Intelligent driving method, device, and vehicle

Abstract

Provided in the present application are an intelligent driving method, a device, and a vehicle, which can be applied to the field of intelligent driving. The method comprises: on the basis of data collected by sensors of a vehicle, acquiring information of road elements around the vehicle, the vehicle being at a first position; on the basis of the first position and the information of the road elements, determining the physical connectivity of the vehicle from the first position to each sampling point among a plurality of sampling points; if the physical connectivity indicates that the vehicle is physically connected from the first position to at least some sampling points among the plurality of sampling points, planning a traveling trajectory for the vehicle to travel to the at least some sampling points; and, on the basis of the traveling trajectory, controlling the vehicle. The present application can be applied to intelligent vehicles or electric vehicles. Planning decision-making of vehicles may not depend on high-definition maps and lane topological relationships, thereby helping to improve driving and riding experience of users.

Description

Intelligent driving method, device and vehicle

[0001] This application claims priority to the Chinese patent application No. 202410866844.8, filed on June 28, 2024, and entitled "Intelligent driving method, device and vehicle", the entire content of which is incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the field of intelligent driving, and more particularly, to an intelligent driving method, device and vehicle. BACKGROUND

[0003] In the field of autonomous driving, the behavior decision of a vehicle often depends on the lane topology relationship in a high-definition (HD) map. On the one hand, the cost of using the HD map is high; on the other hand, the real-time performance of the HD map is poor, and if a real change scenario such as road construction is encountered, the vehicle in an autonomous driving state may have problems such as wrong lane selection and stopping. This will affect the driving safety of users. SUMMARY

[0004] The present application provides an intelligent driving method, device and vehicle, which can not rely on a high-definition map and lane topology relationship when planning and decision-making, which helps to reduce the cost required for intelligent driving and improve the driving safety of users.

[0005] In a first aspect, the present application provides an intelligent driving method, which comprises: acquiring information of road elements around a vehicle according to data collected by a sensor of the vehicle, the vehicle being at a first position; determining physical connectivity of the vehicle from the first position to each sampling point in a plurality of sampling points according to the first position and the information of the road elements; if the physical connectivity indicates that the vehicle is physically connected from the first position to at least part of the sampling points in the plurality of sampling points, planning a driving trajectory of the vehicle to the at least part of the sampling points; and controlling the vehicle according to the driving trajectory.

[0006] Based on the above technical solution, the vehicle can determine the physical connectivity of the vehicle to a plurality of sampling points according to the position and the information of the road elements. If the physical connectivity indicates that the vehicle is physically connected to at least part of the sampling points, the driving trajectory to the at least part of the sampling points can be planned, so as to control the vehicle to drive. In this way, the vehicle can make decision and planning through the information of the road elements obtained in real time, without relying on a high-definition map and lane topology relationship, avoiding problems such as wrong lane selection and stopping of the vehicle, and helping to improve the driving safety of users.

[0007] In some possible implementation manners, the road elements include obstacles and isolated road elements.

[0008] Exemplarily, the obstacle includes one or more of a curb, a median, a flower bed, a fence.

[0009] Exemplarily, the isolated road element includes one or more of a lane line, a road boundary.

[0010] In some possible implementation manners, the physical connectivity can be represented by a path searched in the path search result, or can also be represented by a semantic representation.

[0011] With reference to the first aspect, in some implementation manners of the first aspect, the determining of the physical connectivity of the vehicle from the first position to each of the plurality of sampling points according to the first position and the information of the road element comprises: searching a path from the first position to each of the plurality of sampling points according to the first position and the information of the road element to obtain a path search result; and determining the physical connectivity of the vehicle from the first position to each of the plurality of sampling points according to the path search result.

[0012] In some possible implementation manners, the plurality of sampling points includes a first sampling point, and the determining of the physical connectivity of the vehicle from the first position to each of the plurality of sampling points according to the path search result comprises: when the path from the first position to the first sampling point is searched, determining that the vehicle is physically connected from the first position to the first sampling point.

[0013] The above vehicle physically connected from the first position to the first sampling point can also be understood as that the vehicle is spatially reachable from the first position to the first sampling point.

[0014] In some possible implementation manners, the plurality of sampling points includes a second sampling point, and the determining of the physical connectivity of the vehicle from the first position to each of the plurality of sampling points according to the path search result comprises: when the path from the first position to the second sampling point cannot be searched, determining that the vehicle is not physically connected from the first position to the second sampling point.

[0015] Because the current HD map-based road element detection result is not updated in time, the road element actually appearing in the road can not appear in the HD map. For example, an obstacle (for example, a fence or a cone barrel) set in the road during road construction can not obtain information of the obstacle through the HD map, thereby causing the vehicle to select a wrong lane.

[0016] Based on the above technical solution, by perceiving the real-time detected road element, a path from the current position to each of the sampling points can be searched, so that the physical connectivity of the vehicle to each of the sampling points can be determined based on the path search result. In this way, some sampling points that cannot be reached from the current position can be accurately excluded, so as to avoid the vehicle from selecting a wrong lane and a collision accident, and to help improve the driving safety of the user.

[0017] With reference to the first aspect, in some implementations of the first aspect, the searching the path from the first position to each of the plurality of sampling points comprises: determining a plurality of triangular regions according to the information of the road element; performing cost mapping on edges of each of the plurality of triangular regions; and searching the path from the first position to each of the plurality of sampling points according to the triangular regions after cost mapping.

[0018] Based on the above technical solution, the vehicle does not need to use lane topology relationship when performing path searching, but uses a plurality of triangular regions determined by road elements existing in the real physical world to perform path searching, which helps to solve the problem of inaccurate path searching caused by unreasonable lane center line and lane topology relationship, thereby helping to improve the driving safety of users.

[0019] With reference to the first aspect, in some implementations of the first aspect, the planning the driving trajectory of the vehicle to drive to the at least part of the sampling points comprises: determining constraints for generating the driving trajectory according to the information of the road element; and planning the driving trajectory of the vehicle to drive to the at least part of the sampling points according to the constraints for generating the driving trajectory.

[0020] Based on the above technical solution, the constraints for generating the driving trajectory can be generated, and the driving trajectory of the vehicle to the at least part of the sampling points can be planned. In some possible implementations, when determining the physical connectivity from the current position to each of the sampling points, the vehicle can also extract the corresponding boundary information as the constraints for generating the driving trajectory. For example, the vehicle can model the road element in the Euclidean space, calculate the path from the current position of the vehicle to each of the sampling points, and extract the corresponding boundary information as the constraints for generating the finite element method (FEM) line. In this way, the generated driving trajectory is more human-like, and is no longer a line type that strongly depends on the lane topology relationship.

[0021] In some possible implementations, the method further comprises: when determining the physical connectivity from the current position to the sampling points, extracting the corresponding boundary information as the constraints for generating the driving trajectory.

[0022] With reference to the first aspect, in some implementations of the first aspect, the method further comprises: when the information of the road boundary exists in the road element, determining the plurality of sampling points according to the road trend of the road where the vehicle is located; or when the information of the road boundary does not exist in the road element, determining the plurality of sampling points according to the kinematic model of the vehicle.

[0023] Based on the technical solution, the decision result of the vehicle is no longer dependent on the lane topological relationship, but is based on the road trend or the kinematic model of the vehicle for sampling (for example, sampling in the Euclidean space). In this way, it can be ensured that the strategy space generated according to the driving trajectory is the full space, and cannot be sampled due to the existence of topological separation relationship.

[0024] In some possible implementation manners, the plurality of sampling points are determined according to the road trend of the road where the vehicle is located, including: determining a first set of sampling points according to the road trend of the road where the vehicle is located; and determining the plurality of sampling points from the first set of sampling points according to the kinematic legality of the vehicle.

[0025] In some possible implementation manners, the plurality of sampling points are determined according to the kinematic model of the vehicle, including: determining a first set of sampling points according to the kinematic model of the vehicle; and determining the plurality of sampling points from the first set of sampling points according to the kinematic legality of the vehicle.

[0026] With reference to the first aspect, in some implementation manners of the first aspect, the vehicle is controlled according to the driving trajectory, including: dividing the driving trajectory into one or more behavior clusters to obtain the strategy space; and controlling the vehicle according to the strategy space.

[0027] Based on the technical solution, the vehicle can cluster the driving trajectory to obtain the strategy space. For example, the physical connectivity can be expressed by semantics: change, keep. For example, keep can be further divided into keep in the self-lane after pressing the left virtual line to bypass the obstacle and keep in the self-lane after pressing the right lane line to bypass the obstacle.

[0028] In some possible implementation manners, the driving trajectory is divided into one or more behavior clusters to obtain the strategy space, including: clustering the driving trajectory according to the physical connectivity, and dividing the driving trajectories with the same physical connectivity boundary into the same cluster. In this way, in the strategy space evaluation, the behavior of the vehicle is no longer dependent on the lane topological relationship, but uses the actual behaviors that can be performed by the vehicle, so that both long-time line pressing and similar human behaviors can be performed, and various intelligent driving functions can be executed in the area without lane topological relationship.

[0029] With reference to the first aspect, in some implementation manners of the first aspect, before the information of the road elements around the vehicle is acquired according to the data collected by the sensor of the vehicle, the method further includes: determining that the vehicle is in an intelligent driving state.

[0030] In a second aspect, the present application provides an intelligent driving device, comprising: an acquisition unit, configured to acquire information of road elements around a vehicle according to data collected by sensors of the vehicle, the vehicle being at a first position; a determination unit, configured to determine physical connectivity of the vehicle from the first position to each of a plurality of sampling points according to the first position and the information of the road elements; a trajectory planning unit, configured to plan a driving trajectory of the vehicle to at least part of the sampling points if the physical connectivity indicates that the vehicle is physically connected from the first position to at least part of the sampling points; and a control unit, configured to control the vehicle according to the driving trajectory.

[0031] With reference to the second aspect, in some implementations of the second aspect, the device further comprises: a path searching unit, configured to search a path from the first position to each of the plurality of sampling points according to the first position and the information of the road elements to obtain a path search result; and wherein the determination unit is specifically configured to determine the physical connectivity of the vehicle from the first position to each of the plurality of sampling points according to the path search result.

[0032] With reference to the second aspect, in some implementations of the second aspect, the path searching unit is specifically configured to: determine a plurality of triangular regions according to the information of the road elements; perform cost mapping on edges of each of the plurality of triangular regions; and search the path from the first position to each of the plurality of sampling points according to the triangular regions after cost mapping.

[0033] With reference to the second aspect, in some implementations of the second aspect, the trajectory planning unit is specifically configured to: determine constraints for generating the driving trajectory according to the information of the road elements; and plan the driving trajectory of the vehicle to at least part of the sampling points according to the constraints for generating the driving trajectory.

[0034] With reference to the second aspect, in some implementations of the second aspect, the determination unit is further configured to: determine the plurality of sampling points according to a road trend of a road where the vehicle is located when there is information of a road boundary in the road elements; or determine the plurality of sampling points according to a kinematic model of the vehicle when there is no information of the road boundary in the road elements.

[0035] With reference to the second aspect, in some implementations of the second aspect, the control unit is specifically configured to: divide the driving trajectory into one or more behavior clusters to obtain a strategy space; and control the vehicle according to the strategy space.

[0036] With reference to the second aspect, in some implementations of the second aspect, the determination unit is further configured to determine that the vehicle is in an intelligent driving state before the acquisition unit acquires the information of the road elements around the vehicle.

[0037] In a possible implementation form of the second aspect, the road element comprises one or more of a static obstacle, a lane line, a road boundary.

[0038] In a third aspect, the present application provides an intelligent driving apparatus, comprising a memory and a processor, the memory is configured to store a computer program, and the processor is configured to execute the computer program in the memory, so that the intelligent driving apparatus can implement the method in the first aspect and any possible implementation form thereof.

[0039] In a fourth aspect, the present application provides an intelligent driving system, comprising a perception system and the apparatus in the second aspect or the third aspect.

[0040] In a fifth aspect, the present application provides a vehicle, comprising the apparatus in the second aspect or the third aspect, or comprising the system in the fourth aspect.

[0041] The vehicle in the present application is a vehicle in a broad sense, which can be a traffic tool (such as a commercial vehicle, a passenger vehicle, a motorcycle, a flying vehicle, a train, etc.), an industrial vehicle (such as a forklift, a trailer, a tractor, etc.), an engineering vehicle (such as an excavator, a bulldozer, a crane, etc.), an agricultural device (such as a mower, a harvester, etc.), a recreational device, a toy vehicle, etc. The type of the vehicle is not limited in the embodiments of the present application.

[0042] In a sixth aspect, the present application provides a computer program product, comprising computer program code, which, when executed on a computer, causes the computer to perform the method in any possible implementation form of the first aspect.

[0043] In a seventh aspect, the present application provides a computer readable storage medium, which stores a computer program, which, when executed on a computer, causes the computer to perform the method in any possible implementation form of the first aspect.

[0044] In an eighth aspect, the present application provides a chip, comprising a circuit, which is configured to perform the method in any possible implementation form of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0045] FIG. 1 is a functional block diagram of a vehicle according to an embodiment of the present application.

[0046] FIG. 2 is a schematic block diagram of an intelligent driving system according to an embodiment of the present application.

[0047] FIG. 3 is a schematic flowchart of an intelligent driving method according to an embodiment of the present application.

[0048] Figure 4 is a schematic diagram of a vehicle acquiring information about its surrounding environment according to an embodiment of this application.

[0049] Figure 5 is a schematic diagram of sampling in a European space provided in an embodiment of this application.

[0050] Figure 6 is a schematic diagram of filtering the first set of sampling points provided in an embodiment of this application.

[0051] Figure 7 is a schematic diagram of the scenario in which the vehicle is located, as provided in an embodiment of this application.

[0052] Figure 8 is a schematic diagram of triangulation mapping and search based on road elements provided in an embodiment of this application.

[0053] Figure 9 is a schematic diagram of clustering multiple trajectories in an embodiment of this application.

[0054] Figure 10 is another schematic flowchart of the intelligent driving method provided in this application.

[0055] Figure 11 is a schematic block diagram of the intelligent driving device provided in this application. Detailed Implementation

[0056] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B; "and / or" in this document is merely a description of the association relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. "At least one" refers to one or more. For example, "at least one of A and B," similar to "A and / or B," describes the association relationship between related objects, indicating that three relationships can exist. For example, at least one of A and B can represent: A existing alone, A and B existing simultaneously, and B existing alone.

[0057] The prefixes such as "first" and "second" used in this application embodiment are merely for distinguishing different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. The use of ordinal numbers and other prefixes used to distinguish descriptive objects in this application embodiment does not constitute a limitation on the described objects. The description of the described objects is given in the claims or the context of the embodiments, and should not constitute unnecessary restrictions due to the use of such prefixes. Furthermore, in the description of this embodiment, unless otherwise stated, "multiple" means two or more.

[0058] FIG. 1 is a functional block diagram of a vehicle 100 according to an embodiment of the present application. The vehicle 100 can include a perception system 110, a computing platform 120, and a display device 130. The perception system 110 can include one or more sensors that sense information about the environment surrounding the vehicle 100. For example, the perception system 110 can include a positioning system, which can be a global positioning system (GPS), a Beidou system, or another positioning system. For another example, the perception system 110 can include one or more of an inertial measurement unit (IMU), an acceleration sensor, a laser radar, a millimeter wave radar, an ultrasonic radar, and a camera.

[0059] Some or all functions of the vehicle 100 can be controlled by the computing platform 120. The computing platform 120 can include one or more processors, such as processors 121 through 12n (n is a positive integer), which are circuits having a processing capability of signals. In one implementation, the processors can be circuits having an instruction reading and running capability, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a kind of microprocessor), a digital signal processor (DSP), or the like. In another implementation, the processors can be circuits having a certain function implemented by a logic relationship of hardware circuits, which is fixed or reconfigurable. For example, the processors can be hardware circuits implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as a field programmable gate array (FPGA). In the reconfigurable hardware circuit, the processor loads a configuration document to implement the hardware circuit configuration. It can be understood that the processor loads instructions to implement the functions of the above units. In addition, the processors can also be hardware circuits designed for artificial intelligence, which can be understood as a kind of ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), or the like. In addition, the computing platform 120 can also include a memory for storing instructions, and some or all of the processors 121 through 12n can call the instructions in the memory to implement corresponding functions.

[0060] The display device 130 in the cabin is mainly divided into two categories, the first category is a vehicle display screen, and the second category is a projection display screen, such as a head up display (HUD). The vehicle display screen is a physical display screen and is an important component of the in-vehicle infotainment system. Multiple display screens can be provided in the cabin, such as a digital instrument display screen, a center control screen, a display screen in front of a passenger (also referred to as a front passenger) at a co-driver position, a display screen in front of a left rear passenger, and a display screen in front of a right rear passenger, or even a vehicle window can be used as a display screen for display. The head up display, also known as a head-up display system, is mainly used for displaying driving information such as speed, navigation, etc. on a display device (such as a windshield) in front of the driver. This reduces the time for the driver to change his line of sight and avoids changes in the pupil caused by the driver changing his line of sight, thereby improving driving safety and comfort. The HUD includes, for example, a combiner-HUD (C-HUD) system, a windshield-HUD (W-HUD) system, and an augmented reality HUD (AR-HUD). It should be understood that other types of systems can also appear as the technology evolves, and the present application does not limit this.

[0061] The above display device 130 is described by taking the vehicle display screen and the projection display screen as examples, and embodiments of the present application are not limited thereto. For example, the display device 130 can also be a light display screen or a projection screen.

[0062] Optionally, the structure of the vehicle 100 described above is only schematic, and in actual applications, various components in the vehicle 100 described above can be added or deleted according to actual needs.

[0063] The vehicle 100 can include an intelligent driving system, which can include an advanced driving assistant system (ADAS) and an autonomous driving system (ADS). The intelligent driving system uses various sensors (including but not limited to laser radar, millimeter wave radar, camera, ultrasonic sensor, global positioning system, and inertial measurement unit) on the vehicle to obtain information from the surroundings of the vehicle, and analyzes and processes the obtained information to realize functions such as obstacle perception, target recognition, vehicle positioning, path planning, driver monitoring / reminding, etc., thereby improving the safety, automation level, and comfort of vehicle driving.

[0064] For example, FIG. 2 shows a schematic block diagram of an intelligent driving system according to an embodiment of the present application. The intelligent driving system can include three functional modules: a perception module 210, a planning module 220, and a control module 230, wherein the perception module 210 perceives the environment around the vehicle body through sensors and outputs corresponding perception data to the planning module 220. The planning module 220 obtains information of road elements according to the information obtained by the perception module 210. The planning module 220 can determine the physical connectivity of the vehicle from the current position to the sampling point based on the current position of the vehicle and the information of the road elements, and plan a driving trajectory of the vehicle from the current position to the sampling point when the vehicle is physically connected from the current position to the sampling point. The planning module 220 can determine the strategy space of the vehicle according to the driving trajectory. The planning module 220 can send the strategy space to the control module 230. The control module 230 can evaluate in the Euclidean space based on the strategy space, thereby making behavior decisions or interaction decisions of the vehicle.

[0065] The above perception module 210 can be the perception system 110 described above, and the planning module 220 and the control module 230 can be located in the computing platform 120 described above.

[0066] The degree to which a vehicle's driving automation system is capable of performing dynamic driving tasks is divided into levels 0 to 5 (or L0-L5) according to the role allocation in performing dynamic driving tasks and the presence or absence of operational design domain (ODD) limits, i.e., external conditions suitable for the functional operation of the driving automation system as determined when the system is designed, such as roads, traffic, weather, lighting, etc. Among the six levels of driving automation, levels 0-2 are driving assistance, and the system assists humans in performing dynamic driving tasks, and the driving subject is still the driver. Levels 3-5 are autonomous driving, and the system replaces humans to perform dynamic driving tasks under the design operating conditions, and when the function is activated, the driving subject is the system. The names and definitions of each level are as follows:

[0067] A level 0 driving automation (may also be referred to as emergency assistance) system is not capable of sustained driving task execution of vehicle lateral or longitudinal motion control, but has the capability for partial goal and event detection and response in dynamic driving tasks. A level 1 driving automation (may also be referred to as partial driver assistance) system is capable of sustained driving task execution of vehicle lateral or longitudinal motion control, and has the capability for partial goal and event detection and response commensurate with the vehicle lateral or longitudinal motion control executed, under all conditions for which it is designed to operate. A level 2 driving automation (may also be referred to as combined driver assistance) system is capable of sustained driving task execution of vehicle lateral and longitudinal motion control, and has the capability for partial goal and event detection and response commensurate with the vehicle lateral and longitudinal motion control executed, under all conditions for which it is designed to operate. A level 3 driving automation (may also be referred to as conditionally automated driving) system is capable of sustained driving task execution of all dynamic driving tasks under all conditions for which it is designed to operate. A level 4 driving automation (may also be referred to as highly automated driving) system is capable of sustained driving task execution of all dynamic driving tasks and the automated execution of a minimal risk strategy under all conditions for which it is designed to operate. A level 5 driving automation (may also be referred to as fully automated driving) system is capable of sustained driving task execution of all dynamic driving tasks and the automated execution of a minimal risk strategy under all conditions for which it is designed to operate. Generally, intelligent driving systems are generally L2-L5, such as ADAS is L2, and ADS is L3-L5.

[0068] As described previously, in the field of autonomous driving, the behavior decision that a vehicle can make relies on the lane topology information in the HD map. On the one hand, the cost of using the HD map is high; on the other hand, the real-time performance of the HD map is poor, and if a real change scene such as road construction is encountered, a vehicle in an autonomous driving state can have lane selection errors, be stuck, and the like. This can cause the driving safety of a user to be affected.

[0069] Embodiments of the present application provide an intelligent driving method and device and a vehicle. The vehicle can make a behavior decision without relying on an HD map and a lane topology relationship, and a vehicle in an autonomous driving state makes a decision plan without a map. In this way, the problem of vehicle decision errors due to poor real-time performance of the HD map can be avoided, and the driving safety of a user can be improved.

[0070] In addition, if the generation of the strategy space depends on the lane topology relationship, this can cause the strategy space of the vehicle in the autonomous driving state to jump or be missing, thereby causing the vehicle to have problems such as degradation, unexpected steering, and the like. In the embodiment of the present application, since the lane topology relationship is not relied on, this can avoid the strategy space of the vehicle from jumping or being missing, help to avoid problems such as degradation or unexpected steering of the vehicle, and help to improve the driving experience and driving safety of the user.

[0071] For example, the strategy space of the vehicle can be understood as all behaviors that the vehicle can make, for example, including but not limited to one or more of the following:

[0072] (1) lane center keeping;

[0073] (2) in-lane obstacle avoidance;

[0074] (3) lane borrowing obstacle avoidance;

[0075] (4) lane changing to a neighboring lane (including cut-in behavior);

[0076] (5) cross-lane lane changing.

[0077] For example, FIG. 3 shows a schematic flowchart of the intelligent driving method 300 provided by the embodiment of the present application. The method 300 includes:

[0078] S310, obtaining information of one or more road elements around the vehicle.

[0079] For example, the one or more road elements can include lane lines, road boundaries, road edges, flower beds, fences, and the like.

[0080] For example, FIG. 4 shows a schematic diagram of the vehicle obtaining surrounding environment information provided by the embodiment of the present application. The vehicle 100 can determine, through data collected by the sensor, that the road elements around the vehicle 100 include road boundaries, lane lines (for example, dashed lines), and stationary obstacles.

[0081] S320, sampling in the Euclidean space to generate a first set of sampling points.

[0082] Optionally, generating a plurality of sampling points in the Euclidean space includes: generating a first set of sampling points in the Euclidean space according to a road trend; or generating a first set of sampling points in the Euclidean space according to a kinematic model of the vehicle.

[0083] For example, FIG. 5 shows a schematic diagram of sampling in the Euclidean space provided by the embodiment of the present application.

[0084] As shown in FIG. 5, the vehicle 100 can generate a first set of sampling points in the Euclidean space based on the road trend, the first set of sampling points including sampling points 1-25. For example, if the current road trend is straight driving, the vehicle can determine a first distance (e.g., 100 m) based on the current speed (e.g., 60 kph) and a preset time length (e.g., 6 s), and generate the sampling points 1-25 at a distance of 100 m from the front of the vehicle.

[0085] S330, screening the first set of sampling points to obtain a second set of sampling points.

[0086] Optionally, the screening of the first set of sampling points to obtain the second set of sampling points comprises: generating a rough sampling line (curve) from the current position of the vehicle to each sampling point in the first set of sampling points; and screening the first set of sampling points according to the kinematic legality of the vehicle and the rough sampling line corresponding to each sampling point to obtain the second set of sampling points.

[0087] For example, FIG. 6 shows a schematic diagram of the screening of the first set of sampling points according to an embodiment of the present application.

[0088] As shown in FIG. 6, the vehicle 100 can generate a rough sampling line (curve) from the current position to the sampling points 1-25. The sampling points in the first set of sampling points are screened based on the kinematic legality of the vehicle 100. For example, the vehicle 100 determines that there is a safety risk when the vehicle 100 changes lanes across multiple lanes based on the current speed of the vehicle 100 and the information of the lane lines, so that the sampling points 1-5 and the sampling points 21-25 can be screened out, and a second set of sampling points including the sampling points 6-20 is obtained. The sampling lines corresponding to the sampling points 21-25 are unreasonable sampling lines.

[0089] For example, before a certain urban intersection, there is a situation of 3 lanes changing to 5 lanes. If the vehicle needs to enter the right-turn lane before the intersection, it needs to change lanes continuously for two times in a short time, so the sampling points across two lanes need to be retained.

[0090] For example, the vehicle 100 can screen the sampling points according to the kinematic legality of the vehicle 100. For example, the sampling points can be screened by the following steps (1) and (2):

[0091] (1) According to the Ackerman model or the bicycle model, starting from the current pose and motion state of the vehicle, the area that the vehicle can reach is deduced (without considering collision, only subject to the Ackerman model or the bicycle model).

[0092] (2) If the sampling point is within the area deduced in (1), it can be considered that the sampling point meets the kinematic legality of the vehicle.

[0093] S340, determining, according to the information of the one or more road elements, physical connectivity of the vehicle from the current position to each sampling point in the second set of sampling points.

[0094] Optionally, the physical connectivity can be used to indicate spatial reachability of the vehicle from the current position to the sampling point, and the spatial reachability includes spatially reachable or spatially unreachable.

[0095] For example, if there is a path between the current position and the sampling point without colliding with static obstacles and without seriously violating traffic rules, the vehicle 100 can determine that the vehicle 100 is physically connected from the current position to the sampling point, or, the vehicle 100 is spatially reachable from the current position to the sampling point.

[0096] For example, if the path searched from the current position to the sampling point collides with static obstacles or seriously violates traffic rules, the vehicle 100 can determine that the vehicle 100 is not physically connected from the current position to the sampling point, or, the vehicle 100 is not spatially reachable from the current position to the sampling point.

[0097] Optionally, the physical connectivity can be represented by the searched path or can be represented by semantics.

[0098] For example, FIG. 7 shows a schematic diagram of a scenario in which the vehicle 100 is located.

[0099] As shown in (a) of FIG. 7, the physical connectivity of the vehicle 100 from the current position to the sampling point 11 indicates that the vehicle 100 is spatially reachable from the current position to the sampling point 11, and the physical connectivity can be indicated by two searched paths, where path 1 indicates that the vehicle 100 can reach the sampling point 11 from the current position by in-lane obstacle avoidance, and path 2 indicates that the vehicle 100 can first change lanes to the right side (or the wheels of the vehicle 100 can first press the right dashed line), and then drive to the left to reach the sampling point 11.

[0100] Alternatively, the physical connectivity can be expressed by semantics. The physical connectivity of the vehicle 100 from the current position to the sampling point 11 can be expressed by the following semantics:

[0101] (1) reach the sampling point 11 after in-lane obstacle avoidance to the left side;

[0102] (2) first change lanes to the right side, and then change lanes to the left side to reach the sampling point 11.

[0103] As shown in (b) of FIG. 7, the information of one or more road elements around the vehicle 100 includes information of a fence, the vehicle 100 is currently located at the left side of the fence, and the sampling points 16-20 are located at the right side of the fence. The physical connectivity of the vehicle 100 from the current location to any one of the sampling points 16-20 indicates that the vehicle 100 is not spatially reachable from the current location to any one of the sampling points 16-20.

[0104] As shown in (a) of FIG. 7, based on the physical connectivity analysis, it can be determined that the vehicle 100 is physically connected (or, spatially reachable) from the current location to any one of the sampling points 6-20.

[0105] As shown in (b) of FIG. 7, based on the physical connectivity analysis, it can be determined that the vehicle 100 is physically connected (or, spatially reachable) from the current location to any one of the sampling points 6-15, and is not physically connected (or, spatially reachable) from the current location to any one of the sampling points 16-20.

[0106] Optionally, the physical connectivity of the vehicle from the current location to each sampling point in the second set of sampling points is determined according to the information of the one or more road elements, including: the physical connectivity of the vehicle from the current location to each sampling point in the second set of sampling points is determined according to a path from the current location to each sampling point in the second set of sampling points.

[0107] For example, the first sampling point can be any one of the sampling points 6-15 shown in (b) of FIG. 7.

[0108] For example, the second sampling point can be any one of the sampling points 16-20 shown in (b) of FIG. 7.

[0109] The above describes the determination of the physical connectivity of the vehicle from the current location to each sampling point by whether a path from the current location to each sampling point can be searched. The following describes some possible implementation manners of searching the path from the current location to each sampling point.

[0110] For example, the vehicle 100 can determine whether a path from the current location to each sampling point can be searched according to a path searching algorithm.

[0111] Exemplarily, a path search algorithm can be used to search a path from the vehicle position to each sample point. A line segment where each road element is located can be assigned with a corresponding cost value. For example, the cost value of the edge corresponding to the obstacle, the road boundary and the curb can be assigned with infinity, so that the vehicle will not cross the obstacle, the road boundary and the curb in subsequent path search. The cost value of the edge corresponding to the dashed line can be smaller, and the cost value of the edge corresponding to the solid line can be larger. The setting of these cost values will make the vehicle 100 search a reasonable path in subsequent path search.

[0112] The above path search algorithm can include a geometric method, an A* algorithm, a hybrid A* algorithm, a rapid-exploration random tree (RRT) algorithm, etc.

[0113] Optionally, determining whether a path from the current position of the vehicle to each sample point in the second sample point set can be searched includes: determining a plurality of triangular regions according to the one or more road elements; performing cost mapping on the edges of each triangular region in the plurality of triangular regions; and determining whether a path from the current position to each sample point can be searched according to the triangular regions after cost mapping.

[0114] Exemplarily, FIG. 8 shows a schematic diagram of road element-based triangulation mapping and search provided by an embodiment of the present application.

[0115] Exemplarily, as shown in (a) of FIG. 8, the obstacle, the road boundary, the lane dashed line boundary, the lane solid line boundary and the curb compose all the road elements.

[0116] Exemplarily, as shown in (b) of FIG. 8, the left road boundary can be composed of points G, H, I, J, K and L, the right road boundary can be composed of points M, N, O, P and Q, the curb can be composed of points A, B and C, the solid line can be composed of points C, D and E, and the dashed line can be composed of points E and F.

[0117] Exemplarily, the distance between two adjacent points on each road element can be (2m, 6m).

[0118] Exemplarily, the distance between two adjacent points (for example, points G and H) on the left and right road boundaries can be 5m.

[0119] Exemplarily, the distance between two adjacent points (for example, points A and B) on the curb can be 3m.

[0120] Exemplarily, the distance between two adjacent points (for example, points C and D) on the solid line can be 2.5m.

[0121] For example, the distance between two adjacent points (e.g., point E and point F) of the dashed line can be 5 m.

[0122] For example, the obstacle can be a cone barrel, and the corresponding points of the cone barrel can be the vertices of the pentagonal area where the cone barrel is located, such as point R, point S, point T, point U, and point V.

[0123] The above division of points on each road element is only illustrative. For example, when the length of the guardrail is 6 m, the guardrail can be divided into three points according to the preset distance, and the distance between two adjacent points is 3 m. Alternatively, the guardrail can be divided into three points, such as the distance between point A and point B is 3.5 m, and the distance between point B and point C is 2.5 m.

[0124] For example, the obstacle can be arranged in a pentagonal area, or it can be arranged in a quadrilateral or other polygonal area, and the embodiments of the present application do not make specific limitations.

[0125] Optionally, a plurality of environment triangles are generated, including: based on a triangle segmentation algorithm, triangulating the plurality of points corresponding to the road element to obtain a plurality of triangular regions.

[0126] For example, the triangle segmentation algorithm can be a constrained Delaunay triangulation (CDT) algorithm.

[0127] For example, through the plurality of points (such as points A to V) corresponding to the road element, a triangular graph can be constructed through the CDT algorithm, and the triangular graph includes a plurality of triangular regions as shown in (b) of FIG. 8.

[0128] For example, (b) of FIG. 8 shows a plurality of triangular graph regions after triangulation, and each line segment in the road element corresponds to an edge of the triangular region in the triangular graph. Therefore, the triangulation mapping describes the physical environment around the vehicle without precision loss. After mapping, according to the types of the road elements, the edges of the corresponding triangular regions are assigned with corresponding cost values. For example, the cost values of the edges of the triangular regions corresponding to the obstacles, the road boundaries, and the guardrails can be assigned with infinite values, so that the vehicle will not cross the obstacles, the road boundaries, and the guardrails in subsequent path searching. The cost of the triangular edge corresponding to the dashed line can be small, and the cost of the triangular edge corresponding to the solid line can be large. The cost of the triangular edge of the non-road element can be zero. The setting of these costs will guide the search algorithm to search a reasonable path in subsequent searching.

[0129] As shown in (b) of FIG. 8, given a triangulated map describing the physical environment, given the current position of the vehicle as the starting point and given the target point, a path can be searched from the current position to the target point by A* algorithm, and the path is shown as the path points (block points) shown in (b) of FIG. 8. Each expansion node (such as path points 1-7) obtained by the A* algorithm is on the edge of the triangular region. In the process of node expansion, a suitable expansion node can be searched by cost and heuristic value, for example, the cost can include one or more of the distance cost of the expansion node from the starting point, the cost of crossing the edge of the triangular region, or the cost generated by whether the kinematics is in compliance. The heuristic value includes the distance between the expansion node and the target point. As shown in (b) of FIG. 8, the target point (for example, each sampling point in the second sampling point set) is on the adjacent lane, although the crossing cost of the dashed line is small, the search result will give the path from the solid line (for example, path point 4 is located on the solid line), because the kinematic model of the vehicle is considered during the search, and if the path is crossed from the dashed line, a backtracking behavior will be generated, which does not comply with the physical motion rule of the vehicle.

[0130] In the embodiments of the present application, the vehicle does not need to obtain the lane center line and the lane topological relationship when performing path search, but performs path search through a plurality of triangular regions determined by road elements existing in the real physical world, which helps to solve the problem of inaccurate path search caused by unreasonable lane center line and lane topological relationship, thereby helping to improve the driving safety of users.

[0131] Optionally, the vehicle 100 can determine whether a path from the current position to each sampling point in the second sampling point set can be searched based on a gridding method.

[0132] For example, the gridding method can be to determine whether a path from the current position to each sampling point can be generated by an occupancy grid map.

[0133] S350, if the physical connectivity indicates that the vehicle is physically connected from the current position to at least part of the sampling points in the second sampling point set, a driving trajectory from the current position to the at least part of the sampling points is planned.

[0134] Optionally, if the vehicle is physically connected from the current position to at least part of the sampling points in the second sampling point set in S340, the vehicle 100 can obtain the boundary information of each sampling point in the at least part of the sampling points as a constraint for generating the driving trajectory.

[0135] Optionally, planning the driving trajectory from the current position to the at least part of the sampling points comprises: planning the driving trajectory from the current position to the at least part of the sampling points according to the boundary information.

[0136] For example, according to the boundary information, a driving trajectory from the current position to at least part of the sampling points is planned, including: according to the boundary information and the FEM, a FEM line from the current position to at least part of the sampling points is planned. The FEM line can be used as an implementation of the driving trajectory, and the application does not limit the manner of obtaining the driving trajectory.

[0137] Optionally, taking the triangulation method shown in FIG. 8 as an example, after the path from the current position to the sampling point is searched, a smooth driving trajectory (which can include multiple reference points) can be generated by solving a quadratic optimization problem with constraints (for example, the above boundary information). In the optimization problem, the initial guessed reference points can be the path points obtained by the path searching algorithm, the left and right soft boundaries of the path points of the vehicle are formed by the left and right lane lines, and the hard boundary is formed by the boundary formed by the road boundary or the obstacle. The target cost function can include one or more of the smoothness cost of the driving trajectory, the distance cost between the reference points, and the length cost of the entire driving trajectory.

[0138] For example, during the process in which the vehicle selects a driving trajectory and drives along the driving trajectory, if a dynamic obstacle (for example, a pedestrian is suddenly detected) is detected, the driving trajectory for avoiding the dynamic obstacle can be re-planned in the homotopy space (or the connected region). For example, as shown in (b) of FIG. 8, the homotopy space (or the connected region) around the driving trajectory obtained after the secondary optimization can include the regions where △ABG, △BGH, △BCH, △CHI, △CDI, △CDO, △DOP, △DEP and △EPQ are located.

[0139] S360, clustering the driving trajectories to obtain a strategy space of the vehicle.

[0140] For example, the driving trajectory of the vehicle from the current position to each sampling point can be obtained through S360.

[0141] For example, FIG. 9 shows a schematic diagram of clustering multiple trajectories in the application.

[0142] As shown in FIG. 9, for the scenario as shown in (b) of FIG. 7, the vehicle 100 can obtain the driving trajectory 3-12 based on S360 described above, and the strategy space can include a plurality of behavior clusters, such as left lane changing, keeping in the self-lane after passing the left dashed line to bypass the obstacle, and keeping in the self-lane after passing the right dashed line to bypass the obstacle, wherein the driving trajectory 3-7 can be divided into left lane changing; the driving trajectory 8-10 can be divided into keeping in the self-lane after passing the left dashed line to bypass the obstacle; and the driving trajectory 11 and the driving trajectory 12 can be divided into keeping in the self-lane after passing the right dashed line to bypass the obstacle.

[0143] For example, after generating the strategy space, the vehicle 100 can perform evaluation based on the generated strategy space, and make behavior decisions (e.g., whether to change lanes) and interaction decisions (e.g., whether to yield or pass the dynamic obstacle) in the Euclidean space.

[0144] FIG. 10 shows a schematic flowchart of the intelligent driving method 1000 provided by the present application. The method 1000 includes:

[0145] S1010, obtaining information of road elements around a vehicle from data collected by sensors of the vehicle, the vehicle being at a first position.

[0146] Optionally, the road elements include obstacles and isolated road elements. For example, the obstacles include one or more of a road edge, a median strip, a flower bed, and a fence; and the isolated road elements include one or more of a lane line and a road boundary.

[0147] Optionally, before obtaining the information of the road elements around the vehicle from the data collected by the sensors of the vehicle, the method further includes determining that the vehicle is in an intelligent driving state.

[0148] For example, when the vehicle detects that a user has turned on an intelligent driving function, it can be determined that the vehicle is in an intelligent driving state.

[0149] For example, the intelligent driving function includes a navigation cruise assist (NCA) function, an adaptive cruise control (ACC) function, an automated lane keeping (ALK) function, an L3-level high-speed expressway automatic driving function, an L3-level urban trunk road automatic driving function, or an automated valet parking (AVP) function.

[0150] S1020, determining physical connectivity of the vehicle from the first position to each sampling point in a plurality of sampling points according to the first position and the information of the road elements.

[0151] Optionally, before determining the physical connectivity of the vehicle from the first position to each of the plurality of sampling points, the method 1000 further comprises: when there is information of road boundary in the road element, determining the plurality of sampling points according to road trend of the road where the vehicle is located; or when there is no information of road boundary in the road element, determining the plurality of sampling points according to the kinematic model of the vehicle.

[0152] Optionally, the plurality of sampling points are determined according to the road trend of the road where the vehicle is located, comprising: determining a first set of sampling points according to the road trend of the road where the vehicle is located; and determining the plurality of sampling points from the first set of sampling points according to kinematic legality of the vehicle.

[0153] Optionally, the plurality of sampling points are determined according to the kinematic model of the vehicle, comprising: determining a first set of sampling points according to the kinematic model of the vehicle; and determining the plurality of sampling points from the first set of sampling points according to kinematic legality of the vehicle.

[0154] The plurality of sampling points can be the sampling points in the second set of sampling points.

[0155] In the embodiments of the present application, the decision result of the vehicle is no longer dependent on the lane topology relationship, but is based on the road trend or the kinematic model of the vehicle to sample (for example, sample in the Euclidean space). In this way, it can be ensured that the strategy space generated according to the driving trajectory is the full space, and cannot be sampled due to the existence of topological separation relationship.

[0156] Optionally, the physical connectivity can be represented by a path search result, or can also be represented by semantics.

[0157] Optionally, the plurality of sampling points include a first sampling point, and the physical connectivity of the vehicle from the first position to each of the plurality of sampling points is determined according to the path search result, comprising: when a path from the first position to the first sampling point is searched, determining that the vehicle is physically connected from the first position to the first sampling point.

[0158] The above vehicle physically connected from the first position to the first sampling point can also be understood as the vehicle being spatially reachable from the first position to the first sampling point.

[0159] Optionally, the plurality of sampling points include a second sampling point, and the physical connectivity of the vehicle from the first position to each of the plurality of sampling points is determined according to the path search result, comprising: when a path from the first position to the second sampling point cannot be searched, determining that the vehicle is not physically connected from the first position to the second sampling point.

[0160] Optionally, the determining the physical connectivity of the vehicle from the first position to each of the plurality of sampling points comprises: searching a path from the first position to each of the plurality of sampling points according to the first position and the information of the road element, to obtain a path search result; and determining the physical connectivity of the vehicle from the first position to each of the plurality of sampling points according to the path search result.

[0161] In the embodiments of the present application, by perceiving the real-time detected road element, a path from the current position of the vehicle to each sampling point can be searched, so that the physical connectivity of the vehicle to each sampling point can be determined based on the path search result. In this way, some sampling points that cannot be reached from the current position can be accurately excluded, avoiding the vehicle from making a wrong choice and collision accidents, and helping to improve the driving safety of the user.

[0162] Optionally, the searching the path from the first position to each of the plurality of sampling points comprises: determining a plurality of triangular regions according to the information of the road element; performing cost mapping on the edges of each triangular region in the plurality of triangular regions; and searching the path from the first position to each of the plurality of sampling points according to the triangular regions after cost mapping.

[0163] The above process of path searching by triangular regions can refer to the path planning process shown in FIG. 8 and described in the above embodiments, which will not be described here again.

[0164] S1030, if the physical connectivity indicates that the vehicle is physically connected from the first position to at least part of the plurality of sampling points, planning a driving trajectory of the vehicle to drive to the at least part of the plurality of sampling points.

[0165] The at least part of the plurality of sampling points can be the sampling points in the third set of sampling points.

[0166] Optionally, the planning the driving trajectory of the vehicle to drive to the at least part of the plurality of sampling points comprises: determining constraints for generating the driving trajectory according to the information of the road element; and planning the driving trajectory of the vehicle to drive to the at least part of the plurality of sampling points according to the constraints for generating the driving trajectory.

[0167] For example, when determining the physical connectivity of the vehicle from the current position to each sampling point, the vehicle can also extract the corresponding boundary information as the constraint for generating the driving trajectory. For example, the vehicle can model the road element in the Euclidean space, calculate the path from the current position of the vehicle to each sampling point, and extract the corresponding boundary information as the constraint for generating the FEM line. In this way, the generated driving trajectory is more human-like, and is no longer linearly dependent on the lane topology.

[0168] S1040, control the vehicle according to the driving trajectory.

[0169] Optionally, the control of the vehicle according to the driving trajectory comprises: dividing the driving trajectory into one or more behavior clusters to obtain the strategy space; and controlling the vehicle according to the strategy space.

[0170] In the embodiments of the present application, the vehicle can cluster the driving trajectory to obtain the strategy space. For example, the physical connectivity can be expressed by semantics: change, keep.

[0171] For example, as shown in FIG. 9, keep can be further divided into keep in the self lane after changing the lane to the left of the virtual line to bypass the obstacle and keep in the self lane after changing the lane to the right of the lane line to bypass the obstacle.

[0172] Optionally, the dividing of the driving trajectory into one or more behavior clusters to obtain the strategy space comprises: clustering the driving trajectory according to the physical connectivity, and dividing the driving trajectories with the same physical connectivity boundary into the same cluster. In this way, in the strategy space evaluation, the behavior of the vehicle is no longer dependent on the lane topology relationship, but the actual behavior that can be performed by the vehicle, so that the human-like behavior such as long-time lane pressing can be made, and various intelligent driving functions can be executed in the area without road topology.

[0173] In the embodiments of the present application, the vehicle can determine the physical connectivity of the vehicle to the plurality of sampling points according to the information of the position and the road element. If the physical connectivity indicates that the vehicle is physically connected to at least part of the sampling points, the driving trajectory to at least part of the sampling points can be planned, so as to control the vehicle to drive. In this way, the vehicle can make decision and planning through the information of the road element obtained by real-time perception, without relying on the high-precision map and the lane topology relationship, so as to avoid the problems such as wrong lane selection and stop, and help to improve the driving safety of the user.

[0174] The method 300 and the method 1000 can be executed by the vehicle 100 described above; or the method 300 and the method 1000 can be executed by the computing platform 120 described above; or the method 300 and the method 1000 can be executed by the processor, the circuit or the chip in the computing platform 120 described above; or the method 300 and the method 1000 can be executed by the intelligent driving system described above; or the method 300 and the method 1000 can be executed by the planning module 220 described above.

[0175] FIG. 11 shows a schematic block diagram of an intelligent driving device 1100 provided by the present application. The device 1100 comprises: an acquisition unit 1110 configured to acquire information of road elements around a vehicle from data collected by sensors of the vehicle, the vehicle being at a first position; a determination unit 1120 configured to determine physical connectivity of the vehicle from the first position to each of a plurality of sampling points according to the first position and the information of the road elements; a trajectory planning unit 1130 configured to plan a driving trajectory of the vehicle to drive to at least part of the plurality of sampling points when the physical connectivity indicates that the vehicle is physically connected from the first position to the at least part of the plurality of sampling points; and a control unit 1140 configured to control the vehicle according to the driving trajectory.

[0176] Optionally, the device further comprises a path searching unit configured to search a path from the first position to each of the plurality of sampling points according to the first position and the information of the road elements to obtain a path search result; and the determination unit 1120 is specifically configured to determine the physical connectivity of the vehicle from the first position to each of the plurality of sampling points according to the path search result.

[0177] Optionally, the path searching unit is specifically configured to determine a plurality of triangular regions according to the information of the road elements, perform cost mapping on edges of each of the plurality of triangular regions, and search the path from the first position to each of the plurality of sampling points according to the triangular regions after cost mapping.

[0178] Optionally, the trajectory planning unit 1130 is specifically configured to determine a constraint for generating the driving trajectory according to the information of the road elements, and plan the driving trajectory of the vehicle to drive to the at least part of the plurality of sampling points according to the constraint for generating the driving trajectory.

[0179] Optionally, the determination unit 1120 is further configured to determine the plurality of sampling points according to a road trend of a road where the vehicle is located when information of a road boundary exists in the road elements, or determine the plurality of sampling points according to a kinematic model of the vehicle when information of the road boundary does not exist in the road elements.

[0180] Optionally, the control unit 1140 is specifically configured to divide the driving trajectory into one or more behavior clusters to obtain a strategy space, and control the vehicle according to the strategy space.

[0181] Optionally, the determination unit 1120 is further configured to determine that the vehicle is in an intelligent driving state before the acquisition unit acquires the information of the road elements around the vehicle.

[0182] Optionally, the road elements comprise one or more of static obstacles, lane lines, and road boundaries.

[0183] For example, the acquisition unit 1110 can be the computing platform in FIG. 1 or a processing circuit, a processor or a controller in the computing platform. Taking the acquisition unit 1110 as the processor 121 in the computing platform as an example, the processor 121 can acquire the data collected by the perception system 110 of the vehicle and acquire the information of the road element based on the data.

[0184] For another example, the determination unit 1120 can be the computing platform in FIG. 1 or a processing circuit, a processor or a controller in the computing platform. Taking the determination unit 1120 as the processor 122 in the computing platform as an example, the processor 122 can determine the physical connectivity of the vehicle from the current position to each sampling point based on the information of the road element determined by the processor 121 and the current position of the vehicle.

[0185] For another example, the trajectory planning unit 1130 can be the computing platform in FIG. 1 or a processing circuit, a processor or a controller in the computing platform. Taking the trajectory planning unit 1130 as the processor 123 in the computing platform as an example, when the processor 122 determines the physical connectivity from the current position of the vehicle to at least part of the sampling points, the processor 123 can plan a driving trajectory from the current position to at least part of the sampling points.

[0186] For another example, the control unit 1140 can be the computing platform in FIG. 1 or a processing circuit, a processor or a controller in the computing platform. Taking the control unit 1140 as the processor 124 in the computing platform as an example, the processor 124 can control the vehicle to drive based on the driving path planned by the processor 123.

[0187] The functions implemented by the acquisition unit 1110, the functions implemented by the determination unit 1120, the functions implemented by the trajectory planning unit 1130 and the functions implemented by the control unit 1140 can be implemented by different processors, or can be implemented by the same processor, or part of the functions can be implemented by the same processor, and the embodiments of the present application do not limit this.

[0188] It should be understood that the division of units in the above apparatus is only a logical functional division, and all or part of them can be integrated into a physical entity or physically separated when actually implemented. In addition, the units in the apparatus can be implemented in the form of processor calling software; for example, the apparatus includes a processor connected with a memory, and the memory stores instructions, and the processor calls the instructions stored in the memory to implement any of the above methods or realize the functions of the units of the apparatus, wherein the processor is, for example, a general processor such as a CPU or a microprocessor, and the memory is a memory in the apparatus or a memory outside the apparatus. Alternatively, the units in the apparatus can be implemented in the form of hardware circuit, and the functions of part or all of the units can be realized by the design of the hardware circuit, which can be understood as one or more processors; for example, in one implementation, the hardware circuit is an ASIC, and the functions of part or all of the units are realized by the design of the logical relationship of elements in the circuit; for example, in another implementation, the hardware circuit is a PLD, and taking FPGA as an example, it can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by a configuration file, so as to realize the functions of part or all of the units. All units of the above apparatus can be implemented in the form of processor calling software, or all units can be implemented in the form of hardware circuit, or part of the units are implemented in the form of processor calling software, and the remaining part is implemented in the form of hardware circuit.

[0189] In the embodiments of the present application, the processor is a circuit with signal processing capability. In one implementation, the processor can be a circuit with instruction reading and running capability, such as CPU, microprocessor, GPU, or DSP, etc. In another implementation, the processor can realize certain functions through the logical relationship of the hardware circuit, and the logical relationship of the hardware circuit is fixed or can be reconfigured, such as ASIC or PLD implemented hardware circuit, such as FPGA. In the reconfigurable hardware circuit, the processor loads the configuration document to realize the hardware circuit configuration process, which can be understood as the process of the processor loading instructions to realize the functions of part or all of the units. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as a kind of ASIC, such as NPU, TPU, DPU, etc.

[0190] It can be seen that each unit in the above apparatus can be one or more processors (or processing circuits) configured to implement the above method, such as CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0191] In addition, all or part of each unit in the above apparatus can be integrated together or can be independently implemented. In one implementation, the units are integrated together to be implemented in the form of a SoC. The SoC can include at least one processor for implementing any of the above methods or the functions of the units of the apparatus, and the at least one processor can be of different types, such as a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, and the like.

[0192] Embodiments of the present application also provide an intelligent driving apparatus, which includes a processing unit and a storage unit, wherein the storage unit is configured to store instructions, and the processing unit is configured to execute the instructions stored in the storage unit, so that the apparatus executes the method or steps performed by the above embodiments.

[0193] Optionally, if the intelligent driving apparatus is located in a vehicle, the processing unit can be one or more of the processors 121-12n shown in FIG. 1.

[0194] Embodiments of the present application also provide an intelligent driving system, which includes a perception system and a computing platform, and the computing platform includes the above intelligent driving apparatus.

[0195] Embodiments of the present application also provide a vehicle, which can include the above intelligent driving apparatus or the above intelligent driving system.

[0196] Embodiments of the present application also provide a computer program product, which includes computer program code, and when the computer program code is run on a computer, the computer is caused to execute the method in the above embodiments.

[0197] Embodiments of the present application also provide a computer readable medium, which stores program code, and when the computer program code is run on a computer, the computer is caused to execute the method in the above embodiments.

[0198] Embodiments of the present application also provide a chip, which includes a circuit for executing the method in the above embodiments.

[0199] In the implementation process, each step of the above method can be completed by the integrated logic circuit of hardware in the processor or the instruction in the form of software. The method disclosed in the embodiments of the present application can be directly embodied as a hardware processor for execution, or a combination of hardware and software modules in the processor for execution. The software module can be located in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable read-only memory, register, and the like. The storage medium is located in the memory, and the processor reads the information in the memory and combines the hardware to complete the steps of the above method. To avoid repetition, it will not be described in detail here.

[0200] It should be understood that the memory in the embodiments of the present application can include read-only memory and random access memory, and provide instructions and data for the processor.

[0201] It should also be understood that the size of the sequence number of each process described above in various embodiments of the present application does not mean the order of execution, and the execution order of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0202] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0203] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0204] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0205] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0206] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0207] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0208] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for intelligent driving, the method comprising: The method comprises: acquiring information of road elements around a vehicle according to data collected by sensors of the vehicle, the vehicle being at a first position; determining physical connectivity of the vehicle from the first position to each of a plurality of sampling points according to the first position and the information of the road elements; if the physical connectivity indicates that the vehicle is physically connected from the first position to at least part of the plurality of sampling points, planning a driving trajectory of the vehicle to the at least part of the plurality of sampling points; controlling the vehicle according to the driving trajectory.

2. The method of claim 1, wherein, The determining of the physical connectivity of the vehicle from the first position to each of the plurality of sampling points according to the first position and the information of the road elements comprises: searching for a path from the first position to each of the plurality of sampling points according to the first position and the information of the road elements to obtain a path search result; determining the physical connectivity of the vehicle from the first position to each of the plurality of sampling points according to the path search result.

3. The method of claim 2, wherein, The searching for the path from the first position to each of the plurality of sampling points comprises: determining a plurality of triangular regions according to the information of the road elements; performing cost mapping on edges of each of the plurality of triangular regions; searching for the path from the first position to each of the plurality of sampling points according to the triangular regions after the cost mapping.

4. The method according to any one of claims 1 to 3, characterized in that, The planning of the driving trajectory of the vehicle to the at least part of the plurality of sampling points comprises: determining constraints for generating the driving trajectory according to the information of the road elements; planning the driving trajectory of the vehicle to the at least part of the plurality of sampling points according to the constraints for generating the driving trajectory.

5. The method according to any one of claims 1 to 4, characterized in that, The method further comprises: when there is information of a road boundary in the road elements, determining the plurality of sampling points according to a road trend of a road where the vehicle is located; or when there is no information of a road boundary in the road elements, determining the plurality of sampling points according to a kinematic model of the vehicle.

6. The method according to any one of claims 1 to 5, characterized in that, The controlling of the vehicle according to the driving trajectory comprises: dividing the driving trajectory into one or more behavior clusters to obtain a strategy space; controlling the vehicle according to the strategy space.

7. The method according to any one of claims 1 to 6, characterized in that, Before the acquiring of the information of the road elements around the vehicle according to the data collected by the sensors of the vehicle, the method further comprises: determining that the vehicle is in an intelligent driving state.

8. The method according to any one of claims 1 to 7, characterized in that, The road elements comprise one or more of a static obstacle, a lane line, and a road boundary.

9. An intelligent driving apparatus, characterized by comprising: The method comprises: an acquiring unit configured to acquire information of road elements around a vehicle according to data collected by sensors of the vehicle, the vehicle being at a first position; a determining unit configured to determine physical connectivity of the vehicle from the first position to each of a plurality of sampling points according to the first position and the information of the road elements; a trajectory planning unit configured to, if the physical connectivity indicates that the vehicle is physically connected from the first position to at least part of the plurality of sampling points, plan a driving trajectory of the vehicle to the at least part of the plurality of sampling points; and a controlling unit configured to control the vehicle according to the driving trajectory. A control unit is configured to control the vehicle according to the driving trajectory.

10. The apparatus of claim 9, wherein, The device further comprises: A path searching unit is configured to search a path from the first position to each of the plurality of sampling points according to the information of the road element and the first position, to obtain a path searching result. The determination unit is specifically configured to: determine the physical connectivity from the first position to each of the plurality of sampling points according to the path searching result.

11. The apparatus of claim 10, wherein, The path searching unit is specifically configured to: determine a plurality of triangular regions according to the information of the road element; perform cost mapping on the edges of each of the plurality of triangular regions; search a path from the first position to each of the plurality of sampling points according to the triangular regions after cost mapping.

12. The apparatus of any one of claims 9-11, wherein, The trajectory planning unit is specifically configured to: determine constraints for generating the driving trajectory according to the information of the road element; plan a driving trajectory for the vehicle to drive to the at least part of the sampling points according to the constraints for generating the driving trajectory.

13. The device according to any one of claims 9 to 12, wherein: the determination unit is further configured to determine the plurality of sampling points according to a road trend of a road where the vehicle is located when the information of the road boundary exists in the road element; or determine the plurality of sampling points according to a vehicle kinematics model when the information of the road boundary does not exist in the road element. The control unit is specifically configured to:

14. The apparatus of any one of claims 9-13, wherein, divide the driving trajectory into one or more behavior clusters to obtain a strategy space; and control the vehicle according to the strategy space.

15. The device according to any one of claims 9 to 14, wherein: the determination unit is further configured to determine that the vehicle is in an intelligent driving state before the acquisition unit acquires the information of the road element around the vehicle. The road element comprises one or more of a static obstacle, a lane line, and a road boundary.

16. The apparatus of any one of claims 9 to 15, wherein, The device comprises:

17. An intelligent driving apparatus, characterized by comprising: a processor configured to execute a computer program stored in a memory, so that the device performs the method according to any one of claims 1 to 8. The device further comprises the memory.

18. The apparatus of claim 17, wherein, The intelligent driving system comprises a perception system and a computing platform, and the computing platform comprises the device according to any one of claims 9 to 18.

19. An intelligent driving system, characterized by, The device according to any one of claims 9 to 18, or the system according to claim 19.

20. A vehicle characterized by comprising: instructions stored thereon, which, when executed by a processor, cause the processor to implement the method according to any one of claims 1 to 8.

21. A computer-readable storage medium, characterized in that, The computer program product comprises computer program code which, when executed on a computer, causes the computer to implement the method according to any one of claims 1 to 8.

22. A computer program product, characterised in that, The chip comprises a circuit configured to execute the method according to any one of claims 1 to 8.

23. A chip, characterized by ​

Images