Autonomous driving method and apparatus, and intelligent driving device

Abstract

An autonomous driving method and apparatus, and an intelligent driving device are provided. The method includes the following steps. S710: Obtain guidance information, where the guidance information indicates a direction and a location for an intelligent driving device to exit a first intersection, the guidance information is determined based on SD map navigation information, and the SD map navigation information indicates a navigation path from a current location of the intelligent driving device to a target location. S720: Control, based on the guidance information, a speed and / or a pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection. According to the method, when a perception system of a vehicle cannot detect a road boundary and / or a clear lane line at an exit of the intersection, the vehicle may be controlled, based on the guidance information, to travel at a low speed. This improves a capability and efficiency of the vehicle in passing through a complex topology scenario such as an intersection or a ramp without relying on a high definition map, thereby enhancing an autonomous driving capability of the vehicle.

Description

AUTONOMOUS DRIVING METHOD AND APPARATUS, AND INTELLIGENT DRIVING DEVICE

[0001] This application claims priority to Chinese Patent Application No. 202410345295.X, filed with the China National Intellectual Property Administration on March 22, 2024 and entitled "AUTONOMOUS DRIVING METHOD AND APPARATUS, AND INTELLIGENT DRIVING DEVICE", which is incorporated herein by reference in its entirety.TECHNICAL FIELD

[0002] This application relates to the field of intelligent driving, and more specifically, to an autonomous driving method and apparatus, and an intelligent driving device.BACKGROUND

[0003] As the automobile industry rapidly develops, many assisted driving technologies and autonomous driving technologies have emerged, which can reduce driving pressure and improve safety and travel efficiency. Currently, most autonomous driving technologies rely on high definition maps for navigation. However, high definition maps have disadvantages such as high collection and production costs, long processing time, limited coverage, and difficulty in maintaining data freshness. As a result, it is challenging to deploy the autonomous driving technologies that rely on high definition maps on a nationwide or global scale.

[0004] In view of this, there is an urgent need to develop an autonomous driving solution independent of high definition maps.SUMMARY

[0005] This application provides an autonomous driving method and apparatus, and an intelligent driving device, to improve a capability and efficiency of an intelligent driving device in passing through a complex topology scenario such as an intersection or a ramp without relying on a high definition map.

[0006] According to a first aspect, an autonomous driving method is provided. The method may be performed by an intelligent driving device, for example, may be performed by a computing platform of the intelligent driving device, or may be performed by a chip or a circuit used for the intelligent driving device. Alternatively, the method may be performed by a cloud server associated with the intelligent driving device. This is not specifically limited in this application.

[0007] The method includes: obtaining guidance information, where the guidance information indicates a direction and a location for an intelligent driving device to exit a first intersection, the guidance information is determined based on standard definition (standard definition, SD) map navigation information, and the SD map navigation information indicates a navigation path from a current location of the intelligent driving device to a target location; and controlling, based on the guidance information, a speed and / or a pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection.

[0008] In some implementations, the first intersection may include at least two boundaries, for example, a boundary on a side on which the intelligent driving device travels into and a boundary on a side on which the intelligent driving device travels out. Alternatively, the first intersection further includes another boundary in addition to the two boundaries. For example, the first intersection may be an n-way 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 is connected to one road.

[0009] In some implementations, the guidance information may include a guidance road, namely, a road boundary connected to a boundary at which the intelligent driving device exits the first intersection. The guidance road may include at least one lane, and the guidance road may be a unidirectional driving road.

[0010] In the foregoing technical solution, when the intelligent driving device is in an autonomous driving state, and the intelligent driving device is too far from the intersection to obtain intersection-related perception information (for example, a road boundary, a lane line, or other information used for autonomous driving navigation) by using a perception apparatus, or when precision of obtained perception information is insufficient, a direction and a location for exiting the intersection can be determined in advance based on the guidance information, so as to guide heading adjustment of a vehicle body and prevent the intelligent driving device from being stuck near the intersection. This improves a pass rate and a success rate of the intelligent driving device at the intersection without a high definition map, and reduces a likelihood of the intelligent driving device entering an oncoming lane or scraping against a road boundary when exiting the intersection. In addition, adjusting a speed of the intelligent driving device based on the guidance information helps improve driving safety of the intelligent driving device.

[0011] With reference to the first aspect, in some implementations of the first aspect, controlling, based on the guidance information, the speed and / or the pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection includes: controlling, based on the guidance information, the intelligent driving device to travel toward the first intersection or pass through the first intersection at a first speed, 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 a maximum speed limit of the first intersection, or may be another speed determined based on the maximum speed limit of the first intersection. The maximum speed limit of the first intersection may be a maximum speed limit of a road connected to the first intersection, for example, a maximum speed limit of a current traveling road (for example, a road entering the first intersection) of the intelligent driving device, or may be a maximum speed limit of a target traveling road (for example, a road departing from the first intersection) of the intelligent driving device.

[0013] In the foregoing technical solution, the intelligent driving device is controlled, based on the guidance information, to travel at a low speed, so that when the intelligent driving device travels toward a specific boundary (namely, the boundary at which the intelligent driving device exits the first intersection), there is sufficient time to obtain road information at the specific boundary, to increase a probability of obtaining clear road information (for example, a clear lane line). This improves a planning and control capability of the intelligent driving device in an autonomous driving process.

[0014] With reference to the first aspect, in some implementations of the first aspect, a first boundary of the first intersection is a target travel-out boundary of the intelligent driving device, the first boundary is connected to a first road, and the method further includes: obtaining information that is about the first road and that is perceived by the intelligent driving device; and controlling, based on the guidance information, the speed and / or the pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection includes: controlling the speed and / or the pose of the intelligent driving device based on the first road information and the guidance information.

[0015] The information that is about the first road and that is perceived by the intelligent driving device may be understood as information that is about the first road and that is obtained by a perception system (such as a camera apparatus or a radar) of the intelligent driving device.

[0016] It may be understood that when the first boundary is far away from the intelligent driving device and beyond a perception range of the intelligent driving device, the intelligent driving device cannot obtain the first road information, or reliability and / or precision of the first road information obtained by the intelligent driving device are insufficient. As the intelligent driving device travels toward the first boundary, reliability and precision of the first road information obtained by the intelligent driving device gradually increase. Because the guidance information is only information that is generated based on the SD map navigation information and that indicates the location and the direction for exiting the first intersection, and an error may exist between the guidance information and an actual location and a boundary of the first road, when the intelligent driving device is close enough to the first boundary, controlling a pose of the intelligent driving device based on the first road information is more reliable than controlling a pose of the intelligent driving device based on the guidance information. "More reliable" in the foregoing may be understood as being able to prevent the intelligent driving device from traveling on a lane line or a road boundary line, or the like. Therefore, when navigation is performed only based on an SD map, the speed and / or the pose of the intelligent driving device is controlled based on the first road information and the guidance road in a process in which the intelligent driving device travels toward the first intersection and passes through the first intersection. This helps improve a pass rate of the intelligent driving device passing through a complex intersection and robustness of an autonomous driving system of the intelligent driving device.

[0017] With reference to the first aspect, in some implementations of the first aspect, the first road information indicates a boundary of the first road and / or a location of a lane line on the first road; and controlling the speed and / or the pose of the intelligent driving device based on the first road information and the guidance information includes: when reliability of the first road information is less than or equal to a preset threshold, control, based on the guidance information, the intelligent driving device to travel at the first speed; or when the reliability of the first road information is greater than the preset threshold, control, based on the first road information, the intelligent driving device to switch from traveling at the first speed to traveling at a second speed, where the first speed is less than or equal to the second speed.

[0018] In the foregoing technical solution, when the reliability of the first road information is greater than the preset threshold, the intelligent driving device is controlled to travel at a high speed, so that a speed and efficiency of the intelligent driving device in passing through the first intersection can be improved.

[0019] With reference to the first aspect, in some implementations of the first aspect, the guidance information includes a guidance road, and obtaining the guidance information includes: obtaining information about a boundary of the first intersection; determining, based on the SD map navigation information, the direction for exiting the first intersection; determining, based on the information about the boundary and the direction for exiting the first intersection, the location for exiting the first intersection; and generating the guidance road based on the location for exiting the first intersection and the direction for exiting the first intersection.

[0020] It should be noted that a guidance lane and the guidance road are not strictly distinguished in this application. When information about a quantity of lanes of a road (for example, the first road) connected to the location at which the intelligent driving device exits the first intersection is not obtained, the guidance road generated based on the location of the first intersection and the direction for the intelligent driving device to exit the first intersection may include only one guidance lane. When information that the quantity of lanes is n is obtained, the guidance road generated based on the location of the first intersection and the direction for the intelligent driving device to exit the first intersection may include n guidance lanes.

[0021] In the foregoing technical solution, for a road beyond the perception range of the intelligent driving device, a guidance road is generated based on SD map navigation information, so that the intelligent driving device can adjust and move toward a target traveling direction in advance, to reduce a probability that the intelligent driving device drives into an oncoming lane or deviates from a road. This helps improve an autonomous driving capability and driving safety of the intelligent driving device in a scenario without a high definition map.

[0022] With reference to the first aspect, in some implementations of the first aspect, generating the guidance road based on the location for exiting the first intersection and the direction for exiting the first intersection includes: generating an initial guidance road based on the location for exiting the first intersection and the direction for exiting the first intersection; and correcting the initial guidance road based on a location of the first road connected to the boundary at which the intelligent driving device exits the first intersection, to obtain the guidance road.

[0023] For example, the location of the first road may be perceived by the perception system of the intelligent driving device.

[0024] With reference to the first aspect, in some implementations of the first aspect, the method further includes: controlling a display apparatus to display a navigation interface, where the navigation interface includes an image of the guidance road and a first icon, and the first icon indicates a location of the intelligent driving device relative to the guidance road.

[0025] For example, when the speed and / or the pose of the intelligent driving device are / is controlled based on the guidance information, the navigation interface is controlled to display the image of the guidance road. The first icon may be an icon representing the intelligent driving device, and the first icon and the image of the guidance road indicate a relative location relationship between the intelligent driving device and the guidance road.

[0026] In the foregoing technical solution, the image of the guidance road is displayed, so that a user of the intelligent driving device can determine a data source currently used for navigation. A driver of the intelligent driving device may be further warned to take over the intelligent driving device when necessary.

[0027] With reference to the first aspect, in some implementations of the first aspect, the navigation interface further includes an image of the boundary of the first road.

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

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

[0030] In some other implementations, when the image of the guidance road indicates the location of the lane line on the first road, the navigation interface may be controlled to display the image of the guidance road and the image of the boundary of the first road in a superimposing manner.

[0031] In the foregoing technical solution, the image of the boundary of the first road is displayed, so that the user can clearly determine a result of perceiving the current traveling road by the intelligent driving device and a difference between the guidance road and the actual first road, so that the driver takes over the intelligent driving device when necessary.

[0032] With reference to the first aspect, in some implementations of the first aspect, the method further includes: when information about the lane line of the first road is obtained, controlling the navigation interface to switch from the image of the guidance road to an image of the lane line of the first road.

[0033] For example, when the speed and / or the pose of the intelligent driving device are / is controlled based on the lane line of the first road, the navigation interface is controlled to switch from displaying the image of the guidance road to displaying the image of the lane line of the first road.

[0034] "Switching" of content of the navigation interface in this application may be understood as a result of refreshing the navigation interface in real time.

[0035] In the foregoing technical solution, the image is switched, so that the user can determine improvement of a perception capability of the intelligent driving device on the first road. This improves confidence in an autonomous driving function of the intelligent driving device.

[0036] With reference to the first aspect, in some implementations of the first aspect, the method further includes: obtaining second road information within a first range of the intelligent driving device, where the second road information indicates a plurality of perceived paths passing through the first intersection; and determining a primary navigation path from the plurality of perceived paths based on the SD map navigation information; and controlling the speed and / or the pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection includes: controlling a speed and / or a pose at which the intelligent driving device travels along the primary navigation path.

[0037] In the foregoing technical solution, when the intelligent driving device is controlled to travel toward the first intersection and pass through the first intersection, the intelligent driving device needs to be controlled to travel along the primary navigation path, to avoid deviation of the intelligent driving device from the navigation path and improve robustness of an autonomous driving function of the intelligent driving device.

[0038] According to a second aspect, an autonomous driving apparatus is provided. The apparatus includes an obtaining unit and a processing unit, where the obtaining unit is configured to: obtain guidance information, where the guidance information indicates a direction and a location for an intelligent driving device to exit a first intersection, the guidance information is determined based on SD map navigation information, and the SD map navigation information indicates a navigation path from a current location of the intelligent driving device to a target location; and the processing unit is configured to: control, based on the guidance information, a speed and / or a pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection.

[0039] With reference to the second aspect, in some implementations of the second aspect, the processing unit is configured to: control, based on the guidance information, the intelligent driving device to travel at a first speed, where the first speed is less than or equal to a speed threshold.

[0040] With reference to the second aspect, in some implementations of the second aspect, a first boundary of the first intersection is a target travel-out boundary of the intelligent driving device, the first boundary is connected to a first road, and the obtaining unit is further configured to: obtain information that is about the first road and that is perceived by the intelligent driving device; and the processing unit is configured to: control the speed and / or the pose of the intelligent driving device based on the first road information and the guidance information.

[0041] With reference to the second aspect, in some implementations of the second aspect, the first road information indicates a boundary of the first road and / or a location of a lane line on the first road; and the processing unit is configured to: when reliability of the first road information is less than or equal to a preset threshold, control, based on the guidance information, the intelligent driving device to travel at the first speed; or when the reliability of the first road information is greater than the preset threshold, control, based on the first road information, the intelligent driving device to switch from traveling at the first speed to traveling at a second speed, where the first speed is less than or equal to the second speed.

[0042] With reference to the second aspect, in some implementations of the second aspect, the guidance information includes a guidance road, and the obtaining unit is further configured to: obtain information about a boundary of the first intersection; and the processing unit is further configured to: determine, based on the SD map navigation information, the direction for exiting the first intersection; determine, based on the information about the boundary and the direction for exiting the first intersection, the location for exiting the first intersection; and generate the guidance road based on the location for exiting the first intersection and the direction for exiting the first intersection.

[0043] With reference to the second aspect, in some implementations of the second aspect, the processing unit is configured to: generate an initial guidance road based on the location for exiting the first intersection and the direction for exiting the first intersection; and correct the initial guidance road based on a location of the first road connected to the boundary at which the intelligent driving device exits the first intersection, to obtain the guidance road.

[0044] With reference to the second aspect, in some implementations of the second aspect, the processing unit is further configured to: control a display apparatus to display a navigation interface, where the navigation interface includes an image of the guidance road and a first icon, and the first icon indicates a location of the intelligent driving device relative to the guidance road.

[0045] With reference to the second aspect, in some implementations of the second aspect, the navigation interface further includes an image of the boundary of the first road.

[0046] With reference to the second aspect, in some implementations of the second aspect, the processing unit is further configured to: when information about the lane line of the first road is obtained, control the navigation interface to switch from the image of the guidance road to an image of the lane line of the first road.

[0047] With reference to the second aspect, in some implementations of the second aspect, the obtaining unit is further configured to: obtain second road information within a first range of the intelligent driving device, where the second road information indicates a plurality of perceived paths passing through the first intersection; and the processing unit is further configured to: determine a primary navigation path from the plurality of perceived paths based on the SD map navigation information; and control a speed and / or a pose at which the intelligent driving device travels along the primary navigation path.

[0048] According to a third aspect, an autonomous driving apparatus is provided. The apparatus includes a processor, configured to execute a computer program stored in a memory, to enable the apparatus to perform the method according to any possible implementation of the first aspect.

[0049] With reference to the third aspect, in some implementations of the third aspect, the autonomous driving apparatus further includes the memory.

[0050] According to a fourth aspect, an intelligent driving device is provided, and the intelligent driving device includes the apparatus according to any possible implementation of the second aspect and the third aspect.

[0051] With reference to the fourth aspect, in some implementations of the fourth aspect, the intelligent driving device is a vehicle.

[0052] According to a fifth aspect, a computer program product is provided. The computer program product includes computer program code. When the computer program code is run on a computer or a processor, the computer or the processor is enabled to perform the method according to any possible implementation of the first aspect.

[0053] It should be noted that the computer program code may be completely or partially stored in a storage medium. The storage medium may be encapsulated with a processor, or encapsulated separately from a processor.

[0054] According to a sixth aspect, a computer-readable medium is provided. The computer-readable medium stores instructions, and when the instructions are executed by a processor, the processor is enabled to implement the method according to any possible implementation of the first aspect.

[0055] According to a seventh aspect, a chip is provided. The chip includes a circuit, and the circuit is configured to perform the method according to any possible implementation of the first aspect.BRIEF DESCRIPTION OF DRAWINGS

[0056] FIG. 1 is a functional block diagram of an intelligent driving device according to an embodiment of this application;

[0057] FIG. 2 is a diagram of an architecture of an autonomous driving system according to an embodiment of this application;

[0058] FIG. 3 is a schematic flowchart of an autonomous driving method according to an embodiment of this application;

[0059] FIG. 4A and FIG. 4B are diagrams of an application scenario of an autonomous driving method according to an embodiment of this application;

[0060] FIG. 5A to FIG. 5C are diagrams of a matching result between a perceived path and an SD map navigation path according to an embodiment of this application;

[0061] FIG. 6 is another schematic flowchart of an autonomous driving method according to an embodiment of this application;

[0062] FIG. 7A to FIG. 7C are diagrams of a result of determining a location for exiting an intersection according to an embodiment of this application;

[0063] FIG. 8 is still another schematic flowchart of an autonomous driving method according to an embodiment of this application;

[0064] FIG. 9A and FIG. 9B are diagrams of a guidance road according to an embodiment of this application;

[0065] FIG. 10 is still another schematic flowchart of an autonomous driving method according to an embodiment of this application;

[0066] FIG. 11 is still another schematic flowchart of an autonomous driving method according to an embodiment of this application;

[0067] FIG. 12 is a block diagram of an autonomous driving apparatus according to an embodiment of this application; and

[0068] FIG. 13 is another block diagram of an autonomous driving apparatus according to an embodiment of this application.DESCRIPTION OF EMBODIMENTS

[0069] Before the solutions of this application are described, related concepts in this application are first described.

[0070] 1. SD map: Precision is generally at a meter level and richness is low. Generally, the SD map mainly includes road information and point of interest (point of information or point of interest, POI). The POI is point data in an electronic map, and includes at least four attributes: a name, an address, coordinates, and a category.

[0071] 2. High definition (high definition, HD) map: Precision and richness of the HD map are higher than those of an SD map. Absolute precision and relative precision of the HD map are both at a centimeter level. In terms of richness, the HD map provides an environment model of an autonomous driving vehicle, including a static high definition map and other dynamic information. The static high definition map includes a lane model, a road component, a road attribute, and the like. The lane model includes road details, such as a lane line, a lane center line, and a lane attribute change. The other dynamic information includes all dynamic information in an intelligent network system, such as map dynamic information, sensor information, driving behavior, and traffic dynamic information control.

[0072] As described above, currently, most autonomous driving technologies rely on high definition maps for navigation. However, the high definition map has disadvantages such as high collection and production costs, long consumed time, insufficient coverage, and difficulty in ensuring data freshness. As a result, it is difficult to promote the autonomous driving technologies that rely on the high definition maps nationwide or globally. If navigation is performed only based on environment information perceived by a vehicle-end sensor without relying on a high definition map, because a perception range of the vehicle-end sensor is limited, information beyond the perception range cannot be obtained, and consequently, an autonomous driving capability of the vehicle is limited. For example, when a vehicle needs to pass through a large intersection or a specially-shaped intersection, or needs to turn left or right at an intersection, the vehicle cannot obtain information such as a lane location and a road boundary at an exit of the intersection, and consequently, an autonomous driving planning capability of the vehicle is affected. For example, the vehicle cannot determine an offset direction and a traveling direction, and consequently, travel efficiency of the vehicle in the foregoing scenario is low.

[0073] In view of this, embodiments of this application provide an autonomous driving method and apparatus, and an intelligent driving device. When a road ahead of an intelligent driving device includes an intersection, guidance information indicating a direction for the intelligent driving device to exit the intersection is generated based on information about the road ahead (including information about an intersection boundary) obtained by the intelligent driving device and a navigation path generated from an SD map. The guidance information indicates the direction and a location for the intelligent driving device to exit the intersection. The intelligent driving device controls, based on the guidance information, a pose and a speed at which the intelligent driving device passes through the intersection. When a perception system of the intelligent driving device fails to detect a road boundary and / or a clear lane line at the exit of the intersection, the intelligent driving device may be controlled, based on the guidance information, to drive at a low speed. As the intelligent driving device approaches the exit of the intersection, when the perception system of the intelligent driving device can detect the road boundary and / or the clear lane line at the exit of the intersection, the intelligent driving device may be controlled, based on the road boundary and / or the clear lane line, to travel at a high speed, to improve travel efficiency of the intelligent driving device in passing through the intersection without a high definition map, thereby enhancing an autonomous driving capability of the intelligent driving device.

[0074] The following describes technical solutions of this application with reference to accompanying drawings.

[0075] FIG. 1 is a functional block diagram of an intelligent driving device according to an embodiment of this application. As shown in FIG. 1, the intelligent driving device 100 may include a perception system 120, a display apparatus 130, and a computing platform 150. The perception system 120 may include various sensors configured to perceive surrounding environment information of the intelligent driving device 100. For example, the perception system 120 may include a positioning system. The positioning system may be a global positioning system (global positioning system, GPS), a BeiDou system, or another positioning system. For another example, the perception system 120 may further include one or more of an inertial measurement unit (inertial measurement unit, IMU), a lidar, a millimeter-wave radar, an ultrasonic radar, and a camera apparatus.

[0076] The display apparatus 130 is mainly classified into two types: a first type is a vehicle-mounted display, and a second type is a projection display, for example, a head-up display (head-up display, HUD) apparatus. The vehicle-mounted display is a physical display, and is an important part of an in-vehicle infotainment system. The vehicle-mounted display may include a human-machine interface (human-machine interface, HMI). The head-up display, also referred to as a head-up display system, is mainly configured to display driving information such as a speed and navigation on a display device (for example, a windshield) in front of a user, to reduce sight line diversion time of the user, avoid a pupil change caused by sight line diversion of the user, and improve traveling safety and comfort.

[0077] Some or all of functions of the intelligent driving device 100 may be controlled by the computing platform 150. The computing platform 150 may include processors 151 to 15n. The processor is a circuit having a signal processing capability. In an implementation, the processor may be a circuit having an instruction reading and running capability, for example, a central processing unit (central processing unit, CPU), a microprocessor, a graphics processing unit (graphics processing unit, GPU) (which may be understood as a microprocessor), or a digital signal processor (digital signal processor, DSP). In another implementation, the processor may implement a specific function based on a logical relationship of a hardware circuit. The logical relationship of the hardware circuit is fixed or reconfigurable. For example, the processor is a hardware circuit implemented by an application-specific integrated circuit (application-specific integrated circuit, ASIC) or a programmable logic device (programmable logic device, PLD), for example, a field programmable gate array (field programmable gate array, FPGA). In a reconfigurable hardware circuit, a process in which the processor loads a configuration document to implement hardware circuit configuration may be understood as a process in which the processor loads instructions to implement functions of some or all of the units. In addition, the processor may alternatively be a hardware circuit designed for artificial intelligence, and may be understood as an ASIC, for example, a neural network processing unit (neural network processing unit, NPU), a tensor processing unit (tensor processing unit, TPU), or a deep learning processing unit (deep learning processing unit, DPU). In addition, the computing platform 150 may further include a memory. The memory is configured to store instructions. Some or all of the processors 151 to 15n may invoke the instructions in the memory, to implement a corresponding function.

[0078] The intelligent driving device 100 may include an ADAS. The ADAS obtains information from surroundings of the intelligent driving device by using a plurality of sensors (including but not limited to a lidar, a millimeter-wave radar, a camera apparatus, an ultrasonic sensor, a global positioning system, and an inertial measurement unit) on the intelligent driving device, and analyzes and processes the obtained information, to implement functions such as obstacle perception, target recognition, intelligent driving device positioning, path planning, and driver monitoring / reminder, so that driving safety, automation, and comfort of the intelligent driving device are improved.

[0079] In terms of logical functions, the ADAS system usually includes three main functional modules: a perception module, a decision-making module, and an execution module. The perception module perceives an environment around a vehicle body through a sensor, and inputs corresponding real-time data to a processing center at a decision-making layer. The perception module mainly includes a vehicle-mounted camera, an ultrasonic radar, a millimeter-wave radar, a lidar, or the like. The decision-making module uses a computing apparatus and an algorithm to make a corresponding decision based on information obtained by the perception module. After receiving a decision-making signal from the decision-making module, the execution module takes a corresponding action, for example, driving, a lane change, steering, braking, or warning.

[0080] At different autonomous driving levels (L0 to L5), the ADAS may implement different levels of autonomous driving assistance based on information obtained by using an artificial intelligence algorithm and a plurality of sensors. The foregoing autonomous driving levels (L0 to L5) are based on a grading standard of the society of automotive engineers (society of automotive engineers, SAE). The level L0 indicates no automation, the level L1 indicates driving assistance, the level L2 indicates partial automation, the level L3 indicates conditional automation, the level L4 indicates high automation, and the level L5 indicates full automation. Tasks of monitoring and responding to road conditions at the levels L1 to L3 are jointly completed by a driver and a system, and the driver needs to take over a dynamic driving task. At the levels L4 and L5, the driver may be completely transformed into a role of a passenger. Currently, functions that may be implemented by the ADAS mainly include but are not limited to: adaptive cruising, automatic emergency braking, automatic parking, blind spot monitoring, traffic warning / braking at a front crossroad, traffic warning / braking at a rear crossroad, front vehicle collision warning, lane deviation warning, lane keeping assistance, rear vehicle collision warning, traffic sign recognition, traffic jam assistance, highway assistance, and the like. It should be understood that the foregoing functions may have specific modes at different autonomous driving levels (L0 to L5). A higher autonomous driving level corresponds to a more intelligent mode.

[0081] In this embodiment of this application, the computing platform 150 may generate guidance information based on an SD map and information about an environment around the intelligent driving device obtained by the perception system 120, or the computing platform 150 may further control the display apparatus 130 to display the guidance information.

[0082] FIG. 2 is a diagram of an architecture of an autonomous driving system according to an embodiment of this application. As shown in FIG. 2, the system includes a perception module 210, a map information obtaining module 220, a road information generation module 230, a planning and control module 240, and a display module 250. The perception module 210 may include one or more camera apparatuses in the perception system 120 shown in FIG. 1, or may further include one or more radars in the perception system 120. The map information obtaining module 220, the road information generation module 230, and the planning and control module 240 may separately include one or more processors in the computing platform 150 shown in FIG. 1. The display module 250 may include one or more of the display apparatuses 130, for example, the HMI.

[0083] The perception module 210 is configured to obtain information about an environment around an intelligent driving device, and send the environment information to the road information generation module 230. The environment information includes road information, for example, a road boundary and a lane line of a current traveling road of the intelligent driving device. For example, the perception module 210 may further include one or more processors, to process collected perception information to obtain environment information. For example, the perception information is an image. The one or more processors in the perception module 210 may extract environment information such as a road boundary, an intersection boundary, and a lane line from the obtained image.

[0084] The map information obtaining module 220 is configured to determine a navigation path between a current location and a target location of the intelligent driving device based on an SD map, and then send the navigation path to the road information generation module 230.

[0085] The road information generation module 230 includes a primary navigation path determining module 231, a determining module 232 for information for exiting an intersection, and a guidance information generation module 233. The primary navigation path determining module is configured to determine a primary navigation path from a plurality of perceived paths based on a matching degree between the navigation path and the plurality of perceived paths detected by the perception module 210. The determining module 232 for information for exiting an intersection is configured to determine, based on an intersection boundary of an intersection that is passed through by the primary navigation path detected by the perception module 210, a location and a direction for exiting the intersection. The guidance information generation module 233 is configured to generate a guidance road based on the location and the direction for exiting the intersection, where the guidance road is a generated virtual road. When the perception module 210 determines, based on the perception information, a boundary of an actual road corresponding to the guidance road, the guidance information generation module 233 may further correct a location of the guidance road based on the boundary of the actual road. Further, the road information generation module 230 sends the guidance road to the planning and control module 240.

[0086] The planning and control module 240 may further obtain the boundary, determined by the perception module 210, of the actual road corresponding to the guidance road, and a clear lane line, collected by the perception module 210, of the actual road corresponding to the guidance road. Further, the planning and control module 240 may control a speed of the intelligent driving device based on a perception result of the perception module 210. The planning and control module 240 may further control the display module 250 to display different navigation elements, where the navigation element is one or more of the guidance road, the boundary of the actual road, and the clear lane line.

[0087] It should be understood that the foregoing modules are merely examples. During actual application, the foregoing modules may be added or deleted based on an actual requirement. For example, in the architecture of the system shown in FIG. 2, the map information obtaining module 220 and the road information generation module 230 may be combined into one module; or the road information generation module 230 and the planning and control module 240 may be combined into one module. For another example, the processor in the perception module 210 may be an independent processing module, and the processing module may be disposed in the intelligent driving device, or may be disposed in a cloud server that communicates with the intelligent driving device.

[0088] The foregoing describes the architecture of the autonomous driving system provided in embodiments of this application. The following describes in detail a procedure of implementing, based on the autonomous driving system shown in FIG. 2, the autonomous driving method provided in embodiments of this application.

[0089] An intelligent driving device in embodiments of this application may include a transportation tool on a road, a transportation tool on water, an air transportation tool, an industrial device, an agricultural device, an entertainment device, or the like. For example, the intelligent driving device may be a vehicle. The vehicle is a vehicle in a broad sense, and may be a transportation tool (for example, a commercial vehicle, a passenger vehicle, a motorcycle, a flying vehicle, or a train), an industrial vehicle (for example, a forklift truck, a trailer, or a tractor), an engineering vehicle (for example, an excavator, a bulldozer, or a crane), an agricultural device (for example, a lawn mower or a harvester), a recreational device, a toy vehicle, or the like. A type of the vehicle is not specifically limited in embodiments of this application. For ease of understanding, the following uses an example in which the intelligent driving device is a vehicle for description.

[0090] FIG. 3 is a schematic flowchart of an autonomous driving method according to an embodiment of this application. The method 300 may be applied to the intelligent driving device shown in FIG. 1, or the method may be performed by the system shown in FIG. 2. More specifically, the method 300 may be performed by the primary navigation path determining module 231, and the method 300 may include the following steps.

[0091] S301: Determine a plurality of perceived paths based on environment information obtained by a perception system, where each of the plurality of perceived paths is a path that connects one road to another road through an intersection.

[0092] In some implementations, the environment information may include an image of a preset range ahead of a vehicle. Elements such as a road and an intersection are extracted based on the image, and the elements such as the road and the intersection are spliced to obtain the plurality of perceived paths. For example, the preset range may be a farthest range that can be perceived by a camera apparatus or a radar of the vehicle, or the preset range may be a range within 150 meters, or a range within 200 meters, or may be another range ahead of the vehicle.

[0093] For example, as shown in FIG. 4A, in a process in which an ego vehicle travels from a road A to an intersection a, an image that includes the intersection a and a part of the road A to a road D is obtained, and elements such as the road A, the road B, the road C, the road D, and the intersection a are extracted through processing of the image, to determine the following three perceived paths based on a topology relationship between the roads: a path that connects the road A to the road B through the intersection a, a path that connects the road A to the road C through the intersection a, and a path that connects the road A to the road D through the intersection a.

[0094] S302: Determine a primary navigation path from the plurality of perceived paths based on an SD map navigation path.

[0095] The SD map navigation path is a part of or all of a navigation path that is generated based on an SD map and extends from a current location of the vehicle to a target location. For example, the SD map navigation path may be a navigation path that is clipped from a full navigation path and that is within a specific range of the current location of the ego vehicle. The SD map navigation path includes a plurality of shape points, and each shape point indicates a location in geographic space.

[0096] For example, the current location of the ego vehicle may be determined based on GPS data, and an SD navigation path that matches the current location of the ego vehicle may be determined. A start point of the SD map navigation path may be the current location of the ego vehicle, or a start point of the SD map navigation path may be within a preset distance behind the current location of the ego vehicle. For example, the preset distance may be 20 meters, 30 meters, or another distance; and a total length of the SD map navigation path may be 100 meters, or may be 120 meters, or may be another length.

[0097] For example, when traveling toward the target location, the ego vehicle needs to switch from the road A shown in FIG. 4A to the road B through the intersection a, and the SD map navigation path may be shown in FIG. 4B. Further, geometric matching may be performed between the SD map navigation path and each of the plurality of perceived paths, overlapping degrees between the plurality of shape points of the SD navigation path and each perceived path are calculated, and a perceived path with a highest overlapping degree is selected as the primary navigation path. FIG. 5A to FIG. 5C show results of geometric matching between the SD map navigation path and each of the three perceived paths. It can be learned that the SD map navigation path and the path that connects the road A to the road B through the intersection a have a highest overlapping degree. Therefore, it may be determined that the path that connects the road A to the road B through the intersection a is the primary navigation path.

[0098] According to the autonomous driving method provided in this embodiment of this application, the primary navigation path for the vehicle to travel toward the target location can be determined based on the SD map navigation path and road information that is perceived by the vehicle, so that the vehicle can be controlled to travel along the primary navigation path toward the target location without a high definition map.

[0099] FIG. 6 is another schematic flowchart of an autonomous driving method according to an embodiment of this application. The method 400 may be applied to the intelligent driving device shown in FIG. 1, or the method may be performed by the system shown in FIG. 2. More specifically, the method 400 may be performed by the determining module 232 for information for exiting an intersection, and the method 400 may include the following steps.

[00100] S401: Determine, based on environment information obtained by a perception system, an intersection boundary of an intersection through which a primary navigation path passes.

[00101] For example, the intersection through which the primary navigation path passes may include the intersection a in the method 300.

[00102] In some implementations, the environment information may include an image of a preset range ahead of a vehicle. In this case, an image of an intersection region through which the primary navigation path passes may be extracted based on the image, and a perceived intersection boundary is determined based on the image of the intersection region. The perceived intersection boundary may be a boundary of the intersection region that is directly extracted from the image. In some scenarios, because the vehicle is far away from an intersection, an image obtained by the perception system of the vehicle includes only an image of a part of an intersection region. For example, as shown in FIG. 7A, a perceived intersection boundary is a boundary determined based on the image that includes the part of the intersection region and that is obtained by the vehicle. In this case, an initial intersection boundary may be adjusted based on a length of one side boundary (for example, a boundary a) for entering the intersection, to obtain a predicted intersection boundary, so that a shape and a location of the predicted intersection boundary are as close as possible to a shape and a location of an actual intersection boundary, and the predicted intersection boundary is used as the intersection boundary of the intersection through which the primary navigation path passes.

[00103] As the vehicle travels forward, an area of the intersection region included in the image obtained by the vehicle is increasingly large, and the intersection region is increasingly complete. When the area of the intersection region (namely, an area surrounded by the initial intersection boundary) extracted based on the image is greater than or equal to an area surrounded by the predicted intersection boundary, or when the image obtained by the vehicle includes a farthest end (for example, a boundary b) of the intersection region and tracking is stable, it may be determined that the perceived intersection boundary is a complete (or actual) boundary of the intersection region. Further, it may be determined that the perceived intersection boundary is the intersection boundary of the intersection through which the primary navigation path passes.

[00104] It should be noted that the intersection boundary in this application is a region in which a road is connected to an intersection. For example, a stop line at an actual intersection may be considered as a boundary of the intersection.

[00105] S402: Determine a direction for exiting the intersection based on an SD map navigation path.

[00106] For example, the SD map navigation path may be divided into a path for traveling into the intersection and a path for exiting the intersection based on the SD map navigation path and a location of the intersection. The path for traveling into the intersection indicates a direction for entering the intersection through which the primary navigation path passes, and the path for exiting the intersection indicates a direction for exiting the intersection through which the primary navigation path passes. Further, a specific quantity of shape points may be selected on the path for exiting the intersection, and then the direction for exiting the intersection is determined based on the selected shape points. The specific quantity may be 3, 5, or another value.

[00107] For example, as shown in FIG. 7B, three shape points may be separately selected at positions ①, ②, and ③, and fitting is performed based on coordinates indicated by the three shape points, to obtain a direction for exiting the intersection shown in FIG. 7C.

[00108] S403: Determine a location for exiting the intersection based on the intersection boundary and an azimuth for exiting the intersection.

[00109] For example, the intersection boundary may be the predicted intersection boundary, or may be the perceived intersection boundary. As the vehicle travels toward the intersection on the primary navigation path, a specific type of the intersection boundary may be switched. For a specific switching rule, refer to the descriptions in S401. Details are not described herein again.

[00110] In some implementations, an intersection geometric center is determined based on the intersection boundary, and an intersection point between a ray of the intersection geometric center in the direction for exiting the intersection and the intersection boundary (for example, a boundary c) is an initial location for exiting the intersection. Further, the initial location for exiting the intersection is offset by a first distance toward a traveling side of the vehicle to avoid an oncoming lane, to obtain a final location for exiting the intersection. The first distance may be a preset distance, for example, 3 meters or 2 meters, or the preset distance may be another distance determined based on a quantity of lanes of a road at the location for exiting the intersection. The quantity of lanes of the road at the location for exiting the intersection may be a quantity of lanes provided by an electronic horizon provider (electronic horizon provider, EHP).

[00111] It should be noted that, when a traffic rule is a right-side passing principle, the traveling side of the vehicle is the right side, that is, the initial location for exiting the intersection is offset by the first distance toward the right side, to obtain the final location for exiting the intersection. When the traffic rule is a left-side passing principle, the traveling side of the vehicle is the left side, that is, the initial location for exiting the intersection is offset by the first distance toward the left side, to obtain the final location for exiting the intersection.

[00112] According to the autonomous driving method provided in this embodiment of this application, a location and a direction for the vehicle to exit the intersection along the primary navigation path can be determined based on the SD map navigation path and the intersection boundary that is perceived by the vehicle.

[00113] FIG. 8 is still another schematic flowchart of an autonomous driving method according to an embodiment of this application. The method 500 may be applied to the intelligent driving device shown in FIG. 1, or the method may be performed by the system shown in FIG. 2. More specifically, the method 500 may be performed by the guidance information generation module 233, and the method 500 may include the following steps.

[00114] S501: Generate, based on a location for exiting an intersection and a direction for exiting the intersection, a guidance lane / road for exiting the intersection.

[00115] For example, as shown in FIG. 9A and FIG. 9B, a guidance road may be generated by using a straight line, along the direction for exiting the intersection, that passes through the location for exiting the intersection as a road center line, and a width of the guidance road may be a first preset width. For example, the first preset width may be 5 meters, 6 meters, or another value.

[00116] In some implementations, when information about a quantity of lanes included in a road at the exit of the intersection is obtained, the width of the guidance road may be a product of the quantity of lanes and a second preset width. For example, the second preset width may be 3.5 meters, 3.6 meters, or another value. For example, if the quantity of lanes is 2 and the second preset width is 3.5 meters, the width of the guidance road is 2×3.5=7 meters. For another example, if the quantity of lanes is 1 and the second preset width is 3.5 meters, the width of the guidance road is 3.5 meters. Further, a plurality of guidance lanes may be generated on the guidance road based on the quantity of lanes.

[00117] In some implementations, the guidance lane / road is a unidirectional lane / road. The following uses a "guidance road" to represent a "guidance lane / road". When no specific description is provided, the guidance road in the following may represent a guidance road, or may represent a guidance lane.

[00118] S502: Correct a location of the guidance road.

[00119] In some implementations, the location of the guidance road may be dynamically corrected in an inertial filtering manner. For example, the location of the guidance road may meet the following relationship:Q(xi, yi)=AQf(xi, yi)+BQf–1(xi, yi).

[00120] Q(xi, yi) represents coordinates of a point i on a boundary of a corrected guidance road, Qf(xi, yi) represents coordinates of the point i on a boundary of a guidance road determined in a current frame, Qf–1(xi, yi) represents coordinates of the point i on a boundary of a guidance lane determined in a previous frame, A and B respectively represent weights of a result of the current frame and a result of the previous frame, and A:B=1:9, or a ratio of A to B may be another value. For example, a result of the guidance road in one frame may be obtained by running the method 400 and the method 500 once. In other words, a current frame is coordinates of the guidance road obtained through a current run of the method 400 and the method 500, and a previous frame is coordinates of the guidance road obtained through a previous run of the method 400 and the method 500.

[00121] In some other implementations, the location of the guidance road may be further corrected based on a location of an unclear road. The unclear road indicates a road boundary determined based on an environment obtained by a perception system, and the unclear road is a unidirectional road. When a vehicle travels along a primary navigation path toward the exit of the intersection, the vehicle can obtain environment information near the exit of the intersection, and determine a boundary of a road at the exit of the intersection, namely, the unclear road, based on the environment information near the exit of the intersection. There may be a specific deviation between the location of the unclear road and the location of the guidance road generated in S501. In this case, the location of the guidance road may be adjusted based on the location of the unclear road. For example, a location of the guidance road in a current frame may meet the following relationship:Qf(xi, yi)=CQf'(xi, yi)+DQs(xi, yi).

[00122] Qf'(xi, yi) is coordinates of the point i on the boundary of the guidance road generated in the current frame, Qs(xi, yi) is coordinates of the point i on a boundary of the unclear road, C and D respectively represent a weight of Qf'(xi, yi) and a weight of Qs(xi, yi), and C:D=8:2, or a ratio of C to D may be another value.

[00123] According to the autonomous driving method provided in this embodiment of this application, the guidance lane can be generated based on the location and the direction for exiting the intersection that are determined in the method 400, to control a pose and / or a speed at which the vehicle passes through the intersection.

[00124] FIG. 10 is still another schematic flowchart of an autonomous driving method according to an embodiment of this application. The method 600 may be applied to the intelligent driving device shown in FIG. 1, or the method may be performed by the system shown in FIG. 2. More specifically, the method 600 may include S601 and S602. S601 and S602 may be performed by the planning and control module 240; or S601 may be performed by the perception module 210, and S602 may be performed by the planning and control module 240.

[00125] S601: Determine a navigation information source based on a perception result of road information at an exit of an intersection.

[00126] For example, the navigation information source is information used to control a pose and / or a speed of a vehicle to enable the vehicle to travel toward the exit of the intersection. The navigation information source may include one or more of a guidance road, an unclear road, and a clear lane. The unclear road indicates a boundary that is of the road at the exit of the intersection and that is determined based on environment information perceived by the vehicle, and the clear lane indicates a location that is of a lane line of the road at the exit of the intersection and that is determined based on the environment information perceived by the vehicle.

[00127] It may be understood that when the vehicle is far away from the exit of the intersection, effective environment information of the exit of the intersection cannot be obtained. The "effective environment information of the exit of the intersection" means perception information (such as an image or other information) from which a road boundary and / or a lane line can be extracted. As the vehicle gradually travels toward the exit of the intersection, the vehicle can obtain the effective environment information of the exit of the intersection. For example, as a distance between the vehicle and the exit of the intersection decreases, the unclear road and the clear lane may be sequentially extracted from the environment information of the exit of the intersection. For example, when the vehicle is 50 meters away from the exit of the intersection, the unclear road may be extracted from the environment information of the exit of the intersection. However, due to limited perception precision, the clear lane cannot be extracted from the environment information of the exit of the intersection. When the vehicle is 20 meters away from the exit of the intersection, the clear lane may be extracted from the environment information of the exit of the intersection.

[00128] When the vehicle does not obtain information about the unclear road and the clear lane, the vehicle may determine that the guidance road is the navigation information source. When the vehicle obtains the information about the unclear road or the clear lane, the vehicle may switch the navigation information source to the unclear road or the clear lane. For example, the navigation information source may be determined and / or the navigation information source may be switched based on content shown in Table 1.Table 1ConditionNavigation information sourceInformation about an unclear road and information about a clear lane are not obtainedGuidance roadThe information about the unclear road is obtained, but the information about the clear lane is not obtainedSwitch from the guidance road to the unclear roadThe information about the clear lane is obtained, but the information about the unclear road is not obtainedSwitch from the guidance road to the clear laneThe information about the unclear road and the information about the clear lane are obtainedSwitch from the guidance road to the unclear road and from the unclear road to the clear lane in sequenceThe information about the unclear road (or the clear lane) is lostSwitch from the unclear road (or the clear lane) to the guidance road 

[00129] In some implementations, before the navigation information source is switched, whether to switch may be determined based on reliability (or referred to as confidence) of a target switched to, and the navigation information source is then switched when the reliability of the target switched to is greater than or equal to a reliability threshold. The target switched to is a switched navigation information source. For example, when the navigation information source is switched from the guidance road to the unclear road, the unclear road is the target switched to. When the information about the unclear road is obtained, the information about the clear lane is not obtained, and reliability of the unclear road is greater than or equal to the reliability threshold, the navigation information source is switched from the guidance road to the unclear road.

[00130] For example, the reliability of the target switched to may be determined based on a formula P=aP1+bP2, where P represents the reliability, P1 represents reliability determined based on a data statistics result, P2 represents reliability determined based on a machine learning algorithm, a and b represent a weight of P1 and a weight of P2 respectively, values of a and b range from 0 to 1, and a sum of a and b is 1. More specifically, the data statistics result may be a result of collecting statistics on collected location information of the unclear road (or the clear lane) in a plurality of frames. For example, a variance of a sample that includes location information in the plurality of frames is determined, and P1 is determined based on the variance. It may be understood that a larger variance indicates a larger deviation between location information in different frames. Therefore, it may be considered that a collected result is unstable. In this case, P1 is smaller. A model used by the machine learning algorithm may be obtained through training based on a location that is of the road boundary (or the lane line) and that is manually marked. A location that is of the road boundary (or the lane line) and that is collected by a perception system of the vehicle is input into the foregoing model, to obtain a deviation between data collected by the perception system of the vehicle and a location of an actual road (or an actual lane). P2 is determined based on the deviation. It may be understood that a larger deviation indicates a more inaccurate result collected by the vehicle. In this case, P2 is smaller.

[00131] For example, when the navigation information source is the guidance road and reliability of the unclear road is greater than a reliability threshold 1, the navigation information source is switched from the guidance road to the unclear road. When the navigation information source is the guidance road or the unclear road, and reliability of the clear lane is greater than a reliability threshold 2, the navigation information source is switched from the guidance road or the unclear road to the clear lane. The reliability threshold 2 is greater than the reliability threshold 1. For example, the reliability threshold 2 may range from 0.6 to 1, and the reliability threshold 1 may range from 0.3 to 0.5. Alternatively, when the reliability threshold is determined in another method, the reliability threshold 1 and the reliability threshold 2 may be other values.

[00132] In some implementations, after the navigation information source is switched once, the navigation information source is switched again after a specific duration. The specific duration may be 15 seconds, or 20 seconds, or may be another duration.

[00133] S602: Control a pose and / or a speed of the vehicle on a primary navigation path based on the navigation information source.

[00134] In some implementations, when the navigation information source is the guidance road, a pose at which the vehicle enters the intersection is controlled based on the location and a direction for exiting the intersection that are indicated by the guidance road, so that when the vehicle travels along the primary navigation path, the vehicle can travel toward the location and the direction that are indicated by the guidance road, to avoid a scratch between the vehicle and the road boundary at the exit of the intersection or a traffic conflict with an oncoming lane. Further, when the navigation information source is the guidance road, the vehicle is controlled to travel at a speed 1. For example, the speed 1 may range from 5 kph (kilometer per hour, kph) to 15 kph. In this way, the vehicle can have sufficient time to obtain road information (such as the road boundary and the lane line) at the intersection by using the perception system.

[00135] When the navigation information source is the unclear road, the vehicle may be controlled to travel at a speed 2. When the navigation information source is the clear lane, the vehicle may be controlled to travel at a speed 3. The speed 2 may range from 20 kph to 30 kph, and the speed 3 may range from 30 kph to 60 kph. It should be understood that the speed 1, the speed 2, and the speed 3 are merely examples for description. During actual implementation, the speed 1, the speed 2, and the speed 3 may alternatively be speeds determined based on a maximum speed limit of the intersection. For example, the speed 1 may be less than or equal to 70% or 80% of the maximum speed limit of the intersection, the speed 2 may be less than or equal to 80% or 90% of the maximum speed limit of the intersection, and the speed 3 may be less than or equal to a maximum limit of the intersection. The maximum limit of the intersection may be the maximum speed limit of the intersection, or may be a maximum speed limit of a road connected to the intersection, for example, a maximum speed limit of a current traveling road of the vehicle (for example, a road entering the intersection), or may be a maximum speed limit of a target traveling road of the vehicle (for example, a road departing from the intersection). Alternatively, the speed 1, the speed 2, and the speed 3 may be other values that meet the following conditions: The speed 1 is less than or equal to the speed 2, and the speed 2 is less than the speed 3.

[00136] In some implementations, there is a specific deviation between a location of the guidance road and a location indicated by the unclear road or the clear lane. In this case, when the navigation information source is the unclear road or the clear lane, the pose of the vehicle may be further adjusted based on the unclear road and the clear lane, to improve navigation precision.

[00137] According to the autonomous driving method provided in this embodiment of this application, the pose and / or the speed of the vehicle can be controlled based on the result of perceiving, by the vehicle, the road at the exit of the intersection. When the vehicle cannot obtain road information at a boundary for exiting the intersection, the pose of the vehicle is controlled based on guidance information, and the vehicle is controlled to travel at a slow speed, so that the vehicle has sufficient time to obtain and process the road information at the boundary for exiting the intersection. In this way, a probability of obtaining clear road information (for example, a clear road boundary and / or lane line) is increased. As the vehicle approaches the boundary for exiting the intersection, reliability and precision of the road information that is perceived by the vehicle and that is at the boundary for exiting the intersection gradually increase. In this case, the pose of the vehicle is controlled based on the unclear road and / or the clear lane, to help improve navigation precision, and the vehicle is controlled to travel at a fast speed based on the unclear road and / or the clear lane, to help improve efficiency of the vehicle in passing through the intersection.

[00138] It should be noted that the foregoing "unclear road" and "clear lane" are merely used to distinguish between types of perception information. The "unclear road" indicates a road boundary perceived by the vehicle, and the "clear lane" indicates a lane boundary perceived by the vehicle (for example, a lane line). "Unclear" and "clear" should not be understood as a limitation on definition of a perception result.

[00139] FIG. 11 is still another schematic flowchart of an autonomous driving method according to an embodiment of this application. The method 700 may be applied to the intelligent driving device shown in FIG. 1, or the method may be performed by the system shown in FIG. 2. More specifically, the method 700 may include S710 and S720.

[00140] S710: Obtain guidance information, where the guidance information indicates a direction and a location for an intelligent driving device to exit a first intersection, the guidance information is determined based on SD map navigation information, and the SD map navigation information indicates a navigation path from a current location of the intelligent driving device to a target location.

[00141] For example, the intelligent driving device may include the vehicle in the foregoing embodiments, or may be another intelligent driving device. The first intersection may include the intersection a in the foregoing embodiments. The guidance information may include the location and the direction for exiting the intersection that are determined in the method 400, or the guidance information may further include the guidance road determined in the method 500. The SD map navigation information may include the SD map navigation path in the foregoing embodiments.

[00142] S720: Control, based on the guidance information, a speed and / or a pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection.

[00143] In some implementations, second road information within a first range of the intelligent driving device is obtained, where the second road information indicates a plurality of perceived paths passing through the first intersection. A primary navigation path is determined from the plurality of perceived paths based on the SD map navigation information. That the speed and / or the pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection are / is controlled includes: controlling a speed and / or a pose at which the intelligent driving device travels along the primary navigation path.

[00144] For example, the first range may include the preset range in the method 300. When the first intersection is the intersection a, the plurality of perceived paths may include: a path that connects a road A to a road B through the intersection a, a path that connects the road A to a road C through the intersection a, and a path that connects the road A to a road D through the intersection a. For a specific implementation of determining the plurality of perceived paths, refer to the descriptions in the method 300, and details are not described herein again.

[00145] In some implementations, that the speed and / or the pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection are / is controlled based on the guidance information includes: controlling, based on the guidance information, the intelligent driving device to travel toward the first intersection or pass through the first intersection at a first speed. For example, the first speed may include the speed 1 in the method 600.

[00146] In some implementations, a first boundary of the first intersection is a target travel-out boundary of the intelligent driving device, the first boundary is connected to a first road, and the method further includes: obtaining information that is about the first road and that is perceived by the intelligent driving device. That the speed and / or the pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection are / is controlled based on the guidance information includes: controlling the speed and / or the pose of the intelligent driving device based on the first road information and the guidance information.

[00147] For example, when the first intersection is the intersection a, the first boundary may be the boundary c shown in FIG. 7C, the first road connected to the first boundary is the road B, and the first road information includes the information about the unclear road and / or information about the clear lane.

[00148] In some implementations, the first road information indicates a boundary of the first road and / or a location of a lane line on the first road. That the speed and / or the pose of the intelligent driving device are / is controlled based on the first road information and the guidance information includes: when reliability of the first road information is less than or equal to a preset threshold, controlling, based on the guidance information, the intelligent driving device to travel at the first speed; or when the reliability of the first road information is greater than the preset threshold, controlling, based on the first road information, the intelligent driving device to switch from traveling at the first speed to traveling at a second speed, where the first speed is less than or equal to the second speed.

[00149] For example, when the first road information indicates the boundary of the first road (for example, the first road information is information about an unclear road), the second speed may include the speed 2. When the first road information indicates the location of the lane line on the first road (for example, the first road information is information about a clear lane), the second speed may include the speed 3. For a more specific implementation of controlling the speed of the intelligent driving device based on the first road information and the guidance information, refer to the descriptions in the method 600. Details are not described herein again.

[00150] In some implementations, the guidance information includes a guidance road, and the method further includes: obtaining information about a boundary of the first intersection; determining, based on the SD map navigation information, the direction for exiting the first intersection; determining, based on the information about the boundary and the direction for exiting the first intersection, the location for exiting the first intersection; and generating the guidance road based on the location for exiting the first intersection and the direction for exiting the first intersection.

[00151] For example, the information about the boundary of the first intersection may include the intersection boundary in the method 400. For specific implementations of determining the direction for exiting the first intersection and determining the location for exiting the first intersection, refer to the descriptions in the method 400. Details are not described herein again. Further, for a method for generating the guidance road, refer to the descriptions in the method 500. Details are not described herein again.

[00152] In some implementations, that the guidance road is generated based on the location for exiting the first intersection and the direction for exiting the first intersection includes: generating an initial guidance road based on the location for exiting the first intersection and the direction for exiting the first intersection; and correcting the initial guidance road based on a location of the first road connected to the boundary at which the intelligent driving device exits the first intersection, to obtain the guidance road.

[00153] For example, the location of the first road may be indicated by the first road information. For a method for correcting the initial guidance road based on the location of the first road, refer to the descriptions in S502. Details are not described herein again.

[00154] In some implementations, the method further includes: controlling a display apparatus to display a navigation interface, where the navigation interface includes an image of the guidance road and a first icon, and the first icon indicates a location of the intelligent driving device relative to the guidance road.

[00155] In some implementations, the navigation interface further includes an image of the boundary of the first road. During actual implementation, the image of the boundary of the first road may gradually change, for example, gradually become clear, as the intelligent driving device travels toward the first boundary of the first intersection.

[00156] In some implementations, when information about the lane line of the first road is obtained, the navigation interface is controlled to switch from the image of the guidance road to an image of the lane line of the first road.

[00157] In some implementations, when the information about the lane line of the first road is obtained, the navigation interface may be switched from displaying the image of the boundary of the first road to displaying the image of the lane line of the first road.

[00158] According to the autonomous driving method provided in this embodiment of this application, an autonomous driving capability of the intelligent driving device can be improved without relying on a high definition map. Specifically, when the intelligent driving device is in an autonomous driving state, and the intelligent driving device is far away from the intersection and cannot obtain road information at the intersection (for example, information used for autonomous driving navigation such as the road boundary or the lane line) by using a perception apparatus, or precision of obtained perception information is insufficient, a direction and a location for exiting the intersection can be determined in advance based on the guidance information, thereby guiding heading adjustment of a vehicle body and preventing the intelligent driving device from being stuck near the intersection. In this way, a pass rate and a success rate of the intelligent driving device at the intersection without a high definition map are improved, and a probability that the intelligent driving device travels into an oncoming lane or scratches the road boundary when the intelligent driving device exits the intersection can be reduced. In addition, the intelligent driving device is controlled, based on the guidance information, to travel at a low speed, so that when the intelligent driving device travels toward the first boundary of the first intersection, there is sufficient time to obtain road information at the first boundary, to increase a probability of obtaining clear road information (for example, a clear lane line). This helps improve a planning and control capability and driving safety of the intelligent driving device in an autonomous driving process. As the intelligent driving device travels toward the first boundary, reliability and precision of first road information obtained by the intelligent driving device gradually increase. Controlling a pose of the intelligent driving device based on the information (for example, the first road information) about the first road obtained by a perception system of the vehicle is more reliable than controlling a pose of the intelligent driving device based on the guidance information, and helps prevent the intelligent driving device from touching a lane line or a road boundary line when the intelligent driving device travels on the first road. In addition, the intelligent driving device is controlled, based on the first road information, to travel at a high speed, to further improve a speed and efficiency of the intelligent driving device in passing through the first intersection.

[00159] In various embodiments of this application, unless otherwise stated or there is a logic conflict, terms and / or descriptions in various embodiments are consistent and may be mutually referenced, and technical features in different embodiments may be combined based on an internal logical relationship thereof, to form a new embodiment.

[00160] The autonomous driving method provided in embodiments of this application is described above in detail with reference to FIG. 1 to FIG. 11. The following describes in detail an apparatus provided in embodiments of this application with reference to FIG. 12 and FIG. 13. It should be understood that descriptions of apparatus embodiments correspond to the descriptions of the method embodiments. Therefore, for content that is not described in detail, refer to the foregoing method embodiments. For brevity, details are not described herein again.

[00161] FIG. 12 is a block diagram of an autonomous driving apparatus 2000 according to an embodiment of this application. The apparatus 2000 may include units configured to perform the method 300, the method 400, the method 500, the method 600, and the method 700. In addition, the units in the apparatus 2000 are configured to implement corresponding procedures in the foregoing method embodiments. The apparatus 2000 includes an obtaining unit 2010. The obtaining unit 2010 may be configured to implement a corresponding data obtaining or receiving / sending function. The apparatus 2000 further includes a processing unit 2020. The processing unit 2020 may be configured to implement a corresponding processing function.

[00162] Optionally, the apparatus 2000 may further include a storage unit, the storage unit may be configured to store instructions and / or data, and the processing unit 2020 may read the instructions and / or the data in the storage unit, so that the apparatus implements actions in the foregoing method embodiments.

[00163] It should be understood that a specific process in which the units perform the foregoing corresponding steps is described in detail in the foregoing method embodiments, and for brevity, details are not described herein.

[00164] It should be understood that the apparatus 2000 herein is embodied in a form of a functional unit. The term "module" or "unit" herein may mean an application-specific integrated circuit (ASIC), an electronic circuit, a processor (for example, a shared processor, a dedicated processor, or a group processor) configured to execute one or more software or firmware programs, a memory, a merged logic circuit, and / or another appropriate assembly that supports the described function.

[00165] The apparatus in the foregoing solutions has a function of implementing corresponding steps performed by the computing platform 150 in the foregoing methods. The function may be implemented by hardware, or may be implemented by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the foregoing functions. For example, the obtaining unit 2010 may be replaced with a transceiver, and another unit, such as a processing unit, may be replaced with a processor configured to perform related processing operations in the method embodiments.

[00166] For example, the obtaining unit 2010 and the processing unit 2020 may be disposed in the intelligent driving device 100 shown in FIG. 1, or may be disposed in the system shown in FIG. 2. More specifically, the obtaining unit 2010 and the processing unit 2020 may be disposed in the planning and control module 240, or may be disposed in the road information generation module 230. For example, operations performed by the obtaining unit 2010 and the processing unit 2020 may be performed by one processor, or may be performed by different processors. In a specific implementation process, the one or more processors may be processors disposed in the intelligent driving device 100 shown in FIG. 1, or the apparatus 2000 may be a chip disposed in the intelligent driving device 100.

[00167] In a specific implementation process, all or some of the units in the foregoing apparatus may be integrated together, or may be implemented independently. In an implementation, the units are integrated and implemented in a form of a system-on-a-chip (system-on-a-chip, SoC).

[00168] FIG. 13 is another block diagram of an autonomous driving apparatus according to an embodiment of this application. The autonomous driving apparatus 2100 shown in FIG. 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 through an internal connection road. The memory 2130 is configured to store instructions. The processor 2110 is configured to execute the instructions stored in the memory 2130, to implement the method in the foregoing embodiments. Optionally, the memory 2130 may be coupled to the processor 2110 by using an interface, or may be integrated with the processor 2110.

[00169] It should be noted that the transceiver 2120 may include but is not limited to a transceiver apparatus of an input / output interface (input / output interface) type, to implement communication between the apparatus 2100 and another device or a communication network.

[00170] The memory 2130 may be a volatile memory and / or a non-volatile memory. The nonvolatile memory may be a read-only memory (read-only memory, ROM), a programmable read-only memory (programmable ROM, PROM), an erasable programmable read-only memory (erasable PROM, EPROM), an electrically erasable programmable read-only memory (electrically EPROM, EEPROM), or a flash memory. The volatile memory may be a random access memory (random access memory, RAM). For example, the RAM may be used as an external cache. By way of example, but not limitation, the RAM includes a plurality of forms, such as a static random access memory (static RAM, SRAM), a dynamic random access memory (dynamic RAM, DRAM), a synchronous dynamic random access memory (synchronous DRAM, SDRAM), a double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), an enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), a synchlink dynamic random access memory (synchlink DRAM, SLDRAM), and a direct rambus random access memory (direct rambus RAM, DR RAM).

[00171] The transceiver 2120 uses, for example, but is not limited to, a transceiver apparatus of a transceiver type, to implement communication between the apparatus 2100 and another device or a communication network, to receive / send data / information used to implement the method in the foregoing embodiments.

[00172] An embodiment of this application further provides an intelligent driving device. The intelligent driving device includes the autonomous driving apparatus 2000 or the autonomous driving apparatus 2100 in the foregoing embodiments.

[00173] An embodiment of this application further provides a computer program product. The computer program product includes computer program code. When the computer program code is run on a computer, the computer is enabled to implement the method in the foregoing embodiments of this application.

[00174] An embodiment of this application further provides a computer-readable storage medium. The computer-readable medium stores computer instructions. When the computer instructions are run on a computer, the computer is enabled to implement the method in the foregoing embodiments of this application.

[00175] An embodiment of this application further provides a chip that includes a circuit, configured to perform the method in the foregoing embodiments of this application.

[00176] It may be clearly understood by a person skilled in the art that, for the purpose of convenient and brief description, for a detailed working process of the foregoing system, apparatus, and unit, refer to a corresponding process in the foregoing method embodiments. Details are not described herein again.

[00177] In the descriptions of embodiments of this application, " / " means "or" unless otherwise specified. For example, A / B may represent A or B. In this specification, "and / or" describes an association relationship between associated objects and represents that three relationships may exist. For example, A and / or B may represent the following three cases. Only A exists, both A and B exist, and only B exists. In this application, "at least one" means one or more, and "a plurality of" means two or more. "At least one of the following items (pieces)" or a similar expression thereof means any combination of these items, including any combination of singular items (pieces) or plural items (pieces). For example, at least one item (piece) of a, b, or c may indicate: a, b, c, a and b, a and c, b and c, or a, b, and c, where a, b, and c may be singular or plural.

[00178] In embodiments of this application, prefixes such as "first" and "second" are used only to distinguish different described objects, and do not limit a location, a sequence, a priority, a quantity, or content of described objects. In embodiments of this application, use of a prefix word, for example, an ordinal number, used to distinguish between described objects, does not constitute a limitation on the described objects. For descriptions of the described objects, refer to the descriptions of the context in the claims or embodiments. Use of such a prefix word should not constitute a redundant limitation.

[00179] In the several embodiments provided in this application, it should be understood that the disclosed system, apparatus, and method may be implemented in another manner. For example, the described apparatus embodiments are merely examples. For example, the division of units is merely logical function division and may be other division in actual implementation. For example, a plurality of units or components may be combined or integrated into another system, or some features may be ignored or may not be performed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections may be implemented by using some interfaces. The indirect couplings or communication connections between apparatuses or units may be implemented in electronic, mechanical, or other forms.

[00180] In various embodiments of this application, unless otherwise stated or there is a logic conflict, terms and / or descriptions in various embodiments are consistent and may be mutually referenced, and technical features in different embodiments may be combined based on an internal logical relationship thereof, to form a new embodiment.

[00181] The units described as separate components may or may not be physically separate, and components displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. Some or all of the units may be selected based on actual requirements to achieve the objectives of the solutions of embodiments.

[00182] In addition, functional units in embodiments of this application may be integrated into one processing unit, each of the units may exist alone physically, or two or more units may be integrated into one unit.

[00183] The foregoing descriptions are merely specific implementations of this application, but are not intended to limit the protection scope of this application. Any variation or replacement readily figured out by a person skilled in the art within the technical scope disclosed in this application shall fall within the protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

Claims

1. An autonomous driving method, comprising:obtaining guidance information, wherein the guidance information indicates a direction and a location for an intelligent driving device to exit a first intersection, the guidance information is determined based on standard definition SD map navigation information, and the SD map navigation information indicates a navigation path from a current location of the intelligent driving device to a target location; andcontrolling, based on the guidance information, a speed and / or a pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection.

2. The method according to claim 1, wherein controlling, based on the guidance information, the speed and / or the pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection comprises:controlling, based on the guidance information, the intelligent driving device to travel toward the first intersection or pass through the first intersection at a first speed, wherein the first speed is less than or equal to a speed threshold.

3. The method according to claim 1, wherein a first boundary of the first intersection is a target travel-out boundary of the intelligent driving device, the first boundary is connected to a first road, and the method further comprises:obtaining first road information that is perceived by the intelligent driving device; andcontrolling, based on the guidance information, the speed and / or the pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection comprises:controlling the speed and / or the pose of the intelligent driving device based on the first road information and the guidance information.

4. The method according to claim 3, wherein the first road information indicates a boundary of the first road and / or a location of a lane line on the first road; andcontrolling the speed and / or the pose of the intelligent driving device based on the first road information and the guidance information comprises:when reliability of the first road information is less than or equal to a preset threshold, controlling, based on the guidance information, the intelligent driving device to travel at the first speed; orwhen the reliability of the first road information is greater than the preset threshold, controlling, based on the first road information, the intelligent driving device to switch from traveling at the first speed to traveling at a second speed, whereinthe first speed is less than or equal to the second speed.

5. The method according to claim 1, wherein the guidance information comprises a guidance road, and obtaining the guidance information comprises:obtaining information about a boundary of the first intersection;determining, based on the SD map navigation information, the direction for exiting the first intersection;determining, based on the information about the boundary and the direction for exiting the first intersection, the location for exiting the first intersection; andgenerating the guidance road based on the location for exiting the first intersection and the direction for exiting the first intersection.

6. The method according to claim 5, wherein generating the guidance road based on the location for exiting the first intersection and the direction for exiting the first intersection comprises:generating an initial guidance road based on the location for exiting the first intersection and the direction for exiting the first intersection; andcorrecting the initial guidance road based on a location of the first road connected to the boundary at which the intelligent driving device exits the first intersection, to obtain the guidance road.

7. The method according to claim 5, wherein the method further comprises:controlling a display apparatus to display a navigation interface, wherein the navigation interface comprises an image of the guidance road and a first icon, and the first icon indicates a location of the intelligent driving device relative to the guidance road.

8. The method according to claim 7, wherein the navigation interface further comprises an image of the boundary of the first road, and the first road is a road connected to the target travel-out boundary of the intelligent driving device at the first intersection.

9. The method according to claim 7, wherein the method further comprises:when information about the lane line of the first road is obtained, controlling the navigation interface to switch from the image of the guidance road to an image of the lane line of the first road, wherein the first road is the road connected to the target travel-out boundary of the intelligent driving device at the first intersection.

10. The method according to claim 1, wherein the method further comprises:obtaining second road information within a first range of the intelligent driving device, wherein the second road information indicates a plurality of perceived paths passing through the first intersection; anddetermining a primary navigation path from the plurality of perceived paths based on the SD map navigation information; andcontrolling the speed and / or the pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection comprises:controlling a speed and / or a pose at which the intelligent driving device travels along the primary navigation path.

11. An autonomous driving apparatus, comprising:an obtaining unit, configured to obtain guidance information, wherein the guidance information indicates a direction and a location for an intelligent driving device to exit a first intersection, the guidance information is determined based on standard definition SD map navigation information, and the SD map navigation information indicates a navigation path from a current location of the intelligent driving device to a target location; anda processing unit, configured to control, based on the guidance information, a speed and / or a pose at which the intelligent driving device travels toward the first intersection or passes through the first intersection.

12. The apparatus according to claim 11, wherein the processing unit is configured to:control, based on the guidance information, the intelligent driving device to travel at a first speed, wherein the first speed is less than or equal to a speed threshold.

13. The apparatus according to claim 11, wherein a first boundary of the first intersection is a target travel-out boundary of the intelligent driving device, the first boundary is connected to a first road, and the obtaining unit is further configured to:obtain information that is about the first road and that is perceived by the intelligent driving device; andthe processing unit is configured to:control the speed and / or the pose of the intelligent driving device based on the first road information and the guidance information.

14. An autonomous driving apparatus, comprising:a processor, configured to execute a computer program stored in a memory, to enable the apparatus to perform the method according to any one of claims 1 to 10.

15. A computer-readable storage medium, wherein the computer-readable storage medium stores instructions, and when the instructions are executed by a processor, the method according to any one of claims 1 to 10 is implemented. 

Images