Motion planning method, device and system of carrying tool and carrying tool

Abstract

The invention discloses a motion planning method, device and system of a carrying tool and the carrying tool, relates to the field of vehicle-mounted technology, fully considers the influence of factors such as road surface bulge, road surface depression and road surface type on the vertical comfort degree in the vehicle driving process, and provides a motion planning method, device and system of the carrying tool on the basis of a global driving path planned by the vehicle. And more fine-grained vehicle local motion planning, such as local trajectory planning and / or vehicle parameter adjustment influencing comfort, is performed in advance by combining the obtained road vertical environment factors such as the road elevation and / or the road type. Therefore, no matter which driving scene the vehicle is in, which type of road surface and whether the road surface elevation is abnormal or not, the vehicle can carry out more flexible and fine local motion planning of each wheel based on the actual environment in advance, and it is guaranteed that the vehicle always keeps a high-comfort-degree riding state in the driving process.

Description

Technical Field

[0001] The present application relates to the field of vehicle-mounted technology, and in particular to a motion planning method, device, system and vehicle for a vehicle. Background Art

[0002] With the continuous development of intelligent driving technology, the use of vehicle motion planning based on road information is becoming more and more widespread. For example, the vehicle can plan the future driving route of the vehicle based on map information and detected obstacle information on the road.

[0003] However, when performing vehicle motion planning based on current intelligent driving technology, the main consideration is the lateral and longitudinal comfort of passengers during vehicle driving. For example, the impact of acceleration and deceleration on passenger comfort is considered in the lateral direction, and the impact of steering on passenger comfort is considered in the longitudinal direction, while the impact of vertical environmental factors (such as road elevation and other factors) on passenger comfort is ignored.

[0004] For example, when planning vehicle motion based on current intelligent driving technology, Figure 1 (a) shows the case where the road surface includes a speed bump and Figure 1 As shown in (b), there are depressions on the road surface, which cannot reduce the bumps caused to the user when the vehicle passes by, so the user's comfort is poor. Summary of the invention

[0005] The present application provides a motion planning method, device, system and vehicle for a vehicle, which can perform vehicle motion planning based on road vertical environmental factors during vehicle driving, thereby improving the comfort of users when riding in the vehicle.

[0006] In order to achieve the above objectives, this application adopts the following technical solutions:

[0007] In a first aspect, a motion planning method for a vehicle is provided, the method comprising: obtaining a planned path and road information representing the vehicle, wherein the road information includes a drivable area of ​​the vehicle, and a road surface elevation and / or road surface type in the drivable area; performing local trajectory planning and / or adjusting suspension parameters of the vehicle according to the road information and a first driving path.

[0008] Exemplarily, the method can be applied to an in-vehicle intelligent system.

[0009] The road elevation of the drivable area is used to reflect the undulation of the road surface, for example, the road elevation can reflect whether the road surface is damaged, whether it is sunken, whether there is a speed bump, etc. The road surface type of the drivable area can be distinguished based on the paving material of the road surface, such as Belgian road, cobblestone road, asphalt road, gravel road, etc.

[0010] As an example, the first driving path may be a driving path obtained after global path planning.

[0011] The solution provided in the first aspect above can fully consider the impact of factors such as road bumps, road depressions, and road type on the vertical comfort of the vehicle during driving on the basis of the global driving path planned by the vehicle when performing motion planning of the vehicle, and carry out more fine-grained vehicle local motion planning in advance by combining the acquired road elevation and / or road type and other road vertical environmental factors, such as local trajectory planning and / or vehicle parameter adjustment that affects comfort. Based on this, no matter what driving scene the vehicle is in, what type of road surface, or whether the road elevation is abnormal, the vehicle can carry out more flexible and detailed local motion planning of each wheel in advance based on the actual environment to ensure that the vehicle always maintains a high degree of comfort during driving.

[0012] As a possible implementation method, the above-mentioned local trajectory planning based on the road information and the first driving path includes: performing a trajectory search within a preset range of the first driving path in the drivable area to obtain a local planned trajectory of the vehicle, wherein the vertical comfort evaluation value corresponding to the local planned trajectory is greater than the vertical comfort evaluation value corresponding to the first driving path. Based on this, by performing a trajectory search within the preset range of the first driving path, the optimal state point can be found while ensuring that the local motion planning does not deviate too much from the global optimal path, and the local trajectory that can ensure the stability of the vehicle can be obtained while avoiding the waste of computing power and resources and ensuring driving safety.

[0013] As a possible implementation method, the above-mentioned local trajectory planning based on the road information and the first driving path includes: adjusting the first driving path within the preset range of the first driving path according to the road information to obtain the second driving path; performing speed planning on the second driving path to obtain the local planning trajectory of the vehicle, wherein the vertical comfort evaluation value corresponding to the local planning trajectory is greater than the vertical comfort evaluation value corresponding to the first driving path. Based on this, the local driving path can be planned within the preset range of the first driving path to ensure that the local motion planning will not deviate too much from the global optimal path, which can avoid the waste of computing power and vehicle power resources on the one hand, and ensure the safety of the vehicle's driving on the other hand.

[0014] As a possible implementation, the above-mentioned vertical comfort evaluation value is related to one or more of the following: wheel state, suspension state, passenger distribution state, vehicle state; wherein the wheel state indicates the position of multiple wheels of the vehicle, the road elevation and / or road type corresponding to the position, and one or more of the wheel speed, acceleration and tire pressure; the suspension state indicates one or more of the modes, loads, stiffness parameters and damping parameters of multiple suspensions of the vehicle; the passenger distribution state indicates whether there are passengers on multiple seats of the vehicle; the vehicle state indicates one or more of the wheelbase and track of the vehicle. Based on this, by fully considering the influence of factors such as wheel state, suspension state, passenger distribution state, and vehicle state on the vertical comfort of the vehicle during driving, the optimal state point in the trajectory search process is determined based on this, and then a reliable local motion planning result is obtained.

[0015] As a possible implementation, the above-mentioned local trajectory planning and / or suspension parameter adjustment of the vehicle based on the road information and the first driving path includes: when the first driving path includes a first area where the road elevation meets the first preset condition, local trajectory planning is performed based on the position information corresponding to the drivable area and the first area; wherein, when driving based on the first driving path, M wheels of the vehicle pass through the first area, M is a positive integer; when driving based on the local planned trajectory, N wheels of the vehicle pass through the first area, N is an integer, and N<M. Based on this, in some cases, such as when the road surface is too concave or too convex, the concave / convex area can be bypassed so that as few wheels as possible pass through the concave or convex area of ​​the road surface, thereby improving the comfort of the vehicle during driving.

[0016] Exemplarily, the above N can be 0, that is, when driving according to the motion trajectory after local trajectory planning, each tire of the vehicle may not pass through the concave or raised area of ​​the road surface; or, the above N can be greater than 0, that is, when driving according to the motion trajectory after local trajectory planning, the vehicle may not be able to avoid passing through the concave or raised area of ​​the road surface, but as few wheels as possible can pass through the concave or raised area of ​​the road surface.

[0017] As a possible implementation, the above-mentioned local trajectory planning and / or suspension parameter adjustment of the vehicle according to the road information and the first driving path further includes: when the first driving path includes a first area whose road elevation meets the first preset condition, the suspension parameters of the vehicle are adjusted according to the road elevation corresponding to the first area. Based on this, in some cases, such as when the road surface is too concave or too convex, the suspension parameters can be adjusted while bypassing the concave / convex area so that as few wheels as possible pass through the concave or convex area of ​​the road surface, so as to further improve the comfort of the vehicle during driving.

[0018] As a possible implementation, the above-mentioned local trajectory planning and / or suspension parameter adjustment of the vehicle according to the road information and the first driving path further includes: when the first driving path includes a second area with a road surface type of a preset type, the suspension parameters of the vehicle when passing through the second area are adjusted according to the road surface type corresponding to the second area. Based on this, in some cases, for example, when the road surface is uneven due to the road surface type, so that the vehicle will have a large degree of bumps when passing through, the suspension parameters can be adjusted to improve the comfort of the vehicle during driving.

[0019] As an example, the local planning trajectory of the vehicle obtained after the above trajectory search may be a rough local planning trajectory. In order to make the curvature of the local motion trajectory within the following capability of the vehicle, the rough local planning trajectory may also be smoothed.

[0020] As a possible implementation method, the above vertical comfort evaluation value is obtained based on the following formula 1:

[0021] S i =f(C f1 , C f2 , …, C fP , S f1 , S f2 , …, S fL , W); (Formula 1)

[0022] In the above formula 1, f() is the vertical comfort model, P is the number of wheels of the vehicle, and C f1 , C f2 ,……,C fP Respectively represent the wheel states of the first wheel, the second wheel, ..., and the Pth wheel of the vehicle, and the wheel states represent the position of the wheel, the road elevation and / or road type corresponding to the position, and one or more of the wheel speed, acceleration, and tire pressure. Exemplarily, T is the number of sampling points, is the position of the pth wheel, t is the sampling mark, is the speed of the pth wheel, is the acceleration of the pth wheel, is the elevation of the road surface that the pth wheel passes through, is the road surface type of the road surface that the pth wheel passes through, p p is the tire pressure of the pth wheel. L is the number of suspensions of the vehicle, S f1 , S f2 ,……,S fLrepresent the suspension states of the first suspension, the second suspension, ..., and the Lth suspension of the vehicle, respectively, and the suspension state represents one or more of the mode, load, stiffness parameter, and damping parameter of the suspension. fl =[m l ,F l ,k l ,c l ], l∈[1,L],m l is the first suspension mode, F l is the load of the first suspension, k l is the stiffness parameter of the first suspension, c l is the damping parameter of the lth suspension. The suspension modes, such as sports mode, standard mode, comfort mode, etc., depend on the specific hardware configuration of the suspension. W is the vehicle state and / or passenger distribution state. The vehicle state indicates the wheelbase and / or track of the vehicle. The passenger distribution state indicates whether there are passengers in multiple seats of the vehicle. Exemplarily, W = [wb, td, r], wb is the wheelbase of the vehicle, td is the track of the vehicle, and r is the passenger distribution state of the vehicle. Taking the vehicle as a five-seater vehicle as an example, r = [r fl , r fr , r rl , r rm , r rr ], where r fl 、r fr 、r rl 、r rm and r rr Respectively indicate whether there are passengers in the driver's seat, the front passenger seat, the rear left seat, the rear middle seat and the rear right seat. For example, if there are passengers, the corresponding value is 1, and if there are no passengers, the corresponding value is 0. By fully considering the impact of factors such as road bumps, road depressions, road type, wheel status, and suspension status on the vertical comfort of the vehicle during driving, the optimal state point in the trajectory search process is determined based on this, and then a reliable local motion planning result is obtained.

[0023] As a possible implementation, the method further includes: determining a first vertical comfort evaluation value corresponding to the local planned trajectory based on the vertical comfort model; obtaining parameters of the vehicle during the driving process according to the local planned trajectory, the parameters including one or more of the following information: motion parameters at multiple seats, state parameters of the suspension, and tire pressure parameters of the wheels; determining a second vertical comfort evaluation value based on the above parameters based on preset rules; optimizing the vertical comfort model based on the relationship between the second vertical comfort evaluation value and the first vertical comfort evaluation value. Based on this, the network parameters of the supervised training neural network can be updated by comparing the comfort evaluation value obtained by the supervised training neural network and the actual comfort back propagation, so as to obtain an evaluation result that is closer to the actual comfort evaluation model, and realize a virtuous cycle of the reliability of the local motion trajectory planning results.

[0024] As a possible implementation, the method further includes: training and learning the historical driving trajectory of the vehicle and the corresponding first vertical comfort evaluation value to determine the preset rule, wherein the historical driving trajectory represents the historical driving path and the corresponding speed strategy. Based on this, a more accurate algorithm can be obtained to obtain a more accurate actual comfort based on the algorithm.

[0025] As a possible implementation, the method further includes: determining a first parameter based on a first driving path and road information of the vehicle; determining a second parameter based on a local planned trajectory and / or a suspension parameter adjustment strategy; determining a third parameter when the vehicle is traveling according to the local planned trajectory based on a vertical comfort model, the third parameter representing a vertical comfort value; optimizing the vertical comfort model according to the first parameter, the second parameter, and the third parameter. Based on this, a deep fusion of the local trajectory search and evaluation network can be achieved through a neural network trained by reinforcement learning, and a reliable comfort evaluation model optimization solution can be provided while considering the decision logic and capabilities of the search tree and the control capabilities of the suspension when there is a preview.

[0026] As an example, the local planning trajectory of the vehicle obtained after the above trajectory search also satisfies one or more of the following conditions: collision constraint, dynamic constraint (such as curvature constraint, etc.), global deviation constraint, XY comfort constraint. Based on this, the optimal state point in the trajectory search process can be determined, and then a reliable local motion planning result can be obtained.

[0027] As a possible implementation, the vehicle includes a positioning device, and the acquisition of the drivable area of ​​the vehicle includes: collecting the location information of the vehicle through the positioning device; and determining the drivable area according to the location information of the vehicle and the surrounding environment information of the vehicle. Based on this, the drivable area of ​​the vehicle can be collected in real time, providing necessary support for subsequent local motion planning.

[0028] As a possible implementation, the vehicle further includes a sensing device, and the surrounding environment information is obtained based on data collected by the sensing device; wherein the sensing device includes a sensing sensor, and the sensing sensor includes one or more of the following: a laser radar, a visible light camera, and a millimeter wave radar. Based on this, the surrounding environment information of the vehicle can be collected in real time, providing necessary support for subsequent local motion planning.

[0029] As a possible implementation method, the above-mentioned road elevation and road type are obtained based on the data collected by the sensing device. Optionally, the above-mentioned road elevation and road type can also be determined based on the received map (such as a high-precision map, a vector map, a navigation map, etc.) or road information shared by other vehicles. Based on this, the road elevation and road type corresponding to the drivable area of ​​the vehicle can be collected in real time, providing necessary support for subsequent local motion planning.

[0030] As a possible implementation, the suspension of the vehicle is an active suspension, a semi-active suspension or a passive suspension. Exemplarily, the suspension parameters include one or more of the following: suspension height, suspension damping, and suspension stiffness. Based on this, no matter what type of suspension the vehicle is, such as active suspension, semi-active suspension or passive suspension, the vehicle can adjust the suspension parameters in advance based on the actual environment to ensure that the vehicle always maintains a high degree of comfort during driving.

[0031] In a second aspect, a motion trajectory planning device is provided, which includes: a data acquisition module for collecting or acquiring information; a memory for storing computer program instructions; and a processor for executing the computer program instructions to support the motion trajectory planning device to implement a method as in any possible implementation of the first aspect.

[0032] In a third aspect, a vehicle-mounted system is provided, the vehicle-mounted system comprising the motion trajectory planning device as described in the second aspect. The vehicle-mounted system is used to implement the method in any possible implementation manner of the first aspect.

[0033] In a fourth aspect, a vehicle is provided, which includes the motion trajectory planning device provided in the second aspect, or includes the vehicle-mounted system provided in the third aspect.

[0034] According to a fifth aspect, a computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method in any possible implementation of the first aspect is implemented.

[0035] According to a sixth aspect, a computer program product comprising instructions is provided, which, when executed on a computer, enables the computer to implement a method as in any possible implementation of the first aspect.

[0036] In a seventh aspect, a chip system is provided, the chip system comprising a processing circuit and a storage medium, wherein the storage medium stores computer program instructions; when the computer program instructions are executed by the processor, the method in any possible implementation of the first aspect is implemented. The chip system may be composed of a chip, or may include a chip and other discrete devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 Schematic diagram of two vehicle driving scenarios provided in the embodiments of the present application;

[0038] Figure 2 A schematic diagram of a vehicle motion planning process provided in an embodiment of the present application;

[0039] Figure 3 A structural block diagram of a vehicle-mounted intelligent system provided in an embodiment of the present application;

[0040] Figure 4 A schematic diagram of a motion planning process provided in an embodiment of the present application;

[0041] Figure 5 Two parking scene schematic diagrams are provided for the embodiments of the present application;

[0042] Figure 6 Schematic diagrams of two block scenes provided for embodiments of the present application;

[0043] Figure 7 A schematic diagram of a scene in which a road surface type is a preset type provided in an embodiment of the present application;

[0044] Figure 8 A flow chart of a motion planning method for a vehicle provided in an embodiment of the present application;

[0045] Fig. 9 A schematic diagram of a global driving path of a vehicle provided in an embodiment of the present application;

[0046] Fig.10 A grid diagram of a drivable area of ​​a vehicle provided in an embodiment of the present application;

[0047] Fig.11 A schematic diagram of a driving status search process for two vehicles provided in an embodiment of the present application;

[0048] Fig.12 Schematic diagram of two local rough trajectory planning results provided in the embodiments of the present application;

[0049] Fig.13 A schematic diagram of a trajectory search process based on a trajectory search tree provided in an embodiment of the present application;

[0050] Fig.14 A schematic diagram of the working process of a neural network for supervised training provided in an embodiment of the present application;

[0051] Fig.15 A schematic diagram of a parameter optimization process of a supervised training neural network provided in an embodiment of the present application;

[0052] Fig.16 Schematic diagram of three node evaluation examples provided in the embodiments of the present application;

[0053] Fig.17 A vertical comfort model optimization solution implementation architecture diagram provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. In the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, "multiple" means two or more than two.

[0055] In the following, the terms "first", "second", etc. are only used to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. For example, if the described object is a "field", the ordinal number before the "field" in the "first field" and the "second field" does not limit the position or order between the "fields", and the "first" and "second" do not limit whether the "fields" they modify are in the same message, nor do they limit the order of the "first field" and the "second field". For another example, if the described object is a "level", the ordinal number before the "level" in the "first level" and the "second level" does not limit the priority between the "levels". For another example, the number of described objects is not limited by the ordinal number, and can be one or more. Taking the "first device" as an example, the number of "devices" can be one or more. In addition, the objects modified by different prefixes can be the same or different. For example, if the described object is a "device", the "first device" and the "second device" can be the same type of device or different types of devices. For another example, if the described object is "information", the "first information" and the "second information" can be information of the same content or information of different contents. In summary, the use of prefixes such as ordinal numbers to distinguish the described objects in the embodiments of the present application does not constitute a limitation on the described objects. For the statement of the described objects, please refer to the description in the context of the claims or embodiments, and no unnecessary limitation should be constituted due to the use of such prefixes.

[0056] Furthermore, in the embodiments of the present application, "connection" may be a direct connection or an indirect connection; in addition, it may refer to an electrical connection or a communication connection; for example, the connection between two electrical components A and B may refer to a direct connection between A and B, or may refer to an indirect connection between A and B through other electrical components or connecting media, or may refer to an indirect connection between A and B through other communication devices or communication media, as long as communication between A and B can be achieved.

[0057] First of all, it should be noted that the vehicle described in the embodiments of the present application may include land vehicles, water vehicles, air vehicles, industrial equipment, agricultural equipment or entertainment equipment, etc. For example, the vehicle may be a vehicle, which is a vehicle in a broad sense, and may be a vehicle (such as a car, a bus, a subway, a high-speed rail, a motorcycle, a flying car, a train, etc.), an industrial vehicle (such as a forklift, a trailer, a tractor, etc.), an engineering vehicle (such as an excavator, a bulldozer, a crane, etc.), agricultural equipment (such as a mower, a harvester, etc.), amusement equipment, a toy vehicle, a ship, an air cushion vehicle, a submarine, an airplane, a helicopter, etc. The embodiments of the present application do not limit the specific type, form and function of the vehicle. In the following embodiments, "vehicle" will be used as an example to introduce the specific process of motion planning of the vehicle.

[0058] In order to further improve the comfort of users when riding in a vehicle, as a first possible implementation method, such as Figure 2 As shown, the vehicle can collect vertical acceleration information of the vehicle in the vertical direction during the movement, and then perform vehicle movement planning based on the vertical acceleration information, such as adjusting the suspension parameters of the vehicle when the vertical acceleration is too large. As an example, the vehicle can obtain vertical acceleration information through an inertial navigation system, and the specific manner and method for the vehicle to obtain vertical acceleration information is not limited. Figure 2 Just as an example.

[0059] As a second possible implementation, Figure 2 As shown, the vehicle can obtain road information such as line features, point features, and surface features used to characterize the degree of road undulation during movement, and then perform vehicle movement planning based on the above road information, such as lane changing, speed limit, or lane avoidance. For example, the vehicle can change lanes and / or limit speed when the line features meet the first condition, avoid lanes and / or limit speed when the point features meet the second condition, and limit speed when the surface features meet the third condition. As an example, the vehicle can obtain road information such as line features, point features, and surface features of the road through a lidar. The specific way and method for the vehicle to obtain road information is not limited. Figure 2 Just as an example.

[0060] As a third possible implementation, Figure 2 As shown, the vehicle can obtain road information such as line features, point features, and surface features used to characterize the degree of road undulation, as well as vertical acceleration information of the vehicle in the vertical direction during movement, and then combine the above road information and vertical acceleration information to perform vehicle motion planning, such as combining the vertical acceleration information and the line features of the road to perform lane changing and / or vehicle speed limit, combining the vertical acceleration information and the point features of the road to perform vehicle speed limit and / or lane avoidance, and combining the vertical acceleration information and the surface features of the road to perform vehicle speed limit.

[0061] However, in the first and third implementations described above, the suspension parameters are adjusted only when the vehicle is passing through a road with large undulations causing the vehicle to be bumpy. Therefore, the response is delayed, which still brings an uncomfortable riding experience to the passengers.

[0062] In addition, in the second and third implementation methods mentioned above, the vehicle motion planning only involves lane changing, speed limit or lane avoidance. For some scenarios, such as the limited number of lanes or intersections that make it impossible to change lanes, the inability to effectively avoid in lanes, cobblestone roads, etc., this solution will still bring an uncomfortable riding experience to passengers.

[0063] Based on the above considerations, the embodiment of the present application provides a motion planning method for a vehicle with strong applicability and high reliability. The method can, based on the global driving path planned by the vehicle, combine the acquired road elevation and / or road type and other road vertical environmental factors to perform more fine-grained vehicle local motion planning (such as local trajectory planning) and / or adjust vehicle parameters that affect comfort in advance. Based on this solution, the vehicle can perform more detailed local motion planning and / or vehicle parameter adjustment based on the actual environment in advance to ensure that the vehicle always maintains a high degree of comfort during driving.

[0064] The method provided in the embodiment of the present application can be implemented based on the vehicle-mounted intelligent system on the vehicle. For example, the vehicle-mounted intelligent system can be an intelligent control device on the vehicle, or it can be an intelligent control system composed of multiple modules cooperating with each other on the vehicle, which is not limited in the embodiment of the present application.

[0065] For example, please refer to Figure 3 , Figure 3 FIG. 1 shows a structural block diagram of a vehicle-mounted intelligent system provided by an embodiment of the present application. Figure 3 As shown, the vehicle intelligent system may include Figure 3 The system includes a perception module 310 , a localization module 320 , a planning module 330 and a control module 340 .

[0066] Among them, the perception module 310 is used to perceive the surrounding environment of the vehicle, including the drivable roads around the vehicle and obstacle information within the preset range of the roads, obstacle information such as dynamic obstacle information and static obstacle information, etc. For example, dynamic obstacles include moving vehicles, pedestrians, animals, etc., and static obstacles include parking lines, stationary vehicles, trees, curbs, railings, street lights, steps, etc., which are not specifically limited in the embodiments of the present application.

[0067] In the embodiment of the present application, the perception module 310 is also used to perceive the vertical environmental factors of the road, which represent but are not limited to the road surface elevation and / or road surface type. Among them, the road surface elevation is used to reflect the degree of undulation of the road surface. For example, the road surface elevation can reflect whether the road surface is damaged, whether it is depressed, whether there is a speed bump, etc. The road surface type can be distinguished based on the paving material of the road surface, such as Belgian road, cobblestone road, asphalt road, gravel road, etc. Of course, the embodiment of the present application does not limit the specific rules and methods for defining the road surface type. For example, the road surface type can also be distinguished based on the geographical location of the road surface, such as mountain roads, wading roads, highways, etc.

[0068] As an example, the perception module 310 may include one or more sensors, and the perception module 310 may perceive the vehicle surrounding environment and road vertical environmental factors through the one or more sensors. Exemplarily, the one or more sensors are one or more of a time of flight (TOF) sensor, a camera (visible light camera such as an infrared camera), a radar (such as a millimeter wave radar, a laser radar, an ultrasonic radar, etc.), a vehicle posture sensor (such as a gyroscope), an inertial measurement unit (IMU), etc., which are not specifically limited in the embodiments of the present application.

[0069] The positioning module 320 is used to provide the position information of the vehicle in the predefined coordinate system, including position information and motion posture information. The position information represents the position coordinates of the vehicle, and the motion posture information represents the vehicle chassis information such as the turning angle of the vehicle wheels.

[0070] In some embodiments, the positioning module 320 may provide absolute position information of the vehicle based on a global navigation satellite system (GNSS) as the position information of the vehicle.

[0071] Exemplarily, GNSS may include, but is not limited to, a global positioning system (GPS), a global navigation satellite system (GLONASS), a Beidou navigation satellite system (BDS), a quasi-zenith satellite system (QZSS) and / or a satellite based augmentation system (SBAS).

[0072] In some embodiments, for example, in a scenario where GNSS has no coverage (such as an underground parking lot), the positioning module 320 can provide the relative position information of the vehicle as the vehicle's position information based on the vehicle's surrounding environment information perceived by the perception module 310 and the environment map in the positioning module 320. The environment map includes a map of the area where the vehicle is currently located. The environment map may include but is not limited to a high-precision map, a vector map, a navigation map, or a map shared by other vehicles.

[0073] Exemplarily, the environmental map in the positioning module 320 may be pre-set in the positioning module 320, or may be obtained by the positioning module 320 from the cloud in real time through the Internet, or may be drawn by the positioning module 320 through automatic composition during historical use and saved in the positioning module 320, which is not specifically limited in the embodiments of the present application.

[0074] In other embodiments, the positioning module 320 may provide fused position information of the vehicle as the position information of the vehicle based on the absolute position information of the vehicle and the relative position information of the vehicle.

[0075] In some embodiments, the positioning module 320 can obtain relevant information from the vehicle posture sensor, the vehicle chassis, the control module 340, the vehicle steering wheel module, etc. to determine the vehicle's motion posture information based on the relevant information.

[0076] The planning module 330 is used to plan the vehicle's motion path based on the vehicle's surrounding environment information perceived by the perception module 310 and the vehicle's location information provided by the positioning module 320, including global driving path planning and local trajectory planning. Among them, the global driving path can determine the vehicle's future driving route from a global perspective. Exemplarily, the planning module 330 can determine the vehicle's drivable area based on the vehicle's surrounding environment information perceived by the perception module 310, and then perform global driving path planning for the vehicle.

[0077] The control module 340 is used to control the vehicle to follow the corresponding trajectory according to the global driving path or the local planned trajectory planned by the planning module 330.

[0078] In some embodiments of the present application, the control module 340 can also be used to control the vehicle to adjust vehicle parameters that affect comfort based on the global driving path when the road elevation and / or road type perceived by the perception module 310 meets the corresponding preset conditions.

[0079] The vehicle parameters that affect the comfort level described in the embodiments of the present application may include, but are not limited to, vehicle suspension parameters (the following embodiments use suspension parameters as examples). The main function of the vehicle suspension is to transmit the force and torque acting on the wheels and the frame (or body) to mitigate the impact load transmitted from the road surface to the frame (or body), attenuate the vehicle vibration caused by the road surface impact, ensure that the wheels have ideal motion characteristics when the road surface is uneven and the load changes, and ensure the stability of the vehicle during driving.

[0080] Exemplarily, the suspension parameters described in the embodiments of the present application may include, but are not limited to, suspension kinematic parameters and suspension compliance parameters. The suspension kinematics refers to the characteristics of the angular displacement and linear displacement changes of the wheel plane and wheel center point due to the action of the suspension guide mechanism during the reciprocating motion of the wheel in the vertical direction. The suspension compliance refers to the characteristics of the angular displacement and linear displacement changes of the wheel plane and wheel center point caused by the force and torque of the ground acting on the tire. Exemplarily, suspension parameters include suspension height, suspension damping, suspension stiffness, etc.

[0081] As an example, the vehicle-mounted intelligent system may further include a serial bus module. The serial bus module is used as a medium for information and command communication between the control module 340 and the vehicle chassis or suspension. Exemplarily, the serial bus module is a controller area network bus (CANBUS).

[0082] It should be noted that Figure 3 The structure shown in the embodiment of the present application is only an example of a possible structure of an in-vehicle intelligent system, and does not constitute a specific limitation on the in-vehicle intelligent system. In other embodiments of the present application, the in-vehicle intelligent system may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure can be implemented in hardware, software, or a combination of software and hardware.

[0083] For example, the vehicle-mounted intelligent system may also include a prediction module, which is used to predict the future movement trajectory of dynamic obstacles in real time according to the information of targets (such as dynamic obstacles) in the vehicle's surrounding environment perceived by the perception module 310 during the vehicle's movement, so as to support the vehicle to stop when it is predicted that the vehicle is about to collide with the target, achieve emergency obstacle avoidance, ensure driving safety, etc. For example, the information of the target may include but is not limited to the position, size, movement direction and movement speed of the target within the preset range of the vehicle.

[0084] Alternatively, exemplarily, the on-board intelligent system may also include a vehicle parameter planning module, which is used to determine a vehicle parameter adjustment plan that affects comfort based on the global driving path when the road elevation and / or road type perceived by the perception module 310 meets the corresponding preset conditions.

[0085] Alternatively, illustratively, the vehicle-mounted intelligent system may not include a planning module, and the functions of the above-mentioned planning module may be implemented by other devices with path planning functions.

[0086] based on Figure 3 The vehicle intelligent system shown is an example. Please refer to Figure 4 , Figure 4 A schematic diagram of a motion planning process provided in an embodiment of the present application is shown.

[0087] After the perception module 310 perceives the vehicle's surrounding environment information and the road vertical environment factors, the perception module 310 can determine the vehicle's drivable area (such as a lane) and obstacle information within a preset range of the drivable area based on the vehicle's surrounding environment information, and determine the road elevation and / or road type corresponding to the drivable area based on the road vertical environment factors. And, Figure 4 As shown, the perception module 310 can send the vehicle's drivable area (such as a lane), obstacle information within a preset range of the drivable area, and the road surface elevation and / or road surface type corresponding to the drivable area to the planning module 330.

[0088] like Figure 4 As shown, after the positioning module 320 obtains the posture information of the vehicle in the predefined coordinate system, the positioning module 320 can send the posture information to the planning module 330.

[0089] Furthermore, if Figure 4 As shown, the planning module 330 can perform global driving path planning based on the vehicle's drivable area (such as a lane) and obstacle information within a preset range of the drivable area from the perception module 310 and the vehicle's posture information from the positioning module 320. In some embodiments, after obtaining the global driving path, the planning module 330 can also perform speed planning on the global driving path to obtain a global driving trajectory.

[0090] As an example, Figure 4 As shown, the vehicle can plan a local motion trajectory based on specific road information, adjust suspension parameters, or both plan a local motion trajectory and adjust suspension parameters. This is not limited in the present application embodiment and depends on the specific driving scenario and specific comfort requirements. The planning of the local motion trajectory includes the planning of the local motion path and / or the local motion speed.

[0091] In some embodiments, after obtaining the global driving path / global driving trajectory, the planning module 330 may send the global driving path / global driving trajectory to the control module 340 for the control module 340 to perform subsequent vehicle trajectory following based on this. In addition, the planning module 330 may simultaneously perform local motion planning based on the road elevation and / or road type corresponding to the drivable area collected in real time by the perception module 310, and send the planning result to the control module 340 for the control module 340 to perform subsequent vehicle trajectory following based on this. The control module 340 may also adjust the suspension parameters based on the road elevation and / or road type corresponding to the drivable area collected in real time by the perception module 310.

[0092] Among them, when performing local trajectory planning, due to the abnormality of the local road elevation, it is usually difficult to directly obtain a smooth, obstacle-avoiding, efficient and comfortable trajectory from end to end. Therefore, as an example, Figure 4 As shown, the planning module 330 can first perform rough planning of the local motion trajectory to provide a high-quality initial solution, and then smooth the rough-planned local motion trajectory to compensate for the error caused by the rough trajectory being not smooth enough when the control module 340 follows the rough trajectory, so that the curvature of the local motion trajectory is within the following capability of the vehicle.

[0093] The following is an introduction to scenarios to which the embodiments of the present application may be applied.

[0094] For example, please refer to Figure 5 and Figure 6 , Figure 5 The present invention provides two parking scenarios. Figure 5 In the scenario shown, the vehicle's movement goal is to park into the target parking space from the initial position; Figure 6 Two block scene schematic diagrams provided in the embodiments of the present application are shown. Figure 6 In the scenario shown, the vehicle's movement goal is to pass the current road smoothly.

[0095] Specifically, in Figure 5 In the scenario shown in (a), an unavoidable speed bump is provided in the vehicle's drivable lane. Based on the motion planning method for a vehicle provided in an embodiment of the present application, the vehicle can plan suspension parameters and local driving trajectory (such as reducing vehicle speed) in advance according to the identified road elevation of the speed bump to ensure the stability of the vehicle when passing over the speed bump.

[0096] exist Figure 5 In the scenario (b) shown in the figure, a speed bump that can be partially avoided is set in the vehicle's drivable lane. Based on the motion planning method for a vehicle provided in an embodiment of the present application, the vehicle can plan a local trajectory in advance according to the road elevation of the identified speed bump, so that the vehicle is adjusted to the right lane. After the vehicle is adjusted to the right lane, only one left tire of the vehicle passes over the speed bump, which will greatly alleviate the bumps of the vehicle when passing over the speed bump. Figure 5 In the scenario shown in (b), the vehicle can also adjust the suspension parameters while adjusting the vehicle to the right lane to further ensure the stability of the vehicle when passing over the speed bump, thereby ensuring the vertical comfort of the vehicle.

[0097] Or, in Figure 5 In the scenario shown in (b) in FIG. 1 , the vehicle may also adjust the suspension parameters directly without performing local trajectory planning to ensure the stability of the vehicle when passing over the speed bump.

[0098] exist Figure 6 In the scenario (a) shown in the figure, a speed bump that can be partially avoided is set in the vehicle's drivable lane. Based on the motion planning method for a vehicle provided in an embodiment of the present application, the vehicle can plan a local trajectory in advance according to the road elevation of the identified speed bump, so that the vehicle is adjusted to the right lane. After the vehicle is adjusted to the right lane, only one left tire of the vehicle passes over the speed bump, which will greatly alleviate the bumps of the vehicle when passing over the speed bump. Figure 6 In the scenario shown in (b), the vehicle can also adjust the suspension parameters while adjusting the vehicle to the right lane to further ensure the stability of the vehicle when passing over the speed bump, thereby ensuring the vertical comfort of the vehicle.

[0099] exist Figure 6 In the scenario shown in (b), there is a depression in the vehicle's drivable lane where the vehicle's wheels are about to pass. Based on the motion planning method for the vehicle provided in the embodiment of the present application, the vehicle can perform local trajectory planning in advance according to the identified elevation of the depressed road surface, so that the vehicle path is adjusted to the left to avoid the depression, thereby preventing the vehicle's right wheel from being jolted when passing through the depression.

[0100] Or, for example, see Figure 7 , Figure 7 A schematic diagram of a scene in which a road surface type is a preset type provided in an embodiment of the present application is shown. Figure 7 In the scenario shown, the purpose of the vehicle's movement is to pass the current road smoothly. The drivable lane in front of the vehicle is a cobblestone road. Based on the motion planning method of the vehicle provided in the embodiment of the present application, the vehicle can plan the suspension parameters and local driving speed in advance according to the identified road surface type of the drivable lane to ensure the stability of the vehicle when passing through the cobblestone road, thereby ensuring the vertical comfort of the vehicle.

[0101] Of course, the motion planning method for the vehicle provided in the embodiment of the present application can also be applied to other situations where the global planning path passes through areas with abnormal road elevation or the road type is a preset type, which are not listed one by one in this application.

[0102] The following will describe in detail the motion planning method of the vehicle provided in the embodiment of the present application in conjunction with the accompanying drawings.

[0103] As an example, see Figure 8 , Figure 8 Taking the vehicle intelligent system for vehicle motion planning as an example, a flow chart of a vehicle motion planning method provided by an embodiment of the present application is shown. Figure 8 As shown, a motion planning method for a vehicle provided in an embodiment of the present application may include S801-S804:

[0104] S801: Determine whether global path planning has been performed. If yes, continue to execute S802; if no, the algorithm ends.

[0105] Among them, global path planning refers to the planning of the vehicle's driving path in a global range. After global path planning, a global driving path (referred to as the first driving path) can be obtained, and the global driving path can represent the future driving path of the vehicle.

[0106] In some embodiments of the present application, the global driving path of the vehicle includes the global driving path of the vehicle wheel granularity. For example, the global driving path of the vehicle includes the global driving paths of all wheels of the vehicle. For example, taking the vehicle including four wheels as an example, the global driving path of the vehicle may be as follows: Fig. 9 shown.

[0107] In the embodiment of the present application, the global optimal path can be obtained based on the global path planning, providing the most efficient motion path planning for the vehicle movement. In addition, the global driving path, as the basis for subsequent local motion planning, can not only ensure that the local motion planning does not significantly change the global optimal path, but also maximize the safety and comfort of the vehicle during driving.

[0108] As an example, the vehicle intelligent system can perform global path planning based on but not limited to any one or more of the following algorithms: graph search-based algorithms (such as Dijkstra algorithm, A star (A*) algorithm, D star (D*) algorithm, etc.), artificial field-based algorithms, sampling-based algorithms (such as rapid-exploring random tree (RRT) algorithm), and intelligent method-based algorithms (such as fuzzy logic algorithms, neural network algorithms, genetic algorithms, etc.). The specific methods and specific processes of global path planning are not limited in the embodiments of this application, and reference can be made to conventional technologies.

[0109] S802: Determine a local area to be planned according to the acquired road information, wherein the road information includes a drivable area and a road surface elevation and / or road surface type of the drivable area.

[0110] The road information includes the drivable area of ​​the vehicle, and the road elevation and / or road type of the drivable area. For example, the road elevation of the drivable area can be expressed as x=[x1, x2, …, x N ], x n =(x,y,h), n∈[1,N], (x,y) is the coordinate information of grid n, and h is the elevation of grid n.

[0111] As an example, the vehicle intelligent system can Figure 3 The vehicle's surrounding environment is sensed by the sensing module 310, such as the drivable roads around the vehicle and the obstacle information within the preset range of the road to determine the drivable area of ​​the vehicle. Figure 3 The perception module 310) determines the corresponding road surface elevation and / or road surface type based on the road vertical environmental factors of the drivable area perceived by the perception module 310.

[0112] As an example, the on-board intelligent system determines the local area to be planned based on the acquired road information, which may specifically include: the on-board intelligent system determines whether there is an area with a road surface elevation that meets preset conditions on the global driving path within the vehicle's drivable area based on the road surface elevation of the drivable area in the acquired road information; if so, the on-board intelligent system takes the area (such as recorded as the first area) as the local area to be planned.

[0113] Exemplarily, the preset condition is such as the first preset condition, and the first preset condition is: h3<road surface elevation<h4. For example, the road surface elevation of the first area is greater than h3 (such as -5cm), which indicates that the first area is a depression with a depth less than 5cm. The road surface elevation of the first area is less than h4 (such as 5cm), which indicates that the first area is a protrusion with a height less than 5cm.

[0114] Alternatively, exemplarily, the preset condition is such as the first preset condition, and the first preset condition is: road surface elevation ≥ h2 or road surface elevation ≤ h1, h1<0, h2>0. For example, the road surface elevation of the first area is less than or equal to h1 (such as -2 cm), indicating that the first area is a depression with a depth greater than 2 cm, and the road surface elevation of the first area is greater than or equal to h2 (such as 2 cm), indicating that the first area is a protrusion with a height greater than or equal to 2 cm.

[0115] Alternatively, exemplarily, the preset condition is such as the second preset condition, and the second preset condition is: h1<road surface elevation<h2. For example, if the road surface elevation of the first area is greater than h1 (such as -2 cm), it means that the first area is a depression with a depth less than 2 cm, and if the road surface elevation of the first area is less than h2 (such as 2 cm), it means that the first area is a protrusion with a height less than 2 cm.

[0116] It should be noted that the above-mentioned preset conditions, such as the values ​​of the first preset condition or the second preset condition, are only examples. In actual applications, they can be adjusted according to factors such as the performance of the perception module (such as perception ability) and suspension ability, and this application does not make specific limitations.

[0117] As a possible implementation method, the vehicle intelligent system can grid the vehicle's drivable area based on the acquired road information and match the road elevation of the drivable area to the corresponding grid, thereby preparing for determining the local area to be planned based on the road elevation and other subsequent processes. For example, the vehicle intelligent system can divide the vehicle's drivable area into units of △L. Fig.10 The basic unit shown is the discrete space of △L.

[0118] It can be understood that when the vehicle's drivable area is gridded, if the basic unit △L of the gridding is too large, the search may fail due to the low discrete accuracy of the drivable area; however, if the basic unit △L of the gridding is too small, the discrete accuracy of the drivable area may be too high, resulting in excessive calculation, affecting the real-time performance of the algorithm. In an embodiment of the present application, the vehicle-mounted intelligent system can select a suitable △L for gridding the drivable area based on one or more of the following factors: the computing power of the vehicle-mounted intelligent system, the perception power of the perception module, the size of the vehicle's drivable area, the vehicle's surrounding environment, and scene requirements. For example, in a low-speed parking scene, the basic unit △L of the gridding can be set to 0.25m x 0.25m; in a medium- and low-speed unstructured road scene, such as a non-main road, an intersection turn, a rural street, etc., the basic unit △L of the gridding can be set to 1m x 1m; in a medium- and high-speed structured road scene, such as a highway, an urban trunk road, etc., the basic unit △L of the gridding can be set to 1m x 2m.

[0119] As an example, the on-board intelligent system determines the local area to be planned based on the acquired road information, which may specifically include: the on-board intelligent system determines whether there is an area on the global driving path within the vehicle's drivable area whose road surface type meets a preset type based on the acquired road information; if so, the on-board intelligent system uses the area (such as recorded as the second area) as the local area to be planned.

[0120] For example, the preset types are uneven road types such as Belgian roads, cobblestone roads, and gravel roads. It can be understood that for uneven roads, the vehicle will have a greater degree of bumps when passing through. By taking the area where the road type meets the preset type as the local area to be planned for subsequent local motion planning, the bumps of the vehicle when passing through the road can be greatly reduced, ensuring the stability of the vehicle when passing through the road.

[0121] As a possible implementation method, the on-board intelligent system can match the road surface type of the drivable area to the corresponding grid, thereby preparing for determining the local area to be planned based on the road surface type and other subsequent processes.

[0122] S803: Perform local trajectory planning and / or suspension parameter planning according to the road elevation and / or road type corresponding to the local area to be planned. If the planning is successful, continue to execute S804; if the planning fails (such as both the local trajectory planning and the suspension parameter planning fail), the algorithm ends.

[0123] Among them, the decision of the on-board intelligent system to perform local trajectory planning, suspension parameter planning, or both local trajectory planning and suspension parameter planning is determined by the specific road surface elevation and / or road surface type of the local area to be planned.

[0124] For example, in some examples, assuming that the elevation of the road surface in the local area to be planned meets the first preset condition, the vehicle intelligent system can decide to perform local trajectory planning. Based on this, the vehicle's driving trajectory can be changed first when the depression or bulge affects the comfort of passengers during the vehicle's driving, so as to avoid damage to the vehicle and reduce the bumps of the vehicle.

[0125] In other examples, assuming that the elevation of the road surface in the local area to be planned meets the second preset condition and / or the road surface type meets the preset type, the vehicle intelligent system can decide to plan the suspension parameters. Based on this, the suspension parameters can be adjusted preferentially when the depression is shallow or the bulge is not high, so as to maximize the safety and comfort of the vehicle when passing through the above local area without changing the global driving path.

[0126] For example, the vehicle-mounted intelligent system can plan the suspension parameters of the vehicle when passing through the local area to be planned according to the specific road elevation value and / or specific road type of the local area to be planned. As an example, the correspondence between the specific road elevation value or specific road type and the suspension parameters can be pre-set in the vehicle-mounted intelligent system according to the suspension performance, historical experience, etc., and this application does not make specific limitations. Among them, the suspension parameters may include but are not limited to one or more of suspension height, suspension damping, and suspension stiffness.

[0127] When performing local trajectory planning, as an example, the vehicle-mounted intelligent system can adjust the global motion path within a preset range of the global motion path according to the road information to obtain a second driving path, and then perform speed planning on the second driving path to obtain a local planned trajectory of the vehicle. The deviation between the second driving path and the global motion path is less than a preset threshold.

[0128] When performing local trajectory planning, as another example, the vehicle-mounted intelligent system can complete the local trajectory planning by following steps 1 and 2:

[0129] Step 1: Perform rough planning of the local motion trajectory to obtain the local rough trajectory.

[0130] As an example, the vehicle-mounted intelligent system can perform rough planning of local motion trajectories within a preset range of the global motion path. In other words, the deviation between the local trajectory planning and the global motion path needs to be less than a preset threshold. Based on this, it can be ensured that the local trajectory planning will not deviate too much from the global optimal path, which can avoid the waste of computing power and vehicle power resources on the one hand, and ensure the safety of vehicle driving on the other hand. Fig. 9 Taking the scene shown in the figure as an example, the preset range of the global motion path can be as follows Fig. 9 The area between Boundary 1 and Boundary 2 is shown.

[0131] Among them, the conditions for limiting the deviation between local trajectory planning and the global motion path may be related to one or more factors such as the perception capability of the vehicle perception module, the computing capability of the on-board intelligent system, the size of the vehicle's drivable area, the vehicle's surrounding environment, and scene requirements. This application does not make specific limitations.

[0132] Exemplarily, the vehicle-mounted intelligent system can search for driving states within a preset range of the global motion path to find a safer, more comfortable, and more efficient driving state, and then find a safer, more comfortable, and more efficient driving trajectory.

[0133] like Fig.11 (a) or Fig.11 As shown in (b) of Figure 1, after searching the driving state, the vehicle intelligent system has more explorations different from the global motion path within the preset range of the global motion path. Through this exploration, the vehicle intelligent system can find a safer, more comfortable and more efficient driving trajectory. Fig.11 In the driving state search shown in (a), the vehicle intelligent system can explore Fig.12 The local rough trajectory shown in (a); Fig.11 In the driving state search shown in (b), the vehicle intelligent system can explore Fig.12 The local rough trajectory shown in (b).

[0134] It can be seen that if according to Fig.11 The vehicle follows the previously planned global driving path shown in (a) above. All four wheels of the vehicle will pass over the speed bumps on the road. Compared with the global driving path, if the vehicle follows the previously planned global driving path, Fig.12 The left front wheel and the left rear wheel of the vehicle can avoid the speed bumps on the road, which will greatly improve the safety and comfort of the vehicle during driving. Fig.11 In (b) of the figure, the vehicle follows the previously planned global driving path. All four wheels of the vehicle will pass over the speed bumps on the road. Compared with the global driving path, if the vehicle follows Fig.12When driving on the local rough trajectory shown in (b) in the figure, the right front wheel and the right rear wheel of the vehicle can avoid the speed bumps on the road, which will greatly improve the safety and comfort of the vehicle during driving.

[0135] As a possible implementation method, the vehicle-mounted intelligent system can establish a trajectory search tree to perform trajectory search within a preset range of the global motion path to perform rough planning of the local motion trajectory and obtain a local rough trajectory.

[0136] For example, please refer to Fig.13 , Fig.13 Taking the Monte Carlo tree search (MCTS) architecture as an example, a schematic diagram of a trajectory search process based on a trajectory search tree provided by an embodiment of the present application is shown. Fig.13 As shown, the trajectory search process based on the trajectory search tree mainly includes S1301-S1304:

[0137] S1301: Node selection (select).

[0138] Node selection refers to continuously selecting the best child node from the root node (such as the current driving state point of the vehicle) until the unexpanded node. For example, the vehicle intelligent system can continuously select the best child node based on a preset strategy. The specific situation of the preset strategy is not specifically limited in the embodiment of the present application.

[0139] As an example, when selecting nodes, one or more of the following strategies may be followed: branch strategy, hierarchical strategy. The branch strategy may include but is not limited to the strategy of horizontal and vertical coupling, that is, coupling the vertical action with the horizontal action. For example, the vertical action may be based on acceleration sampling, such as {-1.0, 0.0, 1.0} (unit: m / s 2 ); lateral action can be based on the sampling of the vehicle's front wheel angular rate, such as {-5.0, 0.0, 5.0} (° / s). The stratification strategy may include but is not limited to a strategy based on stratification at preset time intervals, such as stratification every 1 second. The preset time interval may be related to one or more factors such as the computing power of the vehicle's intelligent system, the size of the vehicle's drivable area, the vehicle's surrounding environment, and scene requirements, and this application does not make specific limitations.

[0140] S1302: Status expansion (expand).

[0141] State expansion refers to generating a new driving state point according to the result of node selection. For example, the state expansion is as follows: Fig.11 (a) or Fig.11 (b) shows the motion path exploration process.

[0142] S1303: Node evaluation (evaluate).

[0143] In some embodiments, node evaluation is used to evaluate the vertical comfort of a new driving state point generated after state expansion, that is, the impact of the wheel state, suspension state, passenger distribution state, and vehicle state corresponding to the new driving state point on the vertical comfort.

[0144] Exemplarily, the vehicle intelligent system can evaluate whether the vertical comfort of the new driving state point generated after the state expansion is the optimal vertical comfort according to a preset comfort evaluation algorithm. For example, the vehicle intelligent system can evaluate the vertical comfort evaluation value of the new driving state point generated after the state expansion based on the following formula 1:

[0145] S i =f(C f1 ,C f2 ,…,C fP ,S f1 ,S f2 ,…,S fL ,W); (Formula 1)

[0146] In the above formula 1, f() is the vertical comfort model, P is the number of wheels of the vehicle, and C f1 , C f2 ,……,C fP Respectively represent the wheel states of the first wheel, the second wheel, ..., and the Pth wheel of the vehicle, and the wheel states represent the position of the wheel, the road elevation and / or road type corresponding to the position, and one or more of the wheel speed, acceleration, and tire pressure. Exemplarily, p∈[1,P], T is the number of sampling points (such as T = 5), is the position of the pth wheel, t is the sampling mark, is the speed of the pth wheel, is the acceleration of the pth wheel, is the elevation of the road surface that the pth wheel passes through, is the road surface type of the road surface that the pth wheel passes through, p p is the tire pressure of the pth wheel.

[0147] Among them, the above-mentioned number of sampling points can be determined by a preset time step, which may be related to one or more factors such as the computing power of the vehicle-mounted intelligent system, the size of the vehicle's drivable area, the vehicle's surrounding environment, and scene requirements, and this application does not make specific restrictions. For example, in some scenarios where it is necessary to accurately capture road elevation information, the time step can be set as small as possible. For another example, when the computing resources of the vehicle-mounted intelligent system are insufficient, the time step can be set as large as possible.

[0148] In the above formula 1, L is the number of suspensions of the vehicle, S f1 , S f2 ,……,S fL Respectively represent the suspension states of the first suspension, the second suspension, ..., the Lth suspension of the vehicle, and the suspension state represents one or more of the mode, load, stiffness parameter and damping parameter of the suspension. fl =[m l ,F l ,k l ,c l ], l∈[1,L],m l is the first suspension mode, F l is the load of the first suspension, k l is the stiffness parameter of the first suspension, c l is the damping parameter of the first suspension. The suspension modes, such as sports mode, standard mode, comfort mode, etc., depend on the specific hardware configuration of the suspension.

[0149] In the above formula 1, W is the vehicle state and / or the passenger distribution state. The vehicle state indicates the wheelbase and / or track width of the vehicle. The passenger distribution state indicates whether there are passengers in the multiple seats of the vehicle. For example, W = [wb, td, r], wb is the wheelbase of the vehicle, td is the track width of the vehicle, and r is the passenger distribution state of the vehicle. For example, if the vehicle is a five-seater vehicle, r = [r fl , r fr , r rl , r rm , r rr ], where r fl 、r fr 、r rl 、r rm and r rr It indicates whether there are passengers in the driver's seat, the front passenger seat, the rear left seat, the rear middle seat and the rear right seat. For example, if there are passengers, the corresponding value is 1, and if there are no passengers, the corresponding value is 0. For vehicles with other numbers of seats, the example of the five-seat vehicle mentioned above can be referred to, and no further description is given here.

[0150] As a possible implementation, the vertical comfort model f() in Formula 1 may be a manually designed mathematical formula, or a supervised trained neural network or a neural network obtained by reinforcement learning, which is not specifically limited in the present embodiment. Taking f() as an example, Fig.14 As shown, according to the wheel state corresponding to the new driving state point, the road surface elevation, road surface type, suspension state, vehicle state, passenger distribution state and other information, the supervised training neural network can obtain the vertical comfort evaluation value of the new driving state point based on the vertical comfort model f().

[0151] The supervised training neural network is obtained by learning from a large amount of training data. In some embodiments, the vehicle intelligent system can continuously optimize the parameters of the supervised training neural network based on information such as the wheel status of the vehicle during driving, the road elevation of the road, the road type, the suspension status, the vehicle status, and the passenger distribution status. For example, please refer to Fig.15 , Fig.15 A schematic diagram of a parameter optimization process of a supervised training neural network provided in an embodiment of the present application is shown.

[0152] like Fig.15 As shown, it is assumed that the first vertical comfort evaluation value (such as s) can be obtained by evaluating the wheel state, the road elevation of the road passed, the road type, the suspension state, the vehicle state, the passenger distribution state and other information during the vehicle's driving along the local planning trajectory based on the current supervised training neural network. The on-board intelligent system can collect one or more of the following parameters during the vehicle's driving along the local planning trajectory: motion parameters at multiple seats, suspension state parameters, tire pressure parameters of the wheels, and calculate the second vertical comfort evaluation value (such as s') based on the one or more parameters. Further, the on-board intelligent system can optimize the supervised training neural network based on the relationship between the second vertical comfort evaluation value and the first vertical comfort evaluation value. For example, the on-board intelligent system can determine the error (loss) between the second vertical comfort evaluation value and the first vertical comfort evaluation value by comparing the two, and then update the network parameters of the supervised training neural network through back propagation based on the error.

[0153] As an example, the vehicle intelligent system can calculate the second vertical comfort evaluation value based on a preset rule and one or more parameters collected during the vehicle's driving along the local planning trajectory. For example, the preset rule can be a predefined post-processing function g(), and the vehicle intelligent system can calculate the second vertical comfort evaluation value based on the post-processing function g():

[0154] The second vertical comfort evaluation value=g(W).

[0155] Taking a five-seater vehicle as an example, W = [W fl , W fr , W rl , W rm , W rr , S f , P p ],W fl , W fr , W rl , W rm and W rrThey represent the displacement, velocity, acceleration and other motion parameters of the driver's seat, the co-pilot seat, the rear left seat, the rear middle seat and the rear right seat during the vehicle's driving along the local planning trajectory. f represents the state parameters of the suspension when the vehicle is traveling along the local planning trajectory, P p It represents the tire pressure parameter of the wheels when the vehicle is driving along the local planning trajectory. For vehicles with other numbers of seats, the above example of a five-seat vehicle can be referred to, and no further description is given here.

[0156] Among them, the above-mentioned post-processing function g() can be pre-set based on historical experience, related algorithms, etc., and the specific algorithm is not specifically limited in this application, and can be flexibly adjusted according to the situation. In addition, the above embodiment only takes the network parameter optimization of the neural network supervised by the vehicle intelligent system as an example, and the embodiment of this application does not limit the specific execution subject and specific process of the network parameter optimization. For example, the vehicle can also send relevant data to the corresponding cloud server, and the cloud server will perform the corresponding supervised training of the neural network network parameter optimization.

[0157] In some embodiments, the node evaluation is also used to evaluate whether the new driving state point generated after the state expansion satisfies one or more of the following: collision constraints, dynamic constraints (such as curvature constraints, etc.), global deviation constraints, XY comfort constraints, etc. Among them, the collision constraint is used to ensure that the new driving state point is collision-free to ensure safety, for example, to avoid the vehicle at the state point from experiencing the following Fig.16 The dynamic constraint condition is used to ensure the followability of the vehicle to the new driving state point, for example, to avoid the vehicle at this state point from experiencing the following Fig.16 The global deviation constraint is used to ensure that the new driving state point does not deviate too much from the global optimal path, so as to avoid the waste of computing power and vehicle power resources while ensuring the safety of vehicle driving. For example, avoid the vehicle at this state point from experiencing the following Fig.16 The XY comfort constraint condition is used to ensure the lateral / longitudinal comfort of the new driving state point. For example, the lateral / longitudinal comfort of the new driving state point may be as follows: Fig.16 As shown in (c) in .

[0158] It can be understood that, through node evaluation, the optimal state point in the trajectory search process can be determined, and then the convergence direction of the trajectory search tree can be determined.

[0159] S1304: Node backtracking (backup).

[0160] Node backtracking refers to tracing back from the new driving status point to the upper node layer by layer until the root node.

[0161] By continuously repeating the above S1301 - S1304 , a series of state points can be searched within a preset range of the global motion path, and then a local rough trajectory is finally obtained.

[0162] Among them, the on-board intelligent system can repeat the above S1301-S1304 until the predefined conditions are met. For example, the predefined conditions may be exhaustion of computing resources, the algorithm time reaching a limit threshold, or the search tree nodes reaching a limit number, etc. The embodiments of the present application do not make specific limitations.

[0163] It should be noted that the above MCTS architecture is only an example of a trajectory search tree. In practical applications, the embodiments of the present application do not limit the specific architecture of the search tree. For example, the architecture of the search tree may also be a Hybrid A* architecture.

[0164] It can be understood that based on the rough planning of the local motion trajectory, the on-board intelligent system can search the motion state of each wheel within the preset range of the global motion path and obtain the road surface elevation and / or road surface type of each state point passed by the wheel, and use it as input information to evaluate the vertical comfort of the vehicle when passing the above-mentioned state points, thereby assisting in the generation of the final local rough trajectory.

[0165] Step 2: Smooth the motion trajectory after rough planning to obtain a local smooth trajectory.

[0166] It is understandable that the node size obtained based on the rough planning of the local motion trajectory is limited. For example, the rough planning of the local motion trajectory often restricts the state space and / or prunes the node size of the search tree to save computing power as much as possible while providing a high-quality initial solution. However, since the rough planning of the local motion trajectory restricts the node size, the curvature of the obtained local rough trajectory is usually discontinuous, which may cause errors when subsequent vehicles follow the trajectory. Based on this consideration, in an embodiment of the present application, the on-board intelligent system can smooth the motion trajectory after rough planning to obtain a local motion trajectory, so that the curvature of the local motion trajectory is within the following ability of the vehicle, thereby improving the followability of the path.

[0167] As a possible implementation method, the vehicle-mounted intelligent system can use a nonlinear optimization method to smooth the motion trajectory after rough planning. Exemplarily, the nonlinear optimization method can take into account but is not limited to constraint conditions and / or obstacle avoidance conditions.

[0168] As an example, a curvature constraint condition is that the curvature of any state point on the local motion trajectory cannot exceed the steering capability of the vehicle. For example, the minimum curvature radius of the vehicle often determines the steering capability of the vehicle. Based on this, in the embodiment of the present application, the vehicle-mounted intelligent system can use a curvature radius greater than the minimum curvature radius as a curvature constraint condition to smooth the motion trajectory after rough planning.

[0169] As an example, the obstacle avoidance condition can be determined based on an optimization model for precise obstacle avoidance, such as a convex optimization model (such as sequential convex optimization). Among them, the basic idea of ​​convex optimization is to determine the local optimal solution near the rough path based on the result of the rough path search as the initial value. For a specific introduction to the optimization model for precise obstacle avoidance, such as the convex optimization model, reference can be made to conventional technology, and the embodiments of this application will not be repeated.

[0170] In some embodiments, if the planning of step S803 fails, such as both local trajectory planning and suspension parameter planning fail, the on-board intelligent system can also notify the user so that the user can understand the specific road elevation and / or road type details so that the user can take corresponding manual adjustment measures based on this.

[0171] S804: Perform corresponding trajectory following and / or parameter setting according to the planning result.

[0172] For example, if S803 outputs a local motion trajectory, the control module of the vehicle intelligent system (such as Figure 3 The control module 340 shown in the figure can perform subsequent high-precision vehicle trajectory following based on this. For example, the control module of the vehicle-mounted intelligent system can send control instructions to the vehicle chassis through a specific interface (such as a serial bus module) so that the vehicle travels according to the planned local motion trajectory.

[0173] For example, if S803 outputs the suspension parameter adjustment result, the control module of the vehicle intelligent system (such as Figure 3 The control module 340 shown in the figure can perform subsequent suspension parameter settings based on this. Exemplarily, the control module of the vehicle-mounted intelligent system can send parameter setting instructions to the vehicle suspension through a specific interface (such as a serial bus module) to complete the setting of the suspension parameters. For example, if the vehicle's multiple suspensions are independent of each other, the control module can send parameter setting instructions to the multiple suspensions respectively; if the vehicle includes an active stabilizer bar fixed with multiple suspensions, the control module can complete the suspension parameter setting by adjusting the angle and torque of the active stabilizer bar.

[0174] It can be understood that the motion planning method for the vehicle provided in the embodiment of the present application fully considers the influence of factors such as road bumps, road depressions, road type, wheel status, suspension status, etc. on the vertical comfort of the vehicle during driving. On the basis of the global driving path planned by the vehicle, a more fine-grained vehicle local motion planning is performed in advance by combining the acquired road elevation and / or road type and other road vertical environmental factors, such as local trajectory planning and / or vehicle parameter adjustment that affects comfort. Based on this solution, no matter what driving scene the vehicle is in (such as a low-speed parking scene, a medium-to-low-speed unstructured road scene, a medium-to-high-speed structured road scene, etc.), what type of road surface, whether the road elevation is abnormal, what the wheel status is, what the suspension status is, the vehicle can perform more flexible and detailed local motion planning for each wheel in advance based on the actual environment to ensure that the vehicle always maintains a high degree of comfort during driving.

[0175] Among them, the motion planning method of the above-mentioned vehicle provided in the embodiment of the present application can be applicable to any type of suspension, such as active suspension, semi-active suspension or passive suspension.

[0176] It can be understood that the stiffness characteristics and damping characteristics of the active suspension can be dynamically and adaptively adjusted according to the driving conditions of the vehicle, such as the wheel state, road surface state, etc., so that the suspension system is always in the best vibration reduction state. For the active suspension, based on the motion planning method of the vehicle provided in the above embodiment, although the on-board intelligent system can instruct the suspension to set the corresponding suspension parameters according to the results of local motion planning (such as the suspension parameter adjustment results), in this process, the suspension passively responds to the suspension parameter adjustment instructions of the on-board intelligent system, and the advantages of the suspension itself, such as large adjustable space and strong self-adjustment ability, are not well utilized. Based on this consideration, the present application also provides a possible implementation architecture, based on which the suspension state, the adjustable space and self-adjustment ability of the suspension can be fully utilized to achieve a deep integration of the algorithm and the actuator, and based on this, the vertical comfort model can be better optimized.

[0177] As an example, the first parameter can be obtained based on the wheel state of the vehicle during driving, the road surface elevation of the road surface passed, the road surface type, the suspension state, the vehicle state, the passenger distribution state and other information. Furthermore, based on the local planning trajectory and / or the suspension parameter adjustment strategy based on the road information and the global driving path provided in the embodiment of the present application, the second parameter can be obtained, such as the action parameter corresponding to the specific local planning trajectory and / or the suspension parameter adjustment strategy. And, the third parameter, such as the reward sequence corresponding to the vertical comfort evaluation value, can be obtained by evaluating the wheel state of the vehicle during driving along the local planning trajectory, the road surface elevation of the road surface passed, the road surface type, the suspension state, the vehicle state, the passenger distribution state and other information through a preset vertical comfort evaluation function. Finally, the vertical comfort model is trained according to the first parameter, the second parameter and the third parameter, and the network parameters of the vertical comfort model are optimized based on the training results. For example, please refer to Fig.17 , Fig.17 FIG. 1 shows an implementation architecture diagram of a vertical comfort model optimization solution provided by an embodiment of the present application. Fig.17 As shown, the implementation architecture includes an on-board intelligent module 1710 and a vehicle / suspension model module 1720.

[0178] Among them, the vehicle-mounted intelligent module 1710 can perform local motion planning based on the acquired road information (such as the drivable area and the elevation and / or road type of the drivable area), the global driving path, the wheel state, etc., such as performing local trajectory search by constructing a trajectory search tree. In addition, in the process of performing local trajectory search, the vehicle-mounted intelligent module 1710 can simultaneously consider the vehicle's dynamic model and vertical comfort model, and use one or more of the following constraints to evaluate the searched state points to obtain state points that meet certain safety and comfort requirements: collision constraints, dynamic constraints (such as curvature constraints, etc.), global deviation constraints, XY comfort constraints, etc.

[0179] The vehicle / suspension model module 1720 includes but is not limited to the constructed environment model, suspension model, vehicle model and perception model. Among them, the environment model can represent the excitation of the vehicle by the real vehicle surrounding environment and the road vertical environment factors. Of course, the environment model can also be the excitation of the vehicle by the virtual environment constructed by environmental simulation to simulate the real environment, which is not specifically limited in the embodiment of the present application. The suspension model represents the dynamic model and suspension control model of the suspension. The suspension model can be a real vehicle suspension system or a suspension simulation model constructed artificially. The embodiment of the present application does not make specific restrictions. Among them, for the real vehicle suspension system, the suspension control model represents the controller adapted by the suspension and its control logic; for the artificially constructed suspension simulation model, the suspension control model is a model assigned with the control logic. In the embodiment of the present application, the suspension control model can receive the state points searched by the vehicle-mounted intelligent module 1710 in real time as the environmental information preview. The vehicle model represents the dynamic model of the vehicle. The vehicle model can be a dynamic model of a real vehicle or a simulation model of a vehicle constructed artificially. The embodiment of the present application does not make specific restrictions. The perception model represents one or more sensors on the vehicle for perceiving the environment / motion, such as a TOF sensor, a camera (such as an infrared camera or other visible light camera), a radar (such as a millimeter wave radar, a laser radar, an ultrasonic radar, etc.), a vehicle posture sensor (such as a gyroscope), an IMU, etc. The perception model can be one or more sensors on a real vehicle, or a sensor simulation model artificially constructed, and the embodiments of the present application do not make specific limitations. For example, the perception model can be used to perceive the motion parameters of each seat during the movement of the vehicle, such as displacement, velocity, acceleration, etc.

[0180] like Fig.17 As shown, through the interaction between the on-board intelligent module 1710 and the vehicle / suspension model module 1720, a set of state-action sequences can be generated. On this basis, combined with the motion parameters of each seat during the vehicle movement perceived by the perception model, a set of state-action-reward sequences can be obtained through a preset vertical comfort evaluation function, which is called a "track" in reinforcement learning. Using the same method, after multiple interactions between the on-board intelligent module 1710 and the vehicle / suspension model module 1720, multiple "tracks" can be generated in the state-action-reward sequence pool. Further, after multiple "tracks" (i.e., multiple groups of state-action-reward sequences) are trained for the vertical comfort model, the network parameters of the vertical comfort model in the on-board intelligent module 1710 can be optimized based on the training results. The present application can continuously repeat the above process to achieve intensive training of the vertical comfort model. Fig.17 The vertical comfort model is also called a neural network trained by reinforcement learning.

[0181] Understandably, based on Fig.17 The architecture can achieve deep integration of local trajectory search and evaluation network. The output result of the evaluation network is not a simple comfort score, but an overall evaluation that takes into account the decision logic and ability of the search tree and the control ability of the suspension with preview. Therefore, it can further ensure the reliability, safety and efficiency of local motion planning, and further improve the stability of the vehicle during driving, ensuring that the vehicle always maintains a high degree of comfort during driving.

[0182] It should be understood that the various schemes of the embodiments of the present application can be used in reasonable combination, and the explanations or descriptions of the various terms appearing in the embodiments can be mutually referenced or explained in the various embodiments, without limitation.

[0183] It should also be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0184] It is understandable that in order to realize the functions of any of the above-mentioned embodiments, the vehicle-mounted intelligent system includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0185] The embodiment of the present application can divide the vehicle-mounted intelligent system into functional modules. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0186] It should also be understood that each module in the controller or resource allocation simulation device can be implemented in software and / or hardware form, and there is no specific limitation on this. In other words, the electronic device is presented in the form of functional modules. The "module" here can refer to a specific application integrated circuit ASIC, a circuit, a processor and a memory that executes one or more software or firmware programs, an integrated logic circuit, and / or other devices that can provide the above functions.

[0187] In an optional manner, when data transmission is implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is implemented in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disk (DVD)), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0188] The steps of the method or algorithm described in conjunction with the embodiments of the present application can be implemented in hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, mobile hard disk, CD-ROM or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in an on-board intelligent system. Of course, the processor and the storage medium can also exist as discrete components.

[0189] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

Claims

1. A motion planning method for a vehicle, It is characterized in that The method comprises: Acquire a first driving path and road information of a vehicle, wherein the first driving path represents a planned path of the vehicle, and the road information includes a drivable area of ​​the vehicle, and a road elevation and / or road type of the drivable area; Local trajectory planning and / or suspension parameter adjustment of the vehicle are performed according to the road information and the first driving path.

2. The method according to claim 1, It is characterized in that The performing local trajectory planning according to the road information and the first driving path includes: A trajectory search is performed within a preset range of the first driving path in the drivable area to obtain a local planned trajectory of the vehicle, wherein a vertical comfort evaluation value corresponding to the local planned trajectory is greater than a vertical comfort evaluation value corresponding to the first driving path.

3. The method according to claim 1, It is characterized in that The performing local trajectory planning according to the road information and the first driving path includes: Adjusting the first driving path within a preset range of the first driving path according to the road information to obtain a second driving path; Speed ​​planning is performed on the second driving path to obtain a local planned trajectory of the vehicle, and a vertical comfort evaluation value corresponding to the local planned trajectory is greater than a vertical comfort evaluation value corresponding to the first driving path.

4. The method according to claim 2 or 3, It is characterized in that The vertical comfort evaluation value is related to one or more of the following: wheel status, suspension status, passenger distribution status, vehicle status; Among them, the wheel state represents the positions of multiple wheels of the vehicle, the road surface elevation and / or road surface type corresponding to the positions, and one or more of the wheel speed, acceleration and tire pressure; the suspension state represents one or more of the modes, loads, stiffness parameters and damping parameters of multiple suspensions of the vehicle; the passenger distribution state indicates whether there are passengers on multiple seats of the vehicle; the vehicle state represents one or more of the wheelbase and track of the vehicle.

5. The method according to any one of claims 1 to 4, It is characterized in that The performing of local trajectory planning and / or suspension parameter adjustment of the vehicle according to the road information and the first driving path includes: When the first driving path includes a first area whose road elevation satisfies a first preset condition, the local trajectory planning is performed according to the position information corresponding to the drivable area and the first area; Among them, when driving based on the first driving path, M wheels of the vehicle pass through the first area, M is a positive integer; when driving based on the local planned trajectory, N wheels of the vehicle pass through the first area, N is an integer, and N<M.

6. The method according to any one of claims 1 to 5, It is characterized in that The performing of local trajectory planning and / or suspension parameter adjustment of the vehicle according to the road information and the first driving path further includes: When the first driving path includes a first area whose road elevation satisfies a first preset condition, the suspension parameters of the vehicle are adjusted according to the road elevation corresponding to the first area.

7. The method according to any one of claims 1 to 6, It is characterized in that The performing of local trajectory planning and / or suspension parameter adjustment of the vehicle according to the road information and the first driving path further includes: When the first driving path includes a second area whose road surface type is a preset type, the suspension parameters of the vehicle when passing through the second area are adjusted according to the road surface type corresponding to the second area.

8. The method according to any one of claims 2 to 7, It is characterized in that The vertical comfort evaluation value is obtained based on the following formula 1: S i =f(C f1 ,C f2 ,…,C fP ,S f1 ,S f2 ,…,S fL ,W); (Formula 1) Among them, S i is the vertical comfort evaluation value, f() is the vertical comfort model, P is the number of wheels of the vehicle, C f1 , C f2 ,……,C fP represents the wheel status of the first wheel, the second wheel, ..., the Pth wheel of the vehicle respectively, and the wheel status represents the position of the wheel, the road elevation and / or road type corresponding to the position, and one or more of the wheel speed, acceleration and tire pressure; L is the number of suspensions of the vehicle, S f1 , S f2 ,……,S fL They respectively represent the suspension states of the first suspension, the second suspension, ..., the Lth suspension of the vehicle, and the suspension states represent one or more of the suspension mode, load, stiffness parameter and damping parameter; W is the vehicle state and / or the passenger distribution state, the vehicle state represents the wheelbase and / or track of the vehicle, and the passenger distribution state represents whether there are passengers on multiple seats of the vehicle.

9. The method according to claim 8, It is characterized in that The method further comprises: Determining a first vertical comfort evaluation value corresponding to the local planning trajectory based on the vertical comfort model; Acquiring parameters of the vehicle during driving according to the local planned trajectory, the parameters including one or more of the following information: motion parameters at the multiple seats, state parameters of the suspension, and tire pressure parameters of the wheels; Based on a preset rule, determining a second vertical comfort evaluation value according to the parameter; The vertical comfort model is optimized according to the relationship between the second vertical comfort evaluation value and the first vertical comfort evaluation value.

10. The method according to claim 9, It is characterized in that The method further comprises: The historical driving trajectory of the vehicle and the corresponding first vertical comfort evaluation value are trained and learned to determine the preset rule, wherein the historical driving trajectory represents the historical driving path and the corresponding speed strategy.

11. The method according to claim 8, It is characterized in that The method further comprises: Determine a first parameter based on a first driving path and road information of the vehicle: Determine the second parameter based on the local planning trajectory and / or the suspension parameter adjustment strategy: Determining a third parameter when the vehicle travels according to the local planned trajectory based on the vertical comfort model, wherein the third parameter represents a vertical comfort value; The vertical comfort model is optimized according to the first parameter, the second parameter and the third parameter.

12. The method according to any one of claims 2 to 11, It is characterized in that The local planned trajectory of the vehicle also satisfies one or more of the following conditions: collision constraint, dynamic constraint, global deviation constraint, and XY comfort constraint.

13. The method according to any one of claims 1 to 12, It is characterized in that The vehicle includes a positioning device, and obtaining the drivable area of ​​the vehicle includes: Collecting the position information of the vehicle through the positioning device; The drivable area is determined according to the position information of the vehicle and the surrounding environment information of the vehicle.

14. The method according to claim 13, It is characterized in that The vehicle further comprises a sensing device, and the surrounding environment information is obtained based on data collected by the sensing device; The perception device includes a perception sensor, and the perception sensor includes one or more of the following: a laser radar, a visible light camera, and a millimeter wave radar.

15. The method according to claim 14, It is characterized in that The road surface elevation and the road surface type are obtained based on data collected by the sensing device.

16. The method according to claim 14, It is characterized in that The road surface elevation and the road surface type are obtained according to one or more of the following: a received map, and road surface information shared by other vehicles.

17. The method according to any one of claims 1 to 16, It is characterized in that The suspension of the vehicle is an active suspension, a semi-active suspension or a passive suspension.

18. The method according to any one of claims 1 to 17, It is characterized in that The suspension parameters include one or more of the following: suspension height, suspension damping, and suspension stiffness.

19. A motion trajectory planning device, It is characterized in that The motion trajectory planning device comprises: A data acquisition module, used to collect or obtain information; a memory for storing computer program instructions; A processor is used to execute the computer program instructions to support the motion trajectory planning device to implement the method as described in any one of claims 1-18 based on the information collected and / or acquired by the data acquisition module.

20. A vehicle-mounted system, It is characterized in that The vehicle-mounted system includes the motion trajectory planning device as described in claim 19.

21. A vehicle, It is characterized in that The vehicle includes the motion trajectory planning device as described in claim 19, or the vehicle-mounted system as described in claim 20.

22. A readable storage medium, It is characterized in that The readable storage medium stores program instructions, and when the program instructions are executed by the device, the method according to any one of claims 1 to 18 is implemented.

23. A program product comprising instructions, It is characterized in that When the program product is run on a device, the device is caused to execute the method according to any one of claims 1 to 18.

24. A chip system, It is characterized in that The chip system includes a processing circuit and a storage medium, wherein the storage medium stores program instructions; when the program instructions are executed by the processing circuit, the method according to any one of claims 1 to 18 is implemented.

Images