersection to obtain the traction road.

17. The device according to claim 15 or 16, characterized in that The processing unit is further configured to: The display device is controlled to display a navigation interface, where the navigation interface includes an image of the towing road and a first icon, where the first icon indicates a position of the intelligent driving device relative to the towing road.

18. The device according to claim 17, characterized in that The navigation interface further includes an image of a boundary of a first road, where the first road is a road connected to a target exit boundary of the intelligent driving device at the first intersection.

19. The device according to claim 17 or 18, characterized in that The processing unit is further configured to: When the lane line information of the first road is obtained, the navigation interface is controlled to switch from the image of the traction road to the image of the lane line of the first road, where the first road is the road connected to the target exit boundary of the intelligent driving device in the first intersection.

20. The device according to any one of claims 11 to 19, characterized in that The acquisition unit is further configured to: obtaining second road information within the first range of the intelligent driving device, where the second road information indicates a plurality of sensed routes passing through the first intersection; The processing unit is further configured to: determining a main course from the plurality of perceived routes according to the SD map navigation information; Control the speed and / or posture of the intelligent driving device when traveling along the main channel.

21. An automatic driving device, characterized in that: include: A processor, configured to execute a computer program stored in a memory, so that the apparatus performs the method according to any one of claims 1 to 10.

22. The device according to claim 21, characterized in that The apparatus further comprises the memory.

23. An intelligent driving device, characterized in that: Comprising the apparatus of any one of claims 11 to 22.

24. A computer-readable storage medium, characterized in that Instructions are stored thereon, and when the instructions are executed by a processor, the method according to any one of claims 1 to 10 is implemented.

25. A computer program product, characterized in that The computer program product comprises: a computer program code, and when the computer program code is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

26. A chip, characterized in that: The chip comprises a circuit for executing the method according to any one of claims 1 to 10.

Images