WO2022160100A1 - 可移动平台的控制方法和装置 - Google Patents

可移动平台的控制方法和装置 Download PDF

Info

Publication number
WO2022160100A1
WO2022160100A1 PCT/CN2021/073832 CN2021073832W WO2022160100A1 WO 2022160100 A1 WO2022160100 A1 WO 2022160100A1 CN 2021073832 W CN2021073832 W CN 2021073832W WO 2022160100 A1 WO2022160100 A1 WO 2022160100A1
Authority
WO
WIPO (PCT)
Prior art keywords
route
lane
target
segment
block
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2021/073832
Other languages
English (en)
French (fr)
Inventor
罗元福
吴易霖
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
SZ DJI Technology Co Ltd
Original Assignee
SZ DJI Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by SZ DJI Technology Co Ltd filed Critical SZ DJI Technology Co Ltd
Priority to CN202180083247.2A priority Critical patent/CN116745581A/zh
Priority to PCT/CN2021/073832 priority patent/WO2022160100A1/zh
Publication of WO2022160100A1 publication Critical patent/WO2022160100A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34—Route searching; Route guidance

Definitions

  • the present application relates to the technical field of mobile platforms, and more particularly, to a control method and device for a mobile platform.
  • navigation route planning is an integral part.
  • the navigation route planning of autonomous vehicles is to plan routes according to the origin, destination and road network information.
  • a route is a route from an origin to a destination consisting of multiple lane segments.
  • the self-driving vehicle generates a driving strategy according to the route, and controls the driving of the self-driving vehicle through the driving strategy, so that the self-driving vehicle travels along the path of the route.
  • the currently planned routes are single and unable to meet diverse needs.
  • the purpose of the present application is to provide a control method and device for a mobile platform, which are used to solve the problem that the currently planned flight routes are single and cannot meet diversified demands.
  • the present application discloses a control method for a movable platform, including:
  • the multiple routes include a first route and a second route, the first route includes a first route segment, the second route includes a second route segment, and the first route segment and the second route segment are located at the target different lanes of the path;
  • the movable platform is controlled to move along the target path.
  • control device for a movable platform including: a memory and a processor;
  • the multiple routes include a first route and a second route, the first route includes a first route segment, the second route includes a second route segment, and the first route segment and the second route segment are located at the target different lanes of the path;
  • the movable platform is controlled to move along the target path.
  • the present application discloses a movable platform, including the control device of the movable platform according to the second aspect.
  • an embodiment of the present application discloses a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and when the instructions are executed on a computer, the computer executes the method described in the first aspect .
  • an embodiment of the present application discloses a computer program product, including computer instructions, which implement the method of the first aspect when the computer instructions are executed by a processor.
  • the present application discloses a control method and device for a movable platform, by generating multiple routes of the same target route according to the lane distribution information of the target route, and then selecting the target route from the multiple routes to control the automatic driving.
  • the vehicle is moving.
  • the lanes on the same road between different routes are not exactly the same. Therefore, the present application provides more options for routes for autonomous vehicles to meet user preferences and requirements in complex environments.
  • FIG. 1 is a schematic architectural diagram of an autonomous vehicle according to an embodiment of the present application
  • FIG. 2 is a schematic diagram of an application scenario provided by an embodiment of the present application.
  • FIG. 3 is a flowchart of a method for controlling a movable platform according to an embodiment of the present application
  • FIG. 4 is a schematic diagram of lane distribution information provided by an embodiment of the present application.
  • FIG. 5 is a flowchart of a method for controlling a movable platform provided by another embodiment of the present application.
  • FIG. 6 is a schematic schematic diagram of a control method of a movable platform provided by the present application.
  • FIGS. 7a-7c are schematic diagrams of generating a route according to a preset motion strategy according to an embodiment of the present application, respectively;
  • FIG. 8 is a flowchart of a method for controlling a movable platform provided by another embodiment of the present application.
  • FIG. 9 is a schematic diagram of determining a steerable area of a flight route provided by an embodiment of the present application.
  • FIG. 10 is a flowchart of a method for controlling a movable platform provided by another embodiment of the present application.
  • FIG. 11 is a flowchart of a method for controlling a movable platform provided by another embodiment of the present application.
  • FIG. 12 is a schematic structural diagram of a control device for a movable platform provided by an embodiment of the application.
  • FIG. 13 is a schematic structural diagram of a movable platform according to an embodiment of the present application.
  • a component when referred to as being "fixed to" another component, it can be directly on the other component or there may also be a centered component. When a component is considered to be “connected” to another component, it may be directly connected to the other component or there may be an intervening component at the same time.
  • Embodiments of the present application provide a control method and device for a movable platform, where the movable platform may be an unmanned aerial vehicle, an unmanned vehicle, an unmanned ship, a robot, or an autonomous vehicle.
  • FIG. 1 is a schematic architectural diagram of an autonomous driving vehicle according to an embodiment of the present application.
  • the autonomous vehicle 100 may include a perception system 110 , a control system 120 and a mechanical system 130 .
  • the perception system 110 is used to measure the status information of the autonomous vehicle 100 , that is, the perception data of the autonomous vehicle 100 , and the perception data may represent the position information and/or status information of the autonomous vehicle 100 , such as position, angle, speed, acceleration and angular velocity, etc.
  • the perception system 110 may include, for example, vision sensors (eg, including multiple monocular or binocular vision devices), lidars, millimeter-wave radars, inertial measurement units (Inertial Measurement Unit, IMU), global navigation satellite systems, gyroscopes, ultrasonic sensors At least one of sensors such as , electronic compass, and barometer.
  • the global navigation satellite system may be the Global Positioning System (GPS).
  • the sensing data may be transmitted to the control system 120 .
  • the control system 120 is used to make decisions for controlling how the autonomous driving vehicle 100 drives according to the sensing data, for example: how much speed to drive, or how much braking acceleration to brake, or whether to change lanes, or, Left/right turn driving, etc.
  • the control system 120 may include, for example, a computing platform, such as a vehicle-mounted supercomputing platform, or at least one of devices with processing functions, such as a central processing unit, a distributed processing unit, and the like.
  • the control system 120 may also include communication links for various data transmissions on the vehicle.
  • Control system 120 may output one or more control commands to mechanical system 130 according to the determined decision.
  • the mechanical system 130 is used to control the autonomous driving vehicle 100 in response to one or more control commands from the control system 120 to complete the above decision.
  • Driving of the vehicle 100 provides power, wherein the rotational speed of the wheels can affect the speed of the autonomous vehicle.
  • the mechanical system 130 may include, for example, at least one of a mechanical body engine/motor, a controlled wire-by-wire system, and the like.
  • FIG. 2 is a schematic diagram of an application scenario provided by an embodiment of the application.
  • the user sets the origin and destination, and the autonomous vehicle 100 can obtain the origin and destination according to the origin and destination.
  • the target path to the destination for example, through road A, then through road B, and then through road C, etc.
  • the autonomous driving vehicle 100 when the autonomous driving vehicle 100 is driving on the road, since the road may include one or more lanes in the same driving direction, the autonomous driving vehicle will also plan a lane on which lane it should travel on each road. route.
  • the route planned by the autonomous vehicle is only one, which cannot meet the diversified needs.
  • the self-driving vehicle Take the self-driving vehicle as an example that needs to travel through the roundabout road. If the user prefers efficiency, that is, the shorter the driving time of the self-driving vehicle, the best way for the self-driving vehicle to travel along the inner ring lane. If the user prefers comfort, That is, the number of times the autonomous vehicle turns to lane (lane change) is the least, and the autonomous vehicle travels best along the outer ring lane. However, a route currently planned cannot meet the above requirements at the same time.
  • the self-driving vehicle will stop behind the obstacle in order to avoid the obstacle and cannot continue to move forward, which cannot meet the requirements of the self-driving vehicle. requirements in complex environments.
  • the driving of the autonomous vehicle is controlled by generating multiple routes and then selecting a target route from the multiple routes, wherein the lanes on the same road between different routes are not exactly the same, which provides the automatic driving vehicle 100 with more information.
  • the multi-route selectable space can meet the user's preference demand and the demand in a complex environment. For the specific implementation process, refer to the following embodiments of the present application.
  • FIG. 3 is a flowchart of a method for controlling a movable platform provided by an embodiment of the present application. As shown in FIG. 3 , the method in this embodiment may include:
  • the multiple routes include a first route and a second route, the first route includes a first route segment, the second route includes a second route segment, the first route segment and the second route segment Different lanes on the target path.
  • the target path of the automatic driving vehicle is obtained.
  • the starting point, destination and road network information of the automatic driving vehicle can be obtained.
  • the lane distribution information of the target path is obtained, and the lane distribution on each road in the target path can be obtained according to the lane distribution information.
  • the lane distribution information can be obtained through high-precision road network information.
  • multiple routes are generated according to the lane distribution information of the target path. Taking any two routes among the multiple routes as an example, in order to distinguish the two routes, the two routes are respectively referred to as the first route and the second route.
  • the first route includes at least one route segment. Taking one route segment in the first route as an example, the route segment may be referred to as a first route segment.
  • the second route includes at least one route segment.
  • the route segment may be referred to as a second route segment.
  • the first flight segment and the second flight segment are on different lanes of the target path.
  • one route is determined from the generated multiple routes as the target route, and then according to the target route, the autonomous driving vehicle is controlled to travel along the target route.
  • an autonomous vehicle can determine a target route from multiple routes according to an indicator representing the user's preference to meet the needs of different users.
  • an autonomous vehicle can determine a target route from multiple routes according to the environment in which the autonomous vehicle is located, so as to meet the driving needs in different complex environments.
  • the first route segment of the first route is located in the left lane of road A
  • the second route segment of the first route is located in the right lane of road A. If the determined target route is the first route, the autonomous driving vehicle is controlled to drive along the A road in the left lane, and if the determined target route is the second route, the autonomous driving vehicle is controlled to drive along the A road in the right lane.
  • the target path can also be a path on a route, and multiple routes are generated according to the lane distribution information of the road where the target route is located, thereby increasing the types of routes and providing more route selection space for autonomous vehicles. .
  • multiple routes are generated according to the lane distribution information of the target path, the multiple routes include a first route and a second route, the first route includes a first route segment, and the second route includes a second route segment, The first route segment and the second route segment are located in different lanes of the target path, and then the target route is determined according to multiple routes, and then the movable platform is controlled to move along the target route according to the target route.
  • the multiple routes generated in this application belong to the same target path (such as the same road), but are not identical at the lane level, providing more choices for the movable platform to travel along the target path to meet diverse needs.
  • the foregoing lane distribution information includes the position of the target path in one or more lane blocks of each road segment.
  • FIG. 4 is a schematic diagram of lane distribution information provided by an embodiment of the application. As shown in FIG. 4 , taking a road where the target path is located as an example, the road is divided into three lanes in the lateral direction, and the road is longitudinally divided into three lanes. The direction includes multiple road segments, and each road segment includes 3 lane blocks. Among them, lane block 1 to lane block 3 form a road segment, lane block 4 to lane block 6 form a road segment, and lane block 7 to lane Block 9 forms a road segment, and lane blocks 10 to 12 form a road segment.
  • a possible implementation manner of the above S303 is: generating multiple routes according to the positions of the target path in one or more lane blocks of each road section.
  • the first air route and the second air route include at least part of the air route segment, and some air route segments are located on the same road segment.
  • at least one same route segment may exist in the first route and the second route.
  • the multiple routes generated in the above S303 may be multiple parallel routes, and the multiple routes are different routes from the same origin to the same destination. It should be noted that the above-mentioned juxtaposition does not mean that the multiple air routes must be parallel to each other, but that the multiple air routes are all located on the above-mentioned target path and arrive at the same destination from the same starting point. These multiple routes may intersect with each other in some sections, or may be parallel in some sections.
  • each route passes through a plurality of lane blocks along the target path.
  • a possible implementation manner of the above-mentioned S303 is: according to the lane distribution information, determining multiple routes passing through multiple lane blocks along the target path.
  • One of the routes is shown in FIG. 4 , and the route passes through lane block 2 , lane block 5 , lane block 8 , lane block 7 , and lane block 10 in sequence.
  • the other route may also pass through lane area 1, lane block 4, lane block 7, lane block 10 and so on in sequence.
  • the route is formed by connecting multiple route segments, and the multiple route segments are located in the lane blocks of the continuously distributed road segments.
  • the route shown in FIG. 4 is composed of the route segment in lane block 2, the route segment in lane block 5, the route segment in lane block 8, the route segment in lane block 7, and the route segment in lane block 10. within the route segment.
  • the section to which lane block 2 belongs is adjacent to the section to which lane block 5 belongs, and the sections to which lane block 8 and lane block 7 belong are adjacent to the section to which lane block 5 belongs. 8.
  • the road section to which lane block 7 belongs is adjacent to the road section to which lane block 10 belongs.
  • the above-mentioned route includes a steering route segment, the steering route segment includes an adjacent first route segment and a second route segment, and the target route includes the first route segment and the second road segment, the first route segment is located in the first lane block of the first road segment, the second route segment is located in the second lane block of the second road segment, and the first lane block and the second lane block are located in the Adjacent first and second lanes.
  • the first road segment and the second road segment are the same road segment, or the first road segment and the second road segment are two adjacent road segments.
  • the autonomous vehicle turns from the middle lane to the left lane
  • the route segment from lane block 8 to lane block 7 can be called the steering route segment
  • lane block 7 and lane block 8 are located in different lanes.
  • the steering route segment includes the route segment located in the lane block 8 and the route segment located in the lane block 7, and the lane block 8 and the lane location 7 are located in the same road segment. Accordingly, autonomous vehicles can steer from one lane to another within the same road segment. or,
  • the route segment from the lane block 8 to the lane block 10 may be referred to as a turning route segment, the lane block 8 and the lane block 10 are located in different lanes, and the steering route segment includes the route segment located in the lane block 8 and the route segment located in the lane block 8.
  • a route segment within lane block 10, and lane block 8 and lane location 10 are located in different road segments.
  • an autonomous vehicle can steer from one lane to another and need to travel across road segments.
  • the above-mentioned route includes a straight route segment, the straight route segment includes an adjacent first route segment and a second route segment, and the target path includes an adjacent route segment.
  • the autonomous vehicle travels straight from lane block 2 to lane block 5, and the route segment from lane block 2 to lane block 5 can be called a straight route segment, lane block 2 and lane block 5
  • the straight route segment is located in the route segment in the lane block 2 and the route segment in the lane block 8, and the lane block 2 and the lane location 5 are located in different road segments.
  • FIG. 5 is a flowchart of a method for controlling a movable platform provided by another embodiment of the present application. As shown in FIG. 5 , the method in this embodiment may include:
  • each lane block in the lane distribution information is searched to obtain multiple route segments, and multiple routes are obtained according to the multiple route segments, for example, a route is formed by combining multiple route segments.
  • a possible implementation manner of the above S503 and S504 is: searching for each lane block in the lane distribution information respectively according to a variety of preset motion strategies to obtain a plurality of route segments. Then, according to the plurality of route segments searched for each preset movement strategy, the route corresponding to the preset movement strategy is generated.
  • n kinds of preset motion strategies are shown in Fig. 6.
  • the lane distribution information of the target path is obtained based on the road network information, and then each preset motion strategy and lane distribution information are obtained according to each preset motion strategy and lane distribution information.
  • For the routes corresponding to the movement strategy a total of n routes are obtained, and then the target route is determined from the n routes.
  • the multiple preset movement strategies include a least-change lane movement strategy and a lane movement strategy in a target direction, where the target direction is the leftmost direction or the leftmost direction.
  • the preset motion strategy includes the following items: a least-change lane motion strategy, a left-lane motion strategy, and a right-lane motion strategy.
  • the following example describes the specific implementation process.
  • the least-change lane movement strategy also known as the greedy forward strategy
  • search each lane block in the lane distribution information may obtain multiple routes
  • the specific process of the segment may include: starting from the current lane block of the target path in the lane distribution information and searching straight along the target path until the third lane block, wherein the next lane block of the third lane block does not belong to the target path; Determine the straight route segment according to the lane block searched straight; start the steering search from the third lane block until the lane block of the target path is searched, and determine the steering route segment according to the lane block searched by the steering.
  • the target route includes driving from road A to road B.
  • the lane distribution information it can be determined that the lane block 13 and the lane block 14 are located in the two bifurcated lanes, and the road from A to B must be Pass through lane block 13.
  • the current lane block is lane block 2, starting from lane block 2 and searching straight along road A, that is, moving forward greedily.
  • the target route ie, B road
  • traffic rules that is, it is impossible to return from the lane block 14 to the must-pass lane block 13 .
  • the search ends when the bifurcation road is reached, and the searched lane block is the lane block 11, and the lane block 2, the lane block 5, the lane block 8, the lane block 8 and the lane block are searched according to the straight-forward search.
  • Block 11 determines the straight route segment.
  • the straight route segment is a route segment that goes straight from the lane block 2 to the lane block 5, the lane block 8, and the lane block 11 in sequence.
  • the turning route segment is a route segment that is left-turned from the lane block 11 to the lane block 10 .
  • a route from lane block 2 to lane block 10 can be obtained by combining the above-mentioned straight route segment and turning route segment.
  • the specific process of the route segment may include: starting from the current lane block of the target path in the lane distribution information to search for a right turn until the third lane block, wherein the third lane block is a lane block that cannot continue to turn right. ; Determine the steering route segment according to the searched lane block; search from the third lane block to the required lane block in the target path, and determine the return route from the third lane block according to the searched lane block To the route segment that must pass through the lane block.
  • the lane in which the above-mentioned third lane block is located is the drivable rightmost lane of the road on which the autonomous driving vehicle is located when the lane is driving forward. In some cases, no lane exists to the right of the third lane block.
  • the right lane of the third lane block is not a driveable lane for autonomous vehicles; for example, the right lane of the third lane block is an emergency lane, a bus lane, a bicycle lane or a sidewalk etc., or, the drivable direction of the right lane of the third lane block is different from the drivable direction of the third lane block, for example, the drivable direction of the lane where the third lane block is located is straight, while the The drivable direction of the right lane of the block is right-turn driving.
  • the target route includes driving from road A to road B.
  • the lane distribution information it can be determined that the lane block 13 and the lane block 14 are located in two bifurcated lanes, and the road from A to B must be Pass through lane block 13.
  • the current lane block is lane block 2, and the search starts from lane block 2 to the right, until it can no longer turn to the right, that is, when the lane block 3 is searched, it cannot continue to steer to the right to change the lane.
  • Lane Block 2 and Lane Block 3 define the right turn course segment.
  • each lane block in the lane distribution information is searched, and multiple lanes are obtained.
  • the specific process of the route segment may include: starting from the current lane block of the target path in the lane distribution information to search for a left turn until the third lane block, wherein the third lane block is a lane block that cannot continue to turn left. ; Determine the steering route segment according to the searched lane block; search from the third lane block to the required lane block in the target path, and determine the return route from the third lane block according to the searched lane block To the route segment that must pass through the lane block.
  • the lane in which the above-mentioned third lane block is located is the drivable rightmost lane of the road where the autonomous vehicle is located when driving forward in this lane.
  • the left lane of the third lane block is not a driveable lane of the autonomous vehicle;
  • the driving directions are different.
  • the driving direction of the lane where the third lane block is located is straight, while the driving direction of the left lane of the right block of the third lane is left-turn driving.
  • the target route includes driving from road A to road B.
  • the lane distribution information it can be determined that the lane block 13 and the lane block 14 are located on the two bifurcated lanes, and the path from road A to road B must be Pass through lane block 13.
  • the current lane block is lane block 2, start from lane block 2 to turn left and search until it can no longer turn left, that is, after searching for lane block 1, it cannot continue to turn left to change lanes, and then according to Lane Block 2 and Lane Block 1 define the left turn course segment.
  • these 3 routes all include routes from lane block 2 to lane block 16. These 3 routes are parallel routes, not 3 routes that must be parallel to each other. The 3 routes intersect each other in some sections and parallel in some sections.
  • FIG. 8 is a flowchart of a method for controlling a movable platform provided by another embodiment of the present application. As shown in FIG. 8 , the method in this embodiment may include:
  • the steerable area of each route can be obtained according to the lane distribution information of each route and the target route.
  • the movable platform can steer to change lanes.
  • the routes taking one of the routes as an example, first determine the third lane and the fourth lane where the steering route segment in the route is located; according to the lane distribution information of the target route, then obtain the fourth lane block and the necessary lanes in the route.
  • the fifth lane block, the fourth lane block is the lane block in the third lane, and the fifth lane block is the lane block in the fourth lane; then according to the road section of the fourth lane block and the fifth lane block
  • the lane block between the road segments where the block is located determines the steerable area of the route.
  • the route includes the following lane block sequence ⁇ 2, 5, 8, 7, 10 ⁇ , and the route includes a turning route segment, which is turned from lane block 8 to lane block 7
  • the route segment is located in the lane where the lane block 7 is located (the leftmost lane) and the lane where the lane block 8 is located (the middle lane), and the route must pass through the lane block 10 of the leftmost lane, and must pass through the lane block 10 of the leftmost lane.
  • Lane block 2 in the middle lane.
  • the steerable area of the route is determined according to the lane block between the road segment where the lane block 2 is located and the road segment where the lane block 10 is located.
  • the route segments before and after the steering route segment in the above route are ⁇ 2, 5, 8 ⁇ and ⁇ 7, 10 ⁇ respectively.
  • the last lane block in the route segment before turning is the lane block 8. Starting from the lane block 8, go straight ahead and search for the lane block. If the searched lane block can be changed by turning once, the post-steering can be reached. In a certain lane block in the route segment ⁇ 7, 10 ⁇ , it is determined that the searched lane block belongs to the steerable area of the route. For example, when the lane block 11 is searched, the lane block 10 can be reached by changing the lane once the vehicle direction, and the above process is repeated until no lane block that meets the conditions can be searched.
  • the first lane block in the route segment after turning is the lane block 7.
  • the steerable area of route ⁇ 2, 5, 8, 7, 10 ⁇ includes lane blocks 1, 2, 4, 5, 7, 8, 10, 11.
  • control the movable platform to move along the target path according to the target route and the steerable area of the target route.
  • the automatic driving vehicle is controlled to travel along the target route according to the determined target route and the steerable blocks of the target route.
  • the autonomous vehicle is controlled to travel along the target route according to the determined target route, and the environmental information of the autonomous vehicle is detected during the process of the autonomous vehicle traveling along the target route.
  • control the self-driving vehicle according to the environmental information of the self-driving vehicle and the steerable blocks on the target route, control the self-driving vehicle to steer and drive along the target route in the steerable area of the target route, so as to ensure that during the process of controlling the self-driving vehicle according to the target route, According to the current environmental information, timely turn to lane change adjustment to avoid uncontrollable driving of the autonomous vehicle according to the target route.
  • a possible implementation of controlling the self-driving vehicle to steer and drive along the target path in the steerable area of the target route is: The information detects that the self-driving vehicle needs to pass the location of the obstacle, and controls the self-driving vehicle to steer and move in the steerable area according to the steerable area on the target route to avoid the obstacle.
  • an obstacle is detected according to the environmental information, and the self-driving vehicle is controlled to drive according to the target route, and whether the self-driving vehicle needs to pass through the lane block where the obstacle is located, If the self-driving vehicle needs to pass through the lane block where the obstacle is located, according to the steerable area on the target route, the self-driving vehicle is controlled to turn to change lanes in the steerable area, so as to avoid driving through the lane block where the obstacle is located, so as to avoid driving through the lane block where the obstacle is located. Avoid obstacles and avoid the risk of a collision.
  • the obstacles may be, for example, roadblocks (stones, objects, etc.), stopped vehicles (such as vehicles that have collided), and the like.
  • the lane distribution information is returned, the steerable area of each route is obtained respectively, and the movable platform is controlled to move along the target path according to the target route and the steerable area of the target route. Help the movable platform to turn to the movement in time in some unexpected situations, and ensure the safety during the movement.
  • the following uses several embodiments to focus on how to determine a target flight route from a plurality of the flight routes.
  • FIG. 10 is a flowchart of a method for controlling a movable platform provided by another embodiment of the present application. As shown in FIG. 10 , the method in this embodiment may include:
  • the target route can be determined from the multiple routes according to the target reference index selected by the user.
  • the target reference index may be selected by the user according to his/her preference. If the user prefers efficiency, the target reference indicator may be, for example, the exercise duration corresponding to the route; if the user prefers comfort, the target reference indicator may be, for example, the steering corresponding to the route. number of times; if the user prefers safety, the target reference index may be, for example, the number of overtaking lanes and/or emergency lanes corresponding to the route. Therefore, according to the user's preference, this embodiment can adapt the user's preferred target route from the multiple routes to meet the user's preference requirement.
  • determining the target route from the multiple routes may include the following S1004 and S1005.
  • the parameter value of each flight route on the target reference index is determined.
  • the target reference index is the exercise duration corresponding to the route, and the exercise duration value corresponding to the route is determined according to the distance of the route.
  • the target reference index is the number of turns corresponding to the route, and then the value of the number of turns corresponding to the route is determined according to each steering segment in the route.
  • the number of dangerous lanes corresponding to the route is determined according to the lanes passing through the route to determine the number of dangerous lanes corresponding to the route.
  • Dangerous lanes include, for example, at least one of the following: passing lanes, emergency lanes, and collision-prone lanes (such as lanes close to bus lanes or bicycle lanes or freight lanes or sidewalks)
  • the target reference index is the exercise duration corresponding to the route
  • the route corresponding to the route whose exercise duration value is less than the exercise duration threshold is selected as the target route. From these at least two, select the route with the least movement duration value as the target route.
  • the target reference index is the number of turns corresponding to the route
  • the number of dangerous lanes corresponding to the route select the route corresponding to the route with the number of dangerous lanes less than the threshold of the number of dangerous lanes as the target route. For example, if there are at least two routes with the number of dangerous lanes less than the threshold of the number of dangerous lanes, then select Select the route with the least number of dangerous lanes among the at least two as the target route.
  • the route with the optimal parameter value may be selected as the target route, for example, the route with the least movement duration value is the target route, or the route with the least number of dangerous lanes is the target route, or, The route with the least number of turns is the target route.
  • determining the target route from the plurality of routes according to the target reference index selected by the user may include: determining the score of each route on the target reference index according to the target reference index; The score of each route on the target reference index, and the target route is determined from multiple routes. If the target reference index is the exercise duration, the lower the exercise duration value, the higher the score. If the target reference index is the number of turns, the lower the number of turns, the higher the score. If the target reference index is the number of dangerous lanes, the lower the number of dangerous lanes, the higher the score.
  • this embodiment may determine the route with the best score on the target reference index as the target route, or select a route from the scores greater than the score threshold as the target route.
  • the above threshold may be determined in one or more of the following ways:
  • the first way according to the target reference index selected by the user, determine the parameter threshold or score threshold corresponding to the target reference index.
  • the target reference indicators are the duration of movement, the number of turns exceeding the number of turns, and the number of dangerous lanes.
  • the corresponding thresholds may be different.
  • the second way get the parameter threshold or scoring threshold input by the user.
  • the above-mentioned parameter thresholds or scoring thresholds are set by the user, so that the selected target flight route is more suitable for the user's needs.
  • the third way according to the situational requirements of the movable platform moving along the target path, determine the parameter threshold or the scoring threshold.
  • the situational requirement is, for example, that the user rides the automatic driving vehicle to catch a flight, or the elderly and/or children ride the automatic driving vehicle, etc.
  • the fourth method using the machine learning method, the motion result of the movable platform is used as input to determine the parameter threshold or scoring threshold;
  • the fifth way according to the environmental information detected by the movable platform, determine the parameter threshold or the scoring threshold.
  • the vehicle environment information indicates that the lanes on the route are relatively congested, and the motion duration threshold needs to be set longer to adapt to the current environment.
  • the target route determined in this embodiment can satisfy the user's preference and improve the user experience.
  • Fig. 11 is a flowchart of a control method of a movable platform provided by another embodiment of the application. As shown in Fig. 11 , the method of this embodiment may include:
  • the route can be evaluated through a variety of reference indicators.
  • the reference indicators are, for example, the movement duration, the number of turns, and the number of dangerous lanes corresponding to the route.
  • the dangerous lanes include, for example, at least one of the following: overtaking lanes, emergency lanes , collision-prone lanes (such as lanes near bus lanes or bike lanes or freight lanes or sidewalks).
  • the weights of various reference indicators are generated. For example, if the target reference index selected by the user is the exercise duration, the weight of the exercise duration is greater than the weight of the number of turns and the number of dangerous lanes.
  • the second way Obtain the weights of various reference indicators input by the user.
  • the weights of the above reference indicators are set by the user, so that the selected target flight route is more suitable for the user's needs.
  • the third way determine the weight of the reference index according to the situational requirement of the movable platform moving along the target path.
  • the situational requirement is that the user takes an autonomous vehicle to catch a flight, for example, the weight of the exercise duration should be greater.
  • Situational requirements such as the elderly and/or children riding in autonomous vehicles, for example, the weight of the number of turns and the number of dangerous lanes should be greater.
  • the fourth method using the machine learning method, the motion results of the movable platform are used as input to determine the weights of various reference indicators;
  • the fifth way determine the weights of various reference indicators according to the environmental information detected by the mobile platform.
  • the target route is determined from the plurality of routes according to the weights of the various reference indicators, as shown in FIG. 6 .
  • a possible implementation of the above S1105 is: according to multiple reference indicators and multiple routes, obtain the score of each route on each reference index; for example, obtain the score of each route on the exercise duration, Score on the number of turns, score on the dangerous lane. Then, according to the scores of each route on various reference indicators and the weights of various reference indicators, the total score corresponding to each route is obtained; for example, the product of the score of the route on the exercise duration and the weight of the exercise duration is obtained to obtain the route. The product of the score on the number of turns and the weight of the number of turns is obtained to obtain the product of the score on the dangerous lane and the weight of the dangerous lane, and the sum of the above three products is determined as the total score corresponding to the route. Then, according to the total scores corresponding to the multiple routes, the target route is determined from the multiple routes, for example, the route with the highest total score is determined as the target route.
  • the target route is determined from the multiple routes according to the weights of various reference indicators. Therefore, the present application evaluates the routes from various aspects, so as to determine that the routes are better and better in all aspects. Target routes that can better meet user needs and improve user experience.
  • an implementation manner of determining the target route from the multiple routes is: detecting the environment where the movable platform is located information; according to the environmental information, from multiple routes, determine the target route.
  • the environmental information where the movable platform is located can be acquired in real time, and the target route can be determined from multiple routes according to the current environmental information. For example, which lane block has obstacles, the route passing through the lane block can be excluded. for the target route.
  • the environmental information where the movable platform is located is detected, and then the target route is re-determined from multiple routes according to the environmental information.
  • the target route determined earlier may not be the same route.
  • the movable platform is controlled to continue to move along the target path. Therefore, when there are some situations in the environment, so that there are some problems in controlling the movable platform according to the current target route (for example, if it continues to move, there will be a collision), then the present application re-determines a target route from multiple pre-generated routes, which can be adjusted quickly.
  • the movement trajectory of the movable platform does not need to regenerate the route or re-acquire the target path, which saves processing resources and improves processing efficiency.
  • the target route will be determined again from multiple routes, and the re-determined target route will not pass through the location of the obstacle to avoid the obstacle. Ensure the movement safety of the movable platform.
  • Embodiments of the present application further provide a computer storage medium, where instructions are stored in the computer storage medium, and when the instructions are executed on the computer, the computer can execute some or all of the steps of the method described in any of the foregoing embodiments.
  • Computer storage media include: read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disks or optical discs and other media that can store program codes.
  • Embodiments of the present application further provide a computer program product, including computer instructions, which, when executed by a processor, implement some or all of the steps of the method described in any of the foregoing embodiments.
  • FIG. 12 is a schematic structural diagram of a control apparatus for a movable platform according to an embodiment of the present application.
  • the control apparatus 1200 for a movable platform in this embodiment may include: a memory 1201 and a processor 1202 .
  • the memory 1201 and the processor 1202 may be connected by a communication bus.
  • the memory 1201 is used to store instructions.
  • the processor 1202 calls the instructions stored in the memory 1201 to perform the following operations:
  • the target path of the movable platform movement obtain the lane distribution information of the target path; generate multiple routes according to the lane distribution information, the multiple routes include a first route and a second route, the first route includes the first route segment, and the second route A second route segment is included, and the first route segment and the second route segment are located in different lanes of the target route; the target route is determined from the multiple routes; the movable platform is controlled to move along the target route according to the target route.
  • the lane distribution information includes the location of the target route in one or more lane blocks of each road segment.
  • the first air route and the second air route include at least part of the air route segment, and some air route segments are located on the same road segment.
  • each route traverses multiple lane blocks along the target path.
  • a route consists of multiple route segments connected, and multiple route segments are located in the lane blocks of consecutively distributed road segments.
  • the route includes a turning route segment
  • the steering route segment includes an adjacent first route segment and a second route segment
  • the target route includes the first route segment and the second route segment
  • the first route segment is located on the first route segment of the first route segment.
  • One lane block, the second route segment is located in the second lane block of the second road segment, the first lane block and the second lane block are located in the adjacent first lane and the second lane of the route, respectively.
  • the first road segment and the second road segment are the same road segment, or the first road segment and the second road segment are two adjacent road segments.
  • the route includes a straight route segment, the straight route segment includes an adjacent first route segment and a second route segment, the target route includes an adjacent first route segment and a second route segment, and the first route segment is located at the first route segment.
  • the first lane block of the road segment, the second route segment is located in the second lane block of the second road segment, and the first lane block and the second lane block are respectively located in the same lane of the target route.
  • the processor 1202 is specifically configured to: search for each lane block in the lane distribution information to obtain multiple route segments; and generate multiple routes according to the multiple route segments.
  • the processor 1202 is specifically configured to:
  • each lane block in the lane distribution information is searched respectively to obtain multiple flight segments.
  • a variety of preset motion strategies include a least-change lane motion strategy and a target direction lane motion strategy, and the target direction is the leftmost direction or the leftmost direction.
  • the preset movement strategy includes a least-change lane movement strategy.
  • the processor 1202 is specifically used for:
  • the steering search starts from the third lane block until the lane block of the target path is searched, and the steering route segment is determined according to the lane block searched by the steering.
  • the preset movement strategy includes a target-direction lane movement strategy.
  • the processor 1202 is specifically used for:
  • the search starts to turn toward the target direction until the third lane block, wherein the third lane block is a lane block that cannot continue to turn toward the target direction;
  • the processor 1202 is further configured to: obtain the steerable areas of the multiple routes respectively according to the multiple routes and the lane distribution information.
  • the processor 1202 is specifically configured to: control the movable platform to move along the target route according to the target route and the steerable area of the target route.
  • the processor 1202 is specifically configured to:
  • control the movable platform to move along the target path
  • the movable platform is controlled to steer and move within the steerable area along the target path.
  • the processor 1202 is specifically configured to:
  • the movable platform If it is detected according to the environmental information that the movable platform needs to pass through the position of the obstacle, according to the steerable area on the target route, the movable platform is controlled to steer in the steerable area to avoid the obstacle.
  • the processor 1302 is specifically configured to:
  • the fourth lane block is the lane block in the third lane
  • the fifth lane block is the fourth lane Lane blocks in ;
  • the steerable area of the route is determined.
  • the processor 1202 is specifically configured to:
  • the target route is determined from multiple routes.
  • the processor 1202 is specifically configured to:
  • the target reference index determine the parameter value of each route on the target reference index
  • the target route is determined from multiple routes.
  • the processor 1202 is further configured to perform one or more of the following:
  • the parameter threshold corresponding to the target reference index According to the target reference index selected by the user, determine the parameter threshold corresponding to the target reference index
  • the motion results of the movable platform are used as input to determine the parameter thresholds
  • the parameter threshold is determined according to the environmental information detected by the movable platform.
  • the processor 1202 is further configured to: obtain weights of multiple reference indicators;
  • the processor 1202 determines the target route from the multiple routes, it is specifically used for:
  • the target route is determined from multiple routes.
  • the processor 1202 is further configured to perform one or more of the following:
  • the weight of various reference indicators is generated
  • the motion results of the movable platform are used as input to determine the weights of various reference indicators
  • the weights of various reference indicators are determined.
  • the processor 1202 is specifically configured to:
  • the total score corresponding to each route is obtained;
  • the target route is determined from the multiple routes.
  • the processor 1202 is specifically configured to: detect environmental information where the movable platform is located; and determine a target route from multiple routes according to the environmental information.
  • the processor 1202 is further configured to:
  • the movable platform is controlled to continue to move along the target path.
  • the processor 1202 is specifically configured to:
  • the target route is determined again from multiple routes to avoid the obstacle.
  • control apparatus of the movable platform in this embodiment can be used to execute the technical solutions of the above method embodiments of the present application, and the implementation principles and technical effects thereof are similar, and will not be repeated here.
  • FIG. 13 is a schematic structural diagram of a movable platform according to an embodiment of the present application.
  • the body of the movable platform 1300 in this embodiment includes a memory 1301 and a processor 1302 .
  • the memory 1301 and the processor 1302 are connected by a communication bus.
  • the memory 1301 is used to store instructions.
  • the processor 1302 calls the instructions stored in the memory 1301 to perform the following operations:
  • the multiple routes include a first route and a second route, the first route includes a first route segment, the second route includes a second route segment, and the first route segment and the second route segment are located at the target different lanes of the path;
  • the movable platform 1300 is controlled to move along the target path.
  • the lane distribution information includes the location of the target route in one or more lane blocks of each road segment.
  • the first air route and the second air route include at least part of the air route segment, and some air route segments are located on the same road segment.
  • each route traverses multiple lane blocks along the target path.
  • a route consists of multiple route segments connected, and multiple route segments are located in the lane blocks of consecutively distributed road segments.
  • the route includes a turning route segment
  • the steering route segment includes an adjacent first route segment and a second route segment
  • the target route includes a first route segment and a second route segment
  • the first route segment is located on the first route segment of the first route segment.
  • One lane block, the second route segment is located in the second lane block of the second road segment, and the first lane block and the second lane block are located in the adjacent first lane and the second lane of the route, respectively.
  • the first road segment and the second road segment are the same road segment, or the first road segment and the second road segment are two adjacent road segments.
  • the route includes a straight route segment, the straight route segment includes an adjacent first route segment and a second route segment, the target route includes an adjacent first route segment and a second route segment, and the first route segment is located at the first route segment.
  • the first lane block of the road segment, the second route segment is located in the second lane block of the second road segment, and the first lane block and the second lane block are respectively located in the same lane of the target route.
  • the processor 1302 is specifically configured to: search each lane block in the lane distribution information to obtain a plurality of route segments; and generate a plurality of routes according to the plurality of route segments.
  • the processor 1302 is specifically configured to:
  • each lane block in the lane distribution information is searched respectively to obtain multiple route segments.
  • the multiple preset movement strategies include a least-change lane movement strategy and a lane movement strategy in a target direction, where the target direction is the leftmost direction or the leftmost direction.
  • the preset movement strategy includes a least-change lane movement strategy.
  • the processor 1302 is specifically used for:
  • the steering search starts from the third lane block until the lane block of the target path is searched, and the steering route segment is determined according to the lane block searched by the steering.
  • the preset movement strategy includes a target-direction lane movement strategy.
  • the processor 1302 is specifically used for:
  • the search starts to turn toward the target direction until the third lane block, wherein the third lane block is a lane block that can no longer turn toward the target direction;
  • the processor 1302 is further configured to:
  • the steerable areas of multiple routes are obtained respectively;
  • the processor 1302 controls the movable platform 1300 to move along the target path according to the target route, it is specifically used for:
  • the movable platform 1300 is controlled to move along the target path.
  • the processor 1302 is specifically configured to:
  • the movable platform 1300 is controlled to steer and move within the steerable area along the target path.
  • the processor 1302 is specifically configured to:
  • the movable platform 1300 If it is detected according to the environmental information that the movable platform 1300 needs to pass the position of the obstacle, the movable platform 1300 is controlled to steer in the steerable area according to the steerable area on the target route to avoid the obstacle.
  • the processor 1302 is specifically configured to:
  • the fourth lane block is the lane block in the third lane
  • the fifth lane block is the fourth lane Lane blocks in ;
  • the steerable area of the route is determined.
  • the processor 1302 is specifically configured to:
  • the target route is determined from multiple routes.
  • the processor 1302 is specifically configured to:
  • the target reference index determine the parameter value of each route on the target reference index
  • the target route is determined from multiple routes.
  • the processor 1302 is further configured to perform one or more of the following:
  • the parameter threshold corresponding to the target reference index According to the target reference index selected by the user, determine the parameter threshold corresponding to the target reference index
  • the motion result of the movable platform 1300 is used as input to determine the parameter threshold;
  • the parameter threshold is determined according to the environmental information detected by the movable platform 1300 .
  • the processor 1302 is further configured to:
  • the processor 1302 determines the target route from the multiple routes, it is specifically used for:
  • the target route is determined from multiple routes.
  • the processor 1302 is further configured to perform one or more of the following:
  • the weight of various reference indicators is generated
  • the motion results of the movable platform 1300 are used as input to determine the weights of various reference indicators;
  • the weights of various reference indicators are determined.
  • the processor 1302 is specifically configured to:
  • the total score corresponding to each route is obtained;
  • the target route is determined from the multiple routes.
  • the processor 1302 is specifically configured to:
  • the target route from multiple routes, determine the target route.
  • the processor 1302 is further configured to:
  • the movable platform 1300 is controlled to continue to move along the target path.
  • the processor 1302 is specifically configured to:
  • the target route is determined again from multiple routes to avoid the obstacle.
  • the movable platform 1300 of this embodiment may further include an environment sensor 1303, and the environment sensor 1303 is used to collect the environment information mentioned in the above embodiments.
  • the environmental sensor 1303 is, for example, an image sensor or a radar.
  • the movable platform 1300 may be an autonomous vehicle.
  • the movable platform of this embodiment can be used to execute the technical solutions of the above method embodiments of the present application, and the implementation principles and technical effects thereof are similar, and are not repeated here.
  • An embodiment of the present application further provides a movable platform, and the body of the movable platform of this embodiment includes a control device of the movable platform.
  • the control device of the movable platform may adopt the structure of the device embodiment shown in FIG. 12 .
  • the movable platform further includes an environment sensor, and the environment sensor can be connected in communication with the control device of the movable platform.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)
  • Navigation (AREA)

Abstract

一种可移动平台的控制方法和装置，所述方法包括：获取可移动平台运动的目标路径；获取目标路径的车道分布信息；根据车道分布信息生成多条航线，多条航线包括第一航线和第二航线，第一航线包括第一航线段，第二航线包括第二航线段，第一航线段和第二航线段位于目标路径的不同车道；从多条所述航线中，确定目标航线；根据目标航线，控制可移动平台沿目标路径运动。因此，为可移动平台沿目标路径行驶时提供更多的航线选择，以满足多样化需求。

Description

可移动平台的控制方法和装置 技术领域
本申请涉及可移动平台技术领域，更为具体地，涉及一种可移动平台的控制方法和装置。
背景技术
在自动驾驶领域，导航路线规划是不可或缺的一部分。目前自动驾驶车辆的导航线路规划是根据起始地、目的地和路网信息，规划航线。航线是由多个车道路段组成的从起始地到达目的地的一条路线。然后自动驾驶车辆根据该航线生成驾驶策略，通过驾驶策略控制自动驾驶车辆行驶，以便自动驾驶车辆沿着航线的路径行驶。但是，目前规划的航线单一，无法满足多样化需求。
发明内容
本申请的目的在于提供一种可移动平台的控制方法和装置，用于解决目前规划的航线单一而无法满足多样化需求的问题。
第一方面，本申请公开了一种可移动平台的控制方法，包括：
获取可移动平台运动的目标路径；
获取目标路径的车道分布信息；
根据车道分布信息生成多条航线，多条航线包括第一航线和第二航线，第一航线包括第一航线段，第二航线包括第二航线段，第一航线段和第二航线段位于目标路径的不同车道；
从多条航线中，确定目标航线；
根据目标航线，控制可移动平台沿目标路径运动。
第二方面，本申请公开了一种可移动平台的控制装置，包括：存储器和处理器；
存储器，用于存储指令；
处理器，调用存储器存储的指令用于执行以下操作：
获取可移动平台运动的目标路径；
获取目标路径的车道分布信息；
根据车道分布信息生成多条航线，多条航线包括第一航线和第二航线，第一航线包括第一航线段，第二航线包括第二航线段，第一航线段和第二航线段位于目标路径的不同车道；
从多条航线中，确定目标航线；
根据目标航线，控制可移动平台沿目标路径运动。
第三方面，本申请公开了一种可移动平台，包括如第二方面所述的可移动平台的控制装置。
第四方面，本申请实施例公开了一种计算机可读存储介质，所述计算机可读存储介质中存储有指令，当所述指令在计算机上运行时，使得计算机执行第一方面所述的方法。
第五方面，本申请实施例公开了一种计算机程序产品，包括计算机指令，所述计算机指令被处理器执行时实现第一方面所述的方法。
结合上述技术方案，本申请公开了可移动平台的控制方法和装置，通过根据目标路径的车道分布信息，生成同一目标路径的多个航线，然后从这多个航线中选择目标航线来控制自动驾驶车辆行驶。其中，不同航线间在同一道路上的车道不完全相同。因此，本申请为自动驾驶车辆提供更多航线可选择的空间，以满足用户偏好的需求、复杂环境下的需求。
附图说明
图1是根据本申请的实施例的自动驾驶车辆的示意性架构图；
图2为本申请一实施例提供的应用场景示意图；
图3为本申请一实施例提供的可移动平台的控制方法的流程图；
图4为本申请一实施例提供的车道分布信息的一种示意图；
图5为本申请另一实施例提供的可移动平台的控制方法的流程图；
图6为本申请提供的可移动平台的控制方法的一种原理示意图；
图7a-图7c分别为本申请一实施例提供的根据预设运动策略生成航线的示意图；
图8为本申请另一实施例提供的可移动平台的控制方法的流程图；
图9为本申请一实施例提供的确定航线的可转向区域的一种示意图；
图10为本申请另一实施例提供的可移动平台的控制方法的流程图；
图11为本申请另一实施例提供的可移动平台的控制方法的流程图；
图12为本申请一实施例提供的可移动平台的控制装置的结构示意图；
图13为本申请一实施例提供的可移动平台的结构示意图。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
需要说明的是，当组件被称为“固定于”另一个组件，它可以直接在另一个组件上或者也可以存在居中的组件。当一个组件被认为是“连接”另一个组件，它可以是直接连接 到另一个组件或者可能同时存在居中组件。
除非另有定义，本文所使用的所有的技术和科学术语与属于本申请的技术领域的技术人员通常理解的含义相同。本文中在本申请的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本申请。本文所使用的术语“及/或”包括一个或多个相关的所列项目的任意的和所有的组合。
本申请的实施例提供了可移动平台的控制方法和装置，其中，可移动平台可以是无人机、无人车、无人船、机器人或自动驾驶汽车等。
下面对本申请可移动平台的描述使用自动驾驶车辆作为示例。图1是根据本申请的实施例的自动驾驶车辆的示意性架构图。
自动驾驶车辆100可以包括感知系统110、控制系统120和机械系统130。
其中，感知系统110用于测量自动驾驶车辆100的状态信息，即自动驾驶车辆100的感知数据，感知数据可以表示自动驾驶车辆100的位置信息和/或状态信息，例如，位置、角度、速度、加速度和角速度等。感知系统110例如可以包括视觉传感器(例如包括多个单目或双目视觉装置)、激光雷达、毫米波雷达、惯性测量单元(Inertial Measurement Unit，IMU)、全球导航卫星系统、陀螺仪、超声传感器、电子罗盘、和气压计等传感器中的至少一种。例如，全球导航卫星系统可以是全球定位系统(Global Positioning System，GPS)。
感知系统110获取到感知数据后，可以将感知数据传输给控制系统120。其中，控制系统120用于根据感知数据做出用于控制自动驾驶车辆100如何行驶的决策，例如：以多少的速度行驶，或者，以多少的刹车加速度刹车，或者，是否变道行驶，或者，左/右转行驶等。控制系统120例如可以包括：计算平台，例如车载超级计算平台，或者中央处理器、分布式处理单元等具有处理功能器件的至少一种。控制系统120还可以包括车辆上各种数据传输的通信链路。
控制系统120可以根据确定的决策向机械系统130输出一个或多个控制指令。其中，机械系统130用于响应来自控制系统120的一个或多个控制指令对自动驾驶车辆100进行控制，以完成上述决策，例如：机械系统130可以驱动自动驾驶车辆100的车轮转动，从而为自动驾驶车辆100的行驶提供动力，其中，车轮的转动速度可以影响到自动驾驶车辆的速度。其中，机械系统130例如可以包括：机械的车身发动机/电动机、控制的线控系统等等中的至少一种。
应理解，上述对于自动驾驶车辆各组成部分的命名仅是出于标识的目的，并不应理解为对本申请的实施例的限制。
其中，图2为本申请一实施例提供的应用场景示意图，如图2所示，用户设定起始地和目的地，自动驾驶车辆100可以根据起始地和目的地，获取从起始地到目的地的目标路径，比如先经过道路A，再经过道路B，然后再经过道路C等。但是，其中，自动驾驶车辆100在道路上行驶时，由于道路的同一行驶方向上可以包括一条或多条车道，所以自动驾驶车辆还会规划出在每个道路上应该沿哪一车道行驶的一条航线。但是，目前自动驾驶 车辆规划出的航线为一条，无法满足多样化需求。
以自动驾驶车辆需要行驶经过环岛道路为环道为例，如果用户偏好效率，也就是，自动驾驶车辆的行驶时间越短，则自动驾驶车辆沿内环车道行驶最佳，如果用户偏好舒适性，也就是自动驾驶车辆转向车道(变换车道)行驶的次数最少，则自动驾驶车辆沿外环车道行驶最佳。但是目前规划的一条航线无法同时满足上述要求。
又比如，如果自动驾驶车辆需要行驶的车道上有障碍物，由于目前规划的航线为一条，自动驾驶车辆为了避障，自动驾驶车辆会停在障碍物后方而无法继续前进，无法满足自动驾驶车辆处于复杂环境下的需求。
而本申请中，通过生成多个航线，然后从这多个航线中选择目标航线来控制自动驾驶车辆行驶，其中，不同航线间在同一道路上的车道不完全相同，为自动驾驶车辆100提供更多航线可选择的空间，以满足用户偏好的需求、复杂环境下的需求，具体实现过程可以参见本申请下述各实施例所述。
图3为本申请一实施例提供的可移动平台的控制方法的流程图，如图3所示，本实施例的方法可以包括：
S301、获取可移动平台运动的目标路径。
S302、获取目标路径的车道分布信息。
S303、根据车道分布信息生成多条航线，多条航线包括第一航线和第二航线，第一航线包括第一航线段，第二航线包括第二航线段，第一航线段和第二航线段位于目标路径的不同车道。
S304、从多条航线中，确定目标航线。
S305、根据目标航线，控制可移动平台沿目标路径运动。
本实施例中，以可移动平台为自动驾驶车辆为例，获取自动驾驶车辆行驶的目标路径，具体的，可以根据自动驾驶车辆的起始地、目的地和路网信息，获得自动驾驶车辆从起始地到目的地的目标路径，目标路径包括从起始地到目的地途径的至少一条道路，比如从A道路的一位置行驶至A道路的另一位置。
在获取自动驾驶车辆行驶的目标路径后，再获取目标路径的车道分布信息，根据车道分布信息可以获得目标路径中各道路上的车道分布情况。其中，车道分布信息可以通过高精度的路网信息得到。然后根据目标路径的车道分布信息生成多条航线。以多条航线中的任两条航线为例，为了区分两条航线，该这两条航线分别称为第一航线和第二航线。第一航线包括至少一个航线段，以第一航线中的一个航线段为例，该航线段可称为第一航线段。第二航线包括至少一个航线段，以第二航线中的一个航线段为例，该航线段可称为第二航线段。第一航线段和第二航线段位于目标路径的不同车道上。然后从生成的多条航线中确定一条航线为目标航线，再根据该目标航线，控制自动驾驶车辆沿目标路径行驶。比如自动驾驶车辆可以根据表示用户偏好的指标从多条航线中确定目标航线，以满足不同用户的需求。比如自动驾驶车辆可以根据自动驾驶车辆所处的环境，从多条航线中确定目标航线， 以满足不同复杂环境下的行驶需求。
以目标路径包括A道路，A道路包括左车道和右车道为例，第一航线的第一航线段位于A道路的左车道，第一航线的第二航线段位于A道路的右车道。如果确定的目标航线为第一航线，则控制自动驾驶车辆在左车道上沿A道路行驶，如果确定的目标航线为第二航线，则控制自动驾驶车辆在右车道上沿A道路行驶。
可选的，该目标路径也可以是一条航线上的路径，根据该目标路径所在道路的车道分布信息，生成多条航线，从而增加了航线的种类，以为自动驾驶车辆提供更多的航线选择空间。
本实施例提供的方法，通过根据目标路径的车道分布信息生成多条航线，多条航线包括第一航线和第二航线，第一航线包括第一航线段，第二航线包括第二航线段，第一航线段和第二航线段位于目标路径的不同车道，然后根据多条航线确定目标航线，再根据目标航线，控制可移动平台沿目标路径运动。本申请中生成的多条航线属于同一目标路径(比如同一道路)，但是在车道级上是不完全相同的，为可移动平台沿目标路径行驶时提供更多的选择，以满足多样化需求。
在上述实施例的基础上，在一种可选的实施例中，上述的车道分布信息包括目标路径在每一路段的一个或多个的车道区块的位置。图4为本申请一实施例提供的车道分布信息的一种示意图，如图4所示，以目标路径所在的一条道路为例，该道路在横向方向分为三个车道，并且该道路在纵向方向包括多个路段，每个路段包括3个车道区块，其中，车道区块1至车道区块3组成一个路段，车道区块4至车道区块6组成一个路段，车道区块7至车道区块9组成一个路段，车道区块10至车道区块12组成一个路段。上述的S303的一种可能的实现方式为：根据目标路径在每一路段的一个或多个车道区块的位置，生成多条航线。
在上述各实施例的基础上，第一航线和第二航线中至少包括部分航线段，部分航线段位于同一路段上。其中，第一航线与第二航线中可以存在至少一个相同的航线段。相应的，上述S303中生成的多条航线可以是并列的多条航线，这多条航线均为同一起始地到同一目的地的不同航线。需要说明的是，上述并列并不是指这多条航线之间必须互相平行，而是指这多条航线均位于上述目标路径上，由同一起始地到达同一目的地。这多条航线可能在一些路段存在互相交叉，也可能在一些路段平行。
在上述各实施例的基础上，在一种可选的实施例中，每一航线沿目标路径经过多个车道区块。上述的S303的一种可能的实现方式为：根据车道分布信息，确定沿目标路径经过多个车道区块的多条航线。其中，图4中示出了其中一条航线，该条航线依次经过车道区块2、车道区块5、车道区块8、车道区块7、车道区块10等。其它一条航线也可以是依次经过车道区1、车道区块4、车道区块7、车道区块10等。
在上述各实施例的基础上，在一种可选的实施例中，航线由多个航线段连接组成，多个航线段位于连续分布的路段的车道区块。如图4所示的航线是由位于车道区块2 内的航线段、车道区块5内的航线段、车道区块8内的航线段、车道区块7内的航线段、车道区块10内的航线段。车道区块2所属的路段与车道区块5所属的路段是相邻的路段，车道区块8、车道区块7所属的路段与车道区块5所属的路段是相邻的，而且车道区块8、车道区块7所属的路段与车道区块10所属的路段是相邻的。
在上述各实施例的基础上，在一种可选的实施例中，上述的航线包括转向航线段，转向航线段包括相邻的第一航线段和第二航线段，目标路径包括第一路段和第二路段，第一航线段位于第一路段的第一车道区块，第二航线段位于第二路段的第二车道区块，第一车道区块和第二车道区块分别位于路径的相邻的第一车道和第二车道。第一路段和第二路段为同一路段，或者，第一路段与第二路段为相邻的两个路段。
参考图4所示，比如自动驾驶车辆由中间车道转向至左侧车道，车道区块8到车道区块7的航线段可以称为转向航线段，车道区块7与车道区块8位于不同的车道内，该转向航线段包括位于车道区块8内的航线段和位于车道区块7内的航线段，而且车道区块8和车道区位7位于同一路段内。相应地，自动驾驶车辆可以在同一路段内由一车道转向至另一车道。或者，
车道区块8到车道区块10的航线段可以称为转向航线段，车道区块8与车道区块10位于不同的车道内，该转向航线段包括位于车道区块8内的航线段和位于车道区块10内的航线段，而且车道区块8和车道区位10位于不同的路段内。相应地，自动驾驶车辆可以由一车道转向至另一车道内需要跨路段行驶。
在上述各实施例的基础上，在一种可选的实施例中，上述的航线包括直行航线段，直行航线段包括相邻的第一航线段和第二航线段，目标路径包括相邻的第一路段和第二路段，第一航线段位于第一路段的第一车道区块，第二航线段位于第二路段的第二车道区块，第一车道区块和第二车道区块分别位于目标路径的同一车道。
参考图4所示，自动驾驶车辆从车道区块2到车道区块5表示直行，车道区块2到车道区块5的航线段可以称为直行航线段，车道区块2与车道区块5位于同一车道内，该直行航线段位于车道区块2内的航线段和位于车道区块8内的航线段，而且车道区块2和车道区位5位于不同路段内。
在上述各实施例的基础上，图5为本申请另一实施例提供的可移动平台的控制方法的流程图，如图5所示，本实施例的方法可以包括：
S501、获取可移动平台运动的目标路径。
S502、获取目标路径的车道分布信息。
本实施例中，S501和S502的具体实现过程可以参见上述各实施例中的相关描述，此处不再赘述。
S503、搜索车道分布信息中的各车道区块，得到多条航线段。
S504、根据多条航线段，生成多条航线。
本实施例中，在获得车道分布信息后，搜索车道分布信息中的各车道区块，得到 多条航线段，根据多条航线段获得多条航线，比如将多条航线段组合而成航线。
其中，有关航线段、航线的描述可以参见上述各实施例中的相关描述，此处不再赘述。
比如可以搜索车道分布信息中的各车道区块，确定目标路径必经的车道区块，从车道分布信息中的各车道区块中去除不会经过的车道区块，获得可能经过的车道区块，然后根据必经的车道区块和可能经过的车道区块，得到多条航线段。
可选的，上述S503和S504的一种可能的实现方式为：根据多种预设运动策略，分别搜索车道分布信息中的各车道区块，得到多条航线段。然后根据每种预设运动策略搜索到的多条航线段，生成该预设运动策略对应的航线。
如图6所示，图6中示出了n种预设运动策略，目标路径的车道分布信息是基于路网信息获得，然后根据每种预设运动策略和车道分布信息，得到每种预设运动策略对应的航线，共得到n条航线，然后从n条航线中确定出目标航线。
可选的，多种预设运动策略包括最少变更车道运动策略、靠目标方向车道运动策略，目标方向为最左方向或最左方向。比如预设运动策略包括如下多项：最少变更车道运动策略、靠左车道运动策略、靠右车道运动策略。
比如可以搜索车道分布信息中的各车道区块，确定目标路径必经的车道区块，从车道分布信息中的各车道区块中去除不会经过的车道区块，获得可能经过的车道区块，然后根据必经的车道区块和可能经过的车道区块，得到多条航线段。下面举例对具体实现过程进行描述。
在一种可能的实现方式中，以最少变更车道运动策略(也可以称为贪婪向前策略)为例，根据最少变更车道运动策略，搜索车道分布信息中的各车道区块，得到多条航线段的具体过程可以包括：从车道分布信息中目标路径的当前车道区块开始沿目标路径直行搜索直至第三车道区块，其中，第三车道区块的下一车道区块不属于目标路径；根据直行搜索到的车道区块确定直行航线段；从第三车道区块开始转向搜索直至搜索到目标路径的车道区块，并根据转向搜索到的车道区块确定转向航线段。
如图7a所示，以目标路径包括由A道路行驶至B道路，根据车道分布信息可以确定车道区块13与车道区块14位于分叉的两处车道上，而且由A道路到B道路必经过车道区块13。当前车道区块为车道区块2，从车道区块2开始沿A道路直行搜索，也就是贪婪地往前运动，当运动到左边的车道和中间的车道分叉以后，将不能够在不违反交通规则的情况下回到目标路径(即B道路)，也就是从车道区块14无法回到必经车道区块13。故而在到达分叉路时搜索结束，此时搜索到的车道区块为车道区块为车道区块11，根据直行搜索到的车道区块2、车道区块5、车道区块8、车道区块11确定直行航线段。该直行航线段为从车道区块2依次直行到车道区块5、车道区块8、车道区块11的航线段。
然后从车道区块11开始转向搜索，由于必经车道区块13位于车道区块11的左侧， 则搜索由车道区块11左转向到车道区块13的航线段，比如从车道区块朝左搜索，搜索到车道区块10，车道区块10可以直行搜索到车道区块13，所以根据车道区块11、车道区块10确定转向航线段。该转向航线段为从车道区块11左转向到车道区块10的航线段。
比如将上述的直行航线段和转向航线段组合可以得到从车道区块2到转道区块10的航线。
在另一种可能的实现方式中，以靠右车道运动策略(也可以称为贪婪向右策略)为例，根据靠右车道运动策略，搜索车道分布信息中的各车道区块，得到多条航线段的具体过程可以包括：从车道分布信息中目标路径的当前车道区块开始朝右转向搜索直至第三车道区块，其中，第三车道区块为不能再继续朝右转向的车道区块；根据转向搜索到的车道区块确定转向航线段；从第三车道区块开始搜索回到目标路径中的必经车道区块，并根据搜索到的车道区块确定从第三车道区块回到必经车道区块的航线段。其中，上述的第三车道区块所在的车道为自动驾驶车辆位于该车道朝前行驶时所在道路的可行驶最右侧车道。在一些情况下，第三车道区块的右侧不存在车道。或者，在另一些情况下，该第三车道区块的右侧车道不是自动驾驶车辆可行驶的车道；比如该第三车道区块的右侧车道为应急车道、公交车道、自行车道或人行道等，或者，该第三车道区块的右侧车道的可行驶方向与第三车道区块的可行驶方向不同，比如第三车道区块所在车道的可行驶方向为直行，而第三车道右块的右侧车道的可行驶方向为右转行驶。
如图7b所示，以目标路径包括由A道路行驶至B道路，根据车道分布信息可以确定车道区块13与车道区块14位于分叉的两处车道上，而且由A道路到B道路必经过车道区块13。当前车道区块为车道区块2，从车道区块2开始朝右转向搜索，直至不能再继续朝右转向为止，即当搜索到车道区块3后无法再继续朝右转向变更车道，然后根据车道区块2和车道区块3确定右转向航线段。然后从车道区块3开始搜索回到必经的车道区块13，比如从车道区块4-12中搜索能从车道区块3到车道区块13的车道区块，比如通过车道区块6、9、12、11、10可以从车道区块3到车道区块13，因此，可以获得从车道区块3依次直行经过车道区块6、9、12的直行航线段，然后由车道区块12右转向变道至车道区块11，车道区块10的转向航线段。
在另一种可能的实现方式中，以靠左车道运动策略(也可以称为贪婪向左策略)为例，根据靠左车道运动策略，搜索车道分布信息中的各车道区块，得到多条航线段的具体过程可以包括：从车道分布信息中目标路径的当前车道区块开始朝左转向搜索直至第三车道区块，其中，第三车道区块为不能再继续朝左转向的车道区块；根据转向搜索到的车道区块确定转向航线段；从第三车道区块开始搜索回到目标路径中的必经车道区块，并根据搜索到的车道区块确定从第三车道区块回到必经车道区块的航线段。其中，上述的第三车道区块所在的车道为自动驾驶车辆位于该车道朝前行驶时所 在道路的可行驶最右侧车道。在一些情况下，第三车道区块的右侧不存在车道。或者，在另一些情况下，该第三车道区块的左侧车道不是自动驾驶车辆可行驶的车道；比如该第三车道区块的左侧车道的可行驶方向与第三车道区块的可行驶方向不同，比如第三车道区块所在车道的可行驶方向为直行，而第三车道右块的左侧车道的可行驶方向为左转行驶。
如图7c所示，以目标路径包括由A道路行驶至B道路，根据车道分布信息可以确定车道区块13与车道区块14位于分叉的两处车道上，而且由A道路到B道路必经过车道区块13。当前车道区块为车道区块2，从车道区块2开始朝左转向搜索，直至不能再继续朝左转向为止，即当搜索到车道区块1后无法再继续朝左转向变更车道，然后根据车道区块2和车道区块1确定左转向航线段。然后从车道区块1开始搜索回到必经的车道区块13，比如从车道区块4-12中搜索能从车道区块1到车道区块13的车道区块，比如通过车道区块4、5、10可以从车道区块1到车道区块13，因此，可以获得从车道区块1依次直行经过车道区块4、7、10到达车道区块的直行航线段。
其中，由图7a-图7c所示，这3条航线均包括由车道区块2到车道区块16的航线，这3条航线为并列的航线，并不是3条必须相互平行的航线，这3条航线在一些路段存在互相交叉，在一些路段平行。
S505、从多条航线中，确定目标航线。
S506、根据目标航线，控制可移动平台沿目标路径运动。
本实施例中，S505和S506的具体实现过程可以参见上述各实施例中的相关描述，此处不再赘述。
本实施例的方法，通过使用多种预设运动策略，在同一目标路径上可以生成多条不同的航线，以提供可移动平台更多的航线选择空间，以满足不同的运动需求。
图8为本申请另一实施例提供的可移动平台的控制方法的流程图，如图8所示，本实施例的方法可以包括：
S801、获取可移动平台运动的目标路径。
S802、获取目标路径的车道分布信息。
S803、根据车道分布信息生成多条航线。
本实施例中，S801-S803的具体实现过程可以参见上述各实施例中的相关描述，此处不再赘述。
S804、根据多条航线和车道分布信息，分别获得多条航线的可转向区域。
本实施例中，在生成多条航线后，根据每条航线和目标路径的车道分布信息，可以获得每条航线的可转向区域。在可转向区域内，可移动平台可以转向变更车道。
其中，以其中一条航线为例，先确定航线中转向航线段所在的第三车道和第四车道；根据目标路径的车道分布信息，再获取航线中必经的第四车道区块和必经的第五车道区块，第四车道区块为第三车道中的车道区块，第五车道区块为第四车道中的车道区 块；然后根据第四车道区块所在路段与第五车道区块所在路段之间的车道区块，确定航线的可转向区域。
参考图9所示，航线包括如下车道区块序列{2，5，8，7，10}，并且该航线中包括转向航线段，该转向航线段是由车道区块8转向到车道区块7的航线段，这一航线段位于车道区块7所在的车道(最左车道)和车道区块8所在的车道为(中间车道)，并且航线必经最左车道的车道区块10，必经中间车道的车道区块2。然后根据车道区块2所在路段与车道区块10所在路段之间的车道区块，确定航线的可转向区域。
在一种具体的例子中，上述航线中转向航线段前后的航线段分别为{2，5，8}和{7，10}。转向前的航线段中的最后一个车道区块为车道区块8，从车道区块8开始前向直行搜索车道区块，如果搜索到的车道区块能够通过转向更换一次车道即可到达转向后的航线段{7，10}中的某一车道区块，则确定搜索到的车道区块属于该航线的可转向区域。比如搜索到车道区块11可以通过车向更换一次车道即可到达车道区块10，重复上述过程直至搜索不到满足条件的车道区块。
转向后的航线段中的第一个车道区块为车道区块7，从车道区块7开始后向直行搜索车道区块，如果搜索到的车道区块能够通过转向更换一次车道即可到达转向前的航线段{2，5，8}中的某一车道区块，则确定搜索到的车道区块属于该航线的可转向区域。比如搜索到车道区块4可以通过车向更换一次车道即可到达车道区块5，搜索到车道区块1可以通过车向更换一次车道即可到达车道区块2。
因此，航线{2，5，8，7，10}的可转向区域包括车道区块1、2、4、5、7、8、10、11。
S805、从多条航线中，确定目标航线。
本实施例中，S805的具体实现过程可以参见各实施例中的相关描述，此处不再赘述。
S806、根据目标航线和目标航线的可转向区域，控制可移动平台沿目标路径运动。
本实施例中以自动驾驶车辆为例，在确定目标航线后，根据确定的目标航线和目标航线的可转向区块，控制自动驾驶车辆沿目标路径行驶。
在一种可选的实现方式中，在确定目标航线后，根据确定的目标航线，控制自动驾驶车辆沿目标路径行驶，在自动驾驶车辆沿目标路径行驶的过程中，检测自动驾驶车辆的环境信息，根据自动驾驶车辆的环境信息和目标航线上的可转向区块，控制自动驾驶车辆沿目标路径在目标航线的可转向区域内转向行驶，以确保根据目标航线控制自动驾驶车辆行驶的过程中，根据当前的环境信息及时转向变道调整，以避免根据目标航线无法控制自动驾驶车辆行驶。
可选的，根据自动驾驶车辆的环境信息和目标航线上的可转向区块，控制自动驾驶车辆沿目标路径在目标航线的可转向区域内转向行驶的一种可能的实现方式为：若根据环境信息检测到自动驾驶车辆行驶需经过障碍物所在的位置，则根据目标航线上的可转向区域，控制自动驾驶车辆在可转向区域转向运动，以规避障碍物。本实施例中，在检测到自动驾驶车辆的环境信息后，根据该环境信息检测到障碍物，并检测根据目 标航线控制自动驾驶车辆行驶，自动驾驶车辆是否需经过障碍物所在的车道区块，如果自动驾驶车辆需经过障碍物所在的车道区块，则根据目标航线上的可转向区域，控制自动驾驶车辆在可转向区域转向变更车道，以避免行驶经过该障碍物所在的车道区块，以规避障碍物，避免发生碰撞的风险。其中，障碍物比如可以是路障(石头、物品等)、停止的车辆(比如发生碰撞的车辆)等。
本实施例提供的方法，在生成多条航线之后，还车道分布信息，分别获得每条航线的可转向区域，根据目标航线和目标航线的可转向区域，控制可移动平台沿目标路径运动，以帮助可移动平台在一些突发情况下能及时转向运动，保障运动过程中的安全性。
下面采用几个实施例重点对如何从多条所述航线中，确定目标航线进行描述。
图10为本申请另一实施例提供的可移动平台的控制方法的流程图，如图10所示，本实施例的方法可以包括：
S1001、获取可移动平台运动的目标路径。
S1002、获取目标路径的车道分布信息。
S1003、根据车道分布信息生成多条航线。
本实施例中，S1001-S1003的具体实现过程可以参见各实施例中的相关描述，此处不再赘述。
在生成多条航线后，可以根据用户选定的目标参考指标，从这多条航线中确定目标航线。该目标参考指标可以是用户根据其偏好选定的，如果用户偏好效率性，该目标参考指标比如可以是航线对应的运动时长；如果用户偏好舒适性，该目标参考指标比如可以是航线对应的转向次数；如果用户偏好安全性，该目标参考指标比如可以是航线对应的超车道和/或应急车道个数。因此，本实施例可以根据用户的偏好，从这多条航线中适配用户偏好的目标航线，以满足用户的偏好需求。
在一种可选的实施例中，根据用户选定的目标参考指标，从这多条航线中确定目标航线可以包括如下S1004和S1005。
S1004、根据目标参考指标，确定每条航线在目标参考指标上的参数值。
本实施例中，根据上述的目标参考指标，确定每条航线在目标参考指标上的参数值。比如目标参考指标为航线对应的运动时长，则根据该航线的路程长度，确定该航线对应的运动时长值。比如目标参考指标为航线对应的转向次数，则根据该航线中的各转向航线段，确定该航线对应的转向次数值。比如航线对应的危险车道个数，则根据该航线中经过的各车道，确定该航线对应的危险车道个数值。危险车道比如包括如下至少一项：超车道、应急车道、容易发生碰撞的车道(比如靠近公交车道或自行车道或货车道或人行道的车道)
S1005、根据参数值与参数阈值之间的大小关系，从多条航线中确定目标航线。
本实施例中，如果目标参考指标为航线对应的运动时长，则选择航线对应的运动时长值小于运动时长阈值的航线为目标航线，比如运动时长值小于运动时长阈值的航线为至少两条，则从这至少两条中选择运动时长值最少的航线为目标航线。如果目标参考指标为航 线对应的转向次数，则选择航线对应的转向次数小于转向次数阈值的航线为目标航线，比如转向次数值小于转向次数阈值的航线为至少两条，则从这至少两条中选择转向次数值最少的航线为目标航线。如果航线对应的危险车道个数，则选择航线对应的危险车道个数小于危险车道个数阈值的航线为目标航线，比如危险车道个数小于危险车道个数阈值的航线为至少两条，则从这至少两条中选择危险车道个数最少的航线为目标航线。
在上述S1005的一种可替换的方式中，可以选择参数值最优的航线为目标航线，比如运动时长值最少的航线为目标航线，或者，危险车道个数值最少的航线为目标航线，或者，转向次数值最少的航线为目标航线。
在另一种可替换的实施例中，根据用户选定的目标参考指标，从这多条航线中确定目标航线可以包括：根据目标参考指标，确定每条航线在目标参考指标上的评分；根据各航线在目标参考指标上的评分，从多条航线中确定目标航线。如果目标参考指标为运动时长，则运动时长值越少，评分越高。如果目标参考指标为转向次数，则转向次数值越少，评分越高。如果目标参考指标为危险车道个数，则危险车道个数值越少，评分越高。可选的，本实施例可以将目标参考指标上的评分最佳的航线确定为目标航线，或者，从大于评分阈值的评分中选择一条航线为目标航线。
可选的，上述的阈值可以采用如下一种或多种方式确定的：
第一种方式：根据用户选定的目标参考指标，确定与目标参考指标相对应的参数阈值或评分阈值。比如：目标参考指标为运动时长、转向次数超、危险车道个数三种指标对应的阈值可能不相同。
第二种方式：获取用户输入的参数阈值或评分阈值。上述的参数阈值或评分阈值由用户设定，以选择的目标航线更适配用户的需求。
第三种方式：根据可移动平台沿目标路径运动的情景需求，确定参数阈值或评分阈值。其中以可移动平台为自动驾驶车辆为例，情景需求比如是用户乘坐自动驾驶车辆赶飞行，或者，老人和/或小孩乘坐自动驾驶车辆等。
第四种方式：采用机器学习的方式，将可移动平台的运动结果作为输入，确定参数阈值或评分阈值；
第五种方式：根据可移动平台检测到的环境信息，确定参数阈值或评分阈值。比车环境信息表示航线上途经的车道上比较拥堵，运动时长阈值则需要设置得更长，以适配当前环境。
S1006、根据目标航线，控制可移动平台沿目标路径运动。
本实施例中，S1006的具体实现过程可以参见各实施例中的相关描述，此处不再赘述。
本实施例的方法，在生成多条航线后，根据用户选定的目标参考指标，重点关注用户选定的目标参考指标，从多条航线中确定目标航线，其中，用户选定的指标与用户的当前偏好有关，因此本实施例确定的目标航线可满足用户的偏好，提高用户体验。
图11为本申请另一实施例提供的可移动平台的控制方法的流程图，如图11所示，本 实施例的方法可以包括：
S1101、获取可移动平台运动的目标路径。
S1102、获取目标路径的车道分布信息。
S1103、根据车道分布信息生成多条航线。
本实施例中，S1101-S1103的具体实现过程可以参见各实施例中的相关描述，此处不再赘述。
S1104、获取多种参考指标的权重。
本实施例中，可以通过多种参考指标来对航线进行评价，该参考指标比如是航线对应的运动时长、转向次数、危险车道个数，危险车道比如包括如下至少一项：超车道、应急车道、容易发生碰撞的车道(比如靠近公交车道或自行车道或货车道或人行道的车道)。
需要说明的是，S1104与上述S1101、S1102、S1103的执行顺序不做限定。
其中，上述S1104可以如下一种或多种实施方式：
第一种方式：根据用户选定的目标参考指标，生成多种参考指标的权重。比如：用户选定的目标参考指标为运动时长，则运动时长的权重相对于转向次数的权重和危险车道个数的权重更大。
第二种方式：获取用户输入的多种参考指标的权重。上述的各参考指标的权重由用户设定，以选择的目标航线更适配用户的需求。
第三种方式：根据可移动平台沿目标路径运动的情景需求，确定参考指标的权重。其中以可移动平台为自动驾驶车辆为例，情景需求比如是用户乘坐自动驾驶车辆赶飞行，则比如运动时长的权重应更大。情景需求比如是老人和/或小孩乘坐自动驾驶车辆等，则比如转向次数的权重和危险车道个数的权重应更大。
第四种方式：采用机器学习的方式，将可移动平台的运动结果作为输入，确定多种参考指标的权重；
第五种方式：根据可移动平台检测到的环境信息，确定多种参考指标的权重。
S1105、根据多种参考指标的权重，从多条航线中，确定目标航线。
本实施例中，获取多种参考指标的权重后，根据这多种参考指标的权重，从多条航线路确定目标航线，如图6所示。
可选的，上述S1105的一种可能的实现方式为：根据多种参考指标和多条航线，获取每条航线在每种参考指标上的评分；比如获取每条航线在运动时长上的评分、转向次数上的评分、危险车道上的评分。再根据每条航线分别在多种参考指标上的评分以及多种参考指标的权重，获得每条航线对应的总评分；比如获取航线在运动时长上的评分与运动时长的权重的乘积，获取航线在转向次数上的评分与转向次数的权重的乘积，获取航线在危险车道上的评分与危险车道的权重的乘积，将上述三个乘积之和确定为该航线对应的总评分。然后根据多条航线分别对应的总评分，从多条航线中确定目标航线，比如将总评分最高的航线确定为目标航线。
S1106、根据目标航线，控制可移动平台沿目标路径运动。
本实施例中，S1106的具体实现过程可以参见各实施例中的相关描述，此处不再赘述。
本实施例的方法，在生成多条航线后，根据多种参考指标的权重，从多条航线中确定目标航线，因此本申请是从多方面来评估航线，以从中确定各方面均较佳且更能满足用户需求的目标航线，提高用户体验。
在一种可替换上述图10或图11所示方案的实施例中，在生成多条航线后，从多条航线中，确定目标航线的一种实现方式为：检测可移动平台所处的环境信息；根据环境信息，从多条航线，确定目标航线。本实施例可以实时获取可移动平台所处的环境信息，可以根据当前的环境信息，从多条航线中确定目标航线，比如哪个车道区块存在障碍物，则可以排除经过该车道区块的航线为目标航线。
在上述各实施例的基础上，在根据目标航线控制可移动平台运动的过程中，检测可移动平台所处的环境信息，再根据环境信息，从多条航线中再次确定目标航线，该目标航线与前面确定目标航线可以不是同一航线。根据再次确定的目标航线，控制可移动平台继续沿目标路径运动。因此，当环境出现一些状况时，使得根据当前的目标航线控制可移动平台存在一些问题(比如继续运动会发生碰撞)，则本申请从预先生成的多条航线中重新确定一条目标航线，可以快速调整可移动平台的运动轨迹，无需再重新生成航线也无需再重新获取目标路径，节省了处理资源，提高了处理效率。
比如：如果根据环境信息确定可移动平台运动需经过障碍物所在的位置，则从多条航线，再次确定目标航线，再次确定的目标航线不途径该障碍物所在的位置，以规避该障碍物，保障可移动平台的运动安全性。
本申请实施例中还提供了一种计算机存储介质，该计算机存储介质中存储有指令，当指令在计算机上运行时，使得计算机执行上述任一实施例所述方法的部分或全部步骤。计算机存储介质包括：只读内存(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
本申请实施例中还提供了一种计算机程序产品，包括计算机指令，所述计算机指令被处理器执行时实现上述任一实施例所述方法的部分或全部步骤。
图12为本申请一实施例提供的可移动平台的控制装置的结构示意图，如图12所示，本实施例的可移动平台的控制装置1200可以包括：存储器1201和处理器1202。存储器1201和处理器1202可以通过通信总线连接。
存储器1201，用于存储指令。
处理器1202，调用存储器1201存储的指令用于执行以下操作：
获取可移动平台运动的目标路径；获取目标路径的车道分布信息；根据车道分布信息生成多条航线，多条航线包括第一航线和第二航线，第一航线包括第一航线段，第二航线包括第二航线段，第一航线段和第二航线段位于目标路径的不同车道；从多条航线中，确定目标航线；根据目标航线，控制可移动平台沿目标路径运动。
在一些实施例中，车道分布信息包括目标路径在每一路段的一个或多个的车道区块的位置。
在一些实施例中，第一航线和第二航线中至少包括部分航线段，部分航线段位于同一路段上。
在一些实施例中，每一航线沿目标路径经过多个车道区块。航线由多个航线段连接组成，多个航线段位于连续分布的路段的车道区块。
在一些实施例中，航线包括转向航线段，转向航线段包括相邻的第一航线段和第二航线段，目标路径包括第一路段和第二路段，第一航线段位于第一路段的第一车道区块，第二航线段位于第二路段的第二车道区块，第一车道区块和第二车道区块分别位于路径的相邻的第一车道和第二车道。第一路段和第二路段为同一路段，或者，第一路段与第二路段为相邻的两个路段。
在一些实施例中，航线包括直行航线段，直行航线段包括相邻的第一航线段和第二航线段，目标路径包括相邻的第一路段和第二路段，第一航线段位于第一路段的第一车道区块，第二航线段位于第二路段的第二车道区块，第一车道区块和第二车道区块分别位于目标路径的同一车道。
在一些实施例中，处理器1202，具体用于：搜索车道分布信息中的各车道区块，得到多条航线段；根据多条航线段，生成多条航线。
在一些实施例中，处理器1202，具体用于：
根据多种预设运动策略，分别搜索车道分布信息中的各车道区块，得到多条航线段。其中，多种预设运动策略包括最少变更车道运动策略、靠目标方向车道运动策略，目标方向为最左方向或最左方向。
在一些实施例中，预设运动策略包括最少变更车道运动策略。处理器1202，具体用于：
从车道分布信息中目标路径的当前车道区块开始沿目标路径直行搜索直至第三车道区块，其中，第三车道区块的下一车道区块不属于目标路径；
根据直行搜索到的车道区块确定直行航线段；
从第三车道区块开始转向搜索直至搜索到目标路径的车道区块，并根据转向搜索到的车道区块确定转向航线段。
在一些实施例中，预设运动策略包括靠目标方向车道运动策略。处理器1202，具体用于：
从车道分布信息中目标路径的当前车道区块开始朝目标方向转向搜索直至第三车道区块，其中，第三车道区块为不能再继续朝目标方向转向的车道区块；
根据转向搜索到的车道区块确定转向航线段；
从第三车道区块开始搜索回到目标路径中的必经车道区块，并根据搜索到的车道区块确定从第三车道区块回到必经车道区块的航线段。
在一些实施例中，处理器1202，还用于：根据多条航线和车道分布信息，分别获得多条航线的可转向区域。处理器1202在根据目标航线，控制可移动平台沿目标路径运动时，具体用于：根据目标航线和目标航线的可转向区域，控制可移动平台沿目标路径运动。
在一些实施例中，处理器1202，具体用于：
根据目标航线，控制可移动平台沿目标路径运动；
在可移动平台沿目标路径运动的过程中，检测可移动平台的环境信息；
根据环境信息和目标航线上的可转向区域，控制可移动平台沿目标路径在可转向区域内转向运动。
在一些实施例中，处理器1202，具体用于：
若根据环境信息检测到可移动平台运动需经过障碍物所在的位置，则根据目标航线上的可转向区域，控制可移动平台在可转向区域转向运动，以规避障碍物。
在一些实施例中，处理器1302，具体用于：
确定航线中转向航线段所在的第三车道和第四车道；
根据车道分布信息，获取航线中必经的第四车道区块和必经的第五车道区块，第四车道区块为第三车道中的车道区块，第五车道区块为第四车道中的车道区块；
根据第四车道区块所在路段与第五车道区块所在路段之间的车道区块，确定航线的可转向区域。
在一些实施例中，处理器1202，具体用于：
根据用户选定的目标参考指标，从多条航线中确定目标航线。
在一些实施例中，处理器1202，具体用于：
根据目标参考指标，确定每条航线在目标参考指标上的参数值；
根据参数值与参数阈值之间的大小关系，从多条航线中确定目标航线。
在一些实施例中，处理器1202，还用于执行如下一项或多项：
根据用户选定的目标参考指标，确定与目标参考指标相对应的参数阈值；
获取用户输入的参数阈值；
根据可移动平台沿目标路径运动的情景需求，确定参数阈值；
采用机器学习的方式，将可移动平台的运动结果作为输入，确定参数阈值；
根据可移动平台检测到的环境信息，确定参数阈值。
在一些实施例中，处理器1202，还用于：获取多种参考指标的权重；
处理器1202在从多条航线中，确定目标航线时，具体用于：
根据多种参考指标的权重，从多条航线中，确定目标航线。
在一些实施例中，处理器1202，还用于执行如下一项或多项：
根据用户选定的目标参考指标，生成多种参考指标的权重；
获取用户输入的多种参考指标的权重；
根据可移动平台沿目标路径运动的情景需求，确定多种参考指标的权重；
采用机器学习的方式，将可移动平台的运动结果作为输入，确定多种参考指标的权重；
根据可移动平台检测到的环境信息，确定多种参考指标的权重。
在一些实施例中，处理器1202，具体用于：
根据多种参考指标和多条航线，获取每条航线在每种参考指标上的评分；
根据每条航线分别在多种参考指标上的评分以及多种参考指标的权重，获得每条航线对应的总评分；
根据多条航线分别对应的总评分，从多条航线中确定目标航线。
在一些实施例中，处理器1202，具体用于：检测可移动平台所处的环境信息；根据环境信息，从多条航线，确定目标航线。
在一些实施例中，处理器1202，还用于：
在根据目标航线控制可移动平台运动的过程中，检测可移动平台所处的环境信息；
根据环境信息，从多条航线中，再次确定目标航线；
根据再次确定的目标航线，控制可移动平台继续沿目标路径运动。
在一些实施例中，处理器1202，具体用于：
若根据环境信息确定可移动平台运动需经过障碍物所在的位置，则从多条航线，再次确定目标航线，以规避障碍物。
本实施例的可移动平台的控制装置，可以用于执行本申请上述各方法实施例的技术方案，其实现原理和技术效果类似，此处不再赘述。
图13为本申请一实施例提供的可移动平台的结构示意图，如图13所示，本实施例的可移动平台1300的机身内包括存储器1301和处理器1302。存储器1301和处理器1302通过通信总线连接。
存储器1301，用于存储指令。
处理器1302，调用存储器1301存储的指令用于执行以下操作：
获取可移动平台1300运动的目标路径；
获取目标路径的车道分布信息；
根据车道分布信息生成多条航线，多条航线包括第一航线和第二航线，第一航线包括第一航线段，第二航线包括第二航线段，第一航线段和第二航线段位于目标路径的不同车道；
从多条航线中，确定目标航线；
根据目标航线，控制可移动平台1300沿目标路径运动。
在一些实施例中，车道分布信息包括目标路径在每一路段的一个或多个的车道区块的位置。
在一些实施例中，第一航线和第二航线中至少包括部分航线段，部分航线段位于 同一路段上。
在一些实施例中，每一航线沿目标路径经过多个车道区块。航线由多个航线段连接组成，多个航线段位于连续分布的路段的车道区块。
在一些实施例中，航线包括转向航线段，转向航线段包括相邻的第一航线段和第二航线段，目标路径包括第一路段和第二路段，第一航线段位于第一路段的第一车道区块，第二航线段位于第二路段的第二车道区块，第一车道区块和第二车道区块分别位于路径的相邻的第一车道和第二车道。第一路段和第二路段为同一路段，或者，第一路段与第二路段为相邻的两个路段。
在一些实施例中，航线包括直行航线段，直行航线段包括相邻的第一航线段和第二航线段，目标路径包括相邻的第一路段和第二路段，第一航线段位于第一路段的第一车道区块，第二航线段位于第二路段的第二车道区块，第一车道区块和第二车道区块分别位于目标路径的同一车道。
在一些实施例中，处理器1302，具体用于：搜索车道分布信息中的各车道区块，得到多条航线段；根据多条航线段，生成多条航线。
在一些实施例中，处理器1302，具体用于：
根据多种预设运动策略，分别搜索车道分布信息中的各车道区块，得到多条航线段。其中，多种预设运动策略包括最少变更车道运动策略、靠目标方向车道运动策略，目标方向为最左方向或最左方向。
在一些实施例中，预设运动策略包括最少变更车道运动策略。处理器1302，具体用于：
从车道分布信息中目标路径的当前车道区块开始沿目标路径直行搜索直至第三车道区块，其中，第三车道区块的下一车道区块不属于目标路径；
根据直行搜索到的车道区块确定直行航线段；
从第三车道区块开始转向搜索直至搜索到目标路径的车道区块，并根据转向搜索到的车道区块确定转向航线段。
在一些实施例中，预设运动策略包括靠目标方向车道运动策略。处理器1302，具体用于：
从车道分布信息中目标路径的当前车道区块开始朝目标方向转向搜索直至第三车道区块，其中，第三车道区块为不能再继续朝目标方向转向的车道区块；
根据转向搜索到的车道区块确定转向航线段；
从第三车道区块开始搜索回到目标路径中的必经车道区块，并根据搜索到的车道区块确定从第三车道区块回到必经车道区块的航线段。
在一些实施例中，处理器1302，还用于：
根据多条航线和车道分布信息，分别获得多条航线的可转向区域；
处理器1302在根据目标航线，控制可移动平台1300沿目标路径运动时，具体用 于：
根据目标航线和目标航线的可转向区域，控制可移动平台1300沿目标路径运动。
在一些实施例中，处理器1302，具体用于：
根据目标航线，控制可移动平台1300沿目标路径运动；
在可移动平台1300沿目标路径运动的过程中，检测可移动平台1300的环境信息；
根据环境信息和目标航线上的可转向区域，控制可移动平台1300沿目标路径在可转向区域内转向运动。
在一些实施例中，处理器1302，具体用于：
若根据环境信息检测到可移动平台1300运动需经过障碍物所在的位置，则根据目标航线上的可转向区域，控制可移动平台1300在可转向区域转向运动，以规避障碍物。
在一些实施例中，处理器1302，具体用于：
确定航线中转向航线段所在的第三车道和第四车道；
根据车道分布信息，获取航线中必经的第四车道区块和必经的第五车道区块，第四车道区块为第三车道中的车道区块，第五车道区块为第四车道中的车道区块；
根据第四车道区块所在路段与第五车道区块所在路段之间的车道区块，确定航线的可转向区域。
在一些实施例中，处理器1302，具体用于：
根据用户选定的目标参考指标，从多条航线中确定目标航线。
在一些实施例中，处理器1302，具体用于：
根据目标参考指标，确定每条航线在目标参考指标上的参数值；
根据参数值与参数阈值之间的大小关系，从多条航线中确定目标航线。
在一些实施例中，处理器1302，还用于执行如下一项或多项：
根据用户选定的目标参考指标，确定与目标参考指标相对应的参数阈值；
获取用户输入的参数阈值；
根据可移动平台1300沿目标路径运动的情景需求，确定参数阈值；
采用机器学习的方式，将可移动平台1300的运动结果作为输入，确定参数阈值；
根据可移动平台1300检测到的环境信息，确定参数阈值。
在一些实施例中，处理器1302，还用于：
获取多种参考指标的权重；
处理器1302在从多条航线中，确定目标航线时，具体用于：
根据多种参考指标的权重，从多条航线中，确定目标航线。
在一些实施例中，处理器1302，还用于执行如下一项或多项：
根据用户选定的目标参考指标，生成多种参考指标的权重；
获取用户输入的多种参考指标的权重；
根据可移动平台1300沿目标路径运动的情景需求，确定多种参考指标的权重；
采用机器学习的方式，将可移动平台1300的运动结果作为输入，确定多种参考指标的权重；
根据可移动平台1300检测到的环境信息，确定多种参考指标的权重。
在一些实施例中，处理器1302，具体用于：
根据多种参考指标和多条航线，获取每条航线在每种参考指标上的评分；
根据每条航线分别在多种参考指标上的评分以及多种参考指标的权重，获得每条航线对应的总评分；
根据多条航线分别对应的总评分，从多条航线中确定目标航线。
在一些实施例中，处理器1302，具体用于：
检测可移动平台1300所处的环境信息；
根据环境信息，从多条航线，确定目标航线。
在一些实施例中，处理器1302，还用于：
在根据目标航线控制可移动平台1300运动的过程中，检测可移动平台1300所处的环境信息；
根据环境信息，从多条航线中，再次确定目标航线；
根据再次确定的目标航线，控制可移动平台1300继续沿目标路径运动。
在一些实施例中，处理器1302，具体用于：
若根据环境信息确定可移动平台1300运动需经过障碍物所在的位置，则从多条航线，再次确定目标航线，以规避障碍物。
在一些实施例中，本实施例的可移动平台1300还可以包括环境传感器1303，环境传感器1303用于采集上述各实施例提及的环境信息。环境传感器1303比如是图像传感器或者雷达等。
在一种可选的实施例中，可移动平台1300可以是自动驾驶车辆。
本实施例的可移动平台，可以用于执行本申请上述各方法实施例的技术方案，其实现原理和技术效果类似，此处不再赘述。
本申请实施例还提供一种可移动平台，本实施例的可移动平台的机身内包括可移动平台的控制装置。其中，可移动平台的控制装置可以采用图12所示装置实施例的结构。可选的，可移动平台还包括环境传感器，该环境传感器可以与可移动平台的控制装置通信连接。
最后应说明的是：以上各实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述各实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。

Claims (50)

  1. 一种可移动平台的控制方法，其特征在于，包括：
    获取所述可移动平台运动的目标路径；
    获取所述目标路径的车道分布信息；
    根据所述车道分布信息生成多条航线，多条所述航线包括第一航线和第二航线，所述第一航线包括第一航线段，所述第二航线包括第二航线段，所述第一航线段和所述第二航线段位于所述目标路径的不同车道；
    从多条所述航线中，确定目标航线；
    根据所述目标航线，控制所述可移动平台沿所述目标路径运动。
  2. 根据权利要求1所述的方法，其特征在于，所述车道分布信息包括所述目标路径在每一路段的一个或多个的车道区块的位置。
  3. 根据权利要求1或2所述的方法，其特征在于，所述第一航线和所述第二航线中至少包括部分航线段，所述部分航线段位于同一路段上。
  4. 根据权利要求1-3任一项所述的方法，其特征在于，每一所述航线沿所述目标路径经过多个车道区块；
    所述航线由多个航线段连接组成，多个所述航线段位于连续分布的路段的车道区块。
  5. 根据权利要求1-4任一项所述的方法，其特征在于，所述航线包括转向航线段，所述转向航线段包括相邻的第一航线段和第二航线段，所述目标路径包括第一路段和第二路段，所述第一航线段位于所述第一路段的第一车道区块，所述第二航线段位于所述第二路段的第二车道区块，所述第一车道区块和所述第二车道区块分别位于所述路径的相邻的第一车道和第二车道；
    所述第一路段和所述第二路段为同一路段，或者，所述第一路段与所述第二路段为相邻的两个路段。
  6. 根据权利要求1-5任一项所述的方法，其特征在于，所述航线包括直行航线段，所述直行航线段包括相邻的第一航线段和第二航线段，所述目标路径包括相邻的第一路段和第二路段，所述第一航线段位于所述第一路段的第一车道区块，所述第二航线段位于所述第二路段的第二车道区块，所述第一车道区块和所述第二车道区块分别位于所述目标路径的同一车道。
  7. 根据权利要求1-6任一项所述的方法，其特征在于，根据所述车道分布信息生成多条航线，包括：
    搜索所述车道分布信息中的各车道区块，得到多条航线段；
    根据所述多条航线段，生成所述多条航线。
  8. 根据权利要求7所述的方法，其特征在于，所述搜索所述车道分布信息中的各车道区块，得到多条航线段，包括：
    根据多种预设运动策略，分别搜索所述车道分布信息中的各车道区块，得到多条航线段；
    其中，所述多种预设运动策略包括最少变更车道运动策略、靠目标方向车道运动策略，所述目标方向为最左方向或最左方向。
  9. 根据权利要求8所述的方法，其特征在于，所述预设运动策略包括最少变更车道运动策略；
    所述根据多种预设运动策略，分别搜索所述车道分布信息中的各车道区块，得到多条航线段，包括：
    从所述车道分布信息中所述目标路径的当前车道区块开始沿所述目标路径直行搜索直至第三车道区块，其中，所述第三车道区块的下一车道区块不属于所述目标路径；
    根据直行搜索到的车道区块确定直行航线段；
    从所述第三车道区块开始转向搜索直至搜索到所述目标路径的车道区块，并根据转向搜索到的车道区块确定转向航线段。
  10. 根据权利要求9所述的方法，其特征在于，所述预设运动策略包括靠目标方向车道运动策略；
    所述根据多种预设运动策略，分别搜索所述车道分布信息中的各车道区块，得到多条航线段，包括：
    从所述车道分布信息中所述目标路径的当前车道区块开始朝所述目标方向转向搜索直至第三车道区块，其中，所述第三车道区块为不能再继续朝所述目标方向转向的车道区块；
    根据转向搜索到的车道区块确定转向航线段；
    从所述第三车道区块开始搜索回到所述目标路径中的必经车道区块，并根据搜索到的车道区块确定从所述第三车道区块回到所述必经车道区块的航线段。
  11. 根据权利要求1-10任一项所述的方法，其特征在于，所述方法还包括：
    根据多条所述航线和所述车道分布信息，分别获得多条所述航线的可转向区域；
    所述根据所述目标航线，控制所述可移动平台沿所述目标路径运动，包括：
    根据所述目标航线和所述目标航线的可转向区域，控制所述可移动平台沿所述目标路径运动。
  12. 根据权利要求11所述的方法，其特征在于，所述根据所述目标航线和所述目标航线的可转向区域，控制所述可移动平台沿所述目标路径运动，包括：
    根据所述目标航线，控制所述可移动平台沿所述目标路径运动；
    在所述可移动平台沿所述目标路径运动的过程中，检测所述可移动平台的环境信息；
    根据所述环境信息和所述目标航线上的可转向区域，控制所述可移动平台沿所述目标路径在所述可转向区域内转向运动。
  13. 根据权利要求12所述的方法，其特征在于，所述根据所述环境信息和所述目标航线上的可转向区域，控制所述可移动平台沿所述目标路径在所述可转向区域内转向运动，包括：
    若根据所述环境信息检测到所述可移动平台运动需经过障碍物所在的位置，则根据所述目标航线上的可转向区域，控制所述可移动平台在所述可转向区域转向运动，以规避所述障碍物。
  14. 根据权利要求11-13任一项所述的方法，其特征在于，所述根据多条所述航线和所述车道分布信息，分别获得多条所述航线的可转向区域，包括：
    确定所述航线中转向航线段所在的第三车道和第四车道；
    根据所述车道分布信息，获取所述航线中必经的第四车道区块和必经的第五车道区块，所述第四车道区块为所述第三车道中的车道区块，第五车道区块为所述第四车道中的车道区块；
    根据所述第四车道区块所在路段与所述第五车道区块所在路段之间的车道区块，确定所述航线的可转向区域。
  15. 根据权利要求1-14任一项所述的方法，其特征在于，所述从多条所述航线中，确定目标航线，包括：
    根据用户选定的目标参考指标，从所述多条航线中确定所述目标航线。
  16. 根据权利要求15所述的方法，其特征在于，所述根据用户选定的目标参考指标，从所述多条航线中确定所述目标航线，包括：
    根据所述目标参考指标，确定每条航线在所述目标参考指标上的参数值；
    根据参数值与参数阈值之间的大小关系，从所述多条航线中确定所述目标航线。
  17. 根据权利要求16所述的方法，其特征在于，所述方法还包括如下一项或多项：
    根据用户选定的目标参考指标，确定与所述目标参考指标相对应的所述参数阈值；
    获取用户输入的所述参数阈值；
    根据所述可移动平台沿所述目标路径运动的情景需求，确定所述参数阈值；
    采用机器学习的方式，将所述可移动平台的运动结果作为输入，确定所述参数阈值；
    根据所述可移动平台检测到的环境信息，确定所述参数阈值。
  18. 根据权利要求1-14任一项所述的方法，其特征在于，所述方法还包括：
    获取多种参考指标的权重；
    所述从多条所述航线中，确定目标航线，包括：
    根据所述多种参考指标的权重，从所述多条航线中，确定所述目标航线。
  19. 根据权利要求18所述的方法，其特征在于，所述获取多种参考指标的权重，包括如下一项或多项：
    根据用户选定的目标参考指标，生成所述多种参考指标的权重；
    获取用户输入的所述多种参考指标的权重；
    根据所述可移动平台沿所述目标路径运动的情景需求，确定所述多种参考指标的权重；
    采用机器学习的方式，将所述可移动平台的运动结果作为输入，确定所述多种参考指标的权重；
    根据所述可移动平台检测到的环境信息，确定所述多种参考指标的权重。
  20. 根据权利要求18或19所述的方法，其特征在于，所述根据所述多种参考指标的权重，从所述多条航线，确定所述目标航线，包括：
    根据所述多种参考指标和所述多条航线，获取每条航线在每种参考指标上的评分；
    根据每条航线分别在所述多种参考指标上的评分以及所述多种参考指标的权重，获得每条航线对应的总评分；
    根据多条所述航线分别对应的总评分，从所述多条航线中确定目标航线。
  21. 根据权利要求1-20任一项所述的方法，其特征在于，所述从多条所述航线中，确定目标航线，包括：
    检测所述可移动平台所处的环境信息；
    根据所述环境信息，从所述多条航线，确定所述目标航线。
  22. 根据权利要求1-21任一项所述的方法，其特征在于，所述方法还包括：
    在根据所述目标航线控制所述可移动平台运动的过程中，检测所述可移动平台所处的环境信息；
    根据所述环境信息，从所述多条航线中，再次确定目标航线；
    根据再次确定的目标航线，控制所述可移动平台继续沿所述目标路径运动。
  23. 根据权利要求22所述的方法，其特征在于，所述根据所述环境信息，从所述多条航线中，再次确定目标航线，包括：
    若根据所述环境信息确定所述可移动平台运动需经过障碍物所在的位置，则从所述多条航线，再次确定目标航线，以规避所述障碍物。
  24. 一种可移动平台的控制装置，其特征在于，包括：存储器和处理器；
    所述存储器，用于存储指令；
    所述处理器，调用所述存储器存储的指令用于执行以下操作：
    获取所述可移动平台运动的目标路径；
    获取所述目标路径的车道分布信息；
    根据所述车道分布信息生成多条航线，多条所述航线包括第一航线和第二航线，所述第一航线包括第一航线段，所述第二航线包括第二航线段，所述第一航线段和所述第二航线段位于所述目标路径的不同车道；
    从多条所述航线中，确定目标航线；
    根据所述目标航线，控制所述可移动平台沿所述目标路径运动。
  25. 根据权利要求24所述的装置，其特征在于，所述车道分布信息包括所述目标 路径在每一路段的一个或多个的车道区块的位置。
  26. 根据权利要求24或25所述的装置，其特征在于，所述第一航线和所述第二航线中至少包括部分航线段，所述部分航线段位于同一路段上。
  27. 根据权利要求24-26任一项所述的装置，其特征在于，每一所述航线沿所述目标路径经过多个车道区块；
    所述航线由多个航线段连接组成，多个所述航线段位于连续分布的路段的车道区块。
  28. 根据权利要求24-27任一项所述的装置，其特征在于，所述航线包括转向航线段，所述转向航线段包括相邻的第一航线段和第二航线段，所述目标路径包括第一路段和第二路段，所述第一航线段位于所述第一路段的第一车道区块，所述第二航线段位于所述第二路段的第二车道区块，所述第一车道区块和所述第二车道区块分别位于所述路径的相邻的第一车道和第二车道；
    所述第一路段和所述第二路段为同一路段，或者，所述第一路段与所述第二路段为相邻的两个路段。
  29. 根据权利要求24-28任一项所述的装置，其特征在于，所述航线包括直行航线段，所述直行航线段包括相邻的第一航线段和第二航线段，所述目标路径包括相邻的第一路段和第二路段，所述第一航线段位于所述第一路段的第一车道区块，所述第二航线段位于所述第二路段的第二车道区块，所述第一车道区块和所述第二车道区块分别位于所述目标路径的同一车道。
  30. 根据权利要求24-29任一项所述的装置，其特征在于，所述处理器，具体用于：
    搜索所述车道分布信息中的各车道区块，得到多条航线段；
    根据所述多条航线段，生成所述多条航线。
  31. 根据权利要求30所述的装置，其特征在于，所述处理器，具体用于：
    根据多种预设运动策略，分别搜索所述车道分布信息中的各车道区块，得到多条航线段；
    其中，所述多种预设运动策略包括最少变更车道运动策略、靠目标方向车道运动策略，所述目标方向为最左方向或最左方向。
  32. 根据权利要求31所述的装置，其特征在于，所述预设运动策略包括最少变更车道运动策略；
    所述处理器，具体用于：
    从所述车道分布信息中所述目标路径的当前车道区块开始沿所述目标路径直行搜索直至第三车道区块，其中，所述第三车道区块的下一车道区块不属于所述目标路径；
    根据直行搜索到的车道区块确定直行航线段；
    从所述第三车道区块开始转向搜索直至搜索到所述目标路径的车道区块，并根据 转向搜索到的车道区块确定转向航线段。
  33. 根据权利要求32所述的装置，其特征在于，所述预设运动策略包括靠目标方向车道运动策略；
    所述处理器，具体用于：
    从所述车道分布信息中所述目标路径的当前车道区块开始朝所述目标方向转向搜索直至第三车道区块，其中，所述第三车道区块为不能再继续朝所述目标方向转向的车道区块；
    根据转向搜索到的车道区块确定转向航线段；
    从所述第三车道区块开始搜索回到所述目标路径中的必经车道区块，并根据搜索到的车道区块确定从所述第三车道区块回到所述必经车道区块的航线段。
  34. 根据权利要求24-33任一项所述的装置，其特征在于，所述处理器，还用于：
    根据多条所述航线和所述车道分布信息，分别获得多条所述航线的可转向区域；
    所述处理器在根据所述目标航线，控制所述可移动平台沿所述目标路径运动时，具体用于：
    根据所述目标航线和所述目标航线的可转向区域，控制所述可移动平台沿所述目标路径运动。
  35. 根据权利要求34所述的装置，其特征在于，所述处理器，具体用于：
    根据所述目标航线，控制所述可移动平台沿所述目标路径运动；
    在所述可移动平台沿所述目标路径运动的过程中，检测所述可移动平台的环境信息；
    根据所述环境信息和所述目标航线上的可转向区域，控制所述可移动平台沿所述目标路径在所述可转向区域内转向运动。
  36. 根据权利要求35所述的装置，其特征在于，所述处理器，具体用于：
    若根据所述环境信息检测到所述可移动平台运动需经过障碍物所在的位置，则根据所述目标航线上的可转向区域，控制所述可移动平台在所述可转向区域转向运动，以规避所述障碍物。
  37. 根据权利要求34-36任一项所述的装置，其特征在于，所述处理器，具体用于：
    确定所述航线中转向航线段所在的第三车道和第四车道；
    根据所述车道分布信息，获取所述航线中必经的第四车道区块和必经的第五车道区块，所述第四车道区块为所述第三车道中的车道区块，第五车道区块为所述第四车道中的车道区块；
    根据所述第四车道区块所在路段与所述第五车道区块所在路段之间的车道区块，确定所述航线的可转向区域。
  38. 根据权利要求24-37任一项所述的装置，其特征在于，所述处理器，具体用 于：
    根据用户选定的目标参考指标，从所述多条航线中确定所述目标航线。
  39. 根据权利要求38所述的装置，其特征在于，所述处理器，具体用于：
    根据所述目标参考指标，确定每条航线在所述目标参考指标上的参数值；
    根据参数值与参数阈值之间的大小关系，从所述多条航线中确定所述目标航线。
  40. 根据权利要求39所述的装置，其特征在于，所述处理器，还用于执行如下一项或多项：
    根据用户选定的目标参考指标，确定与所述目标参考指标相对应的所述参数阈值；
    获取用户输入的所述参数阈值；
    根据所述可移动平台沿所述目标路径运动的情景需求，确定所述参数阈值；
    采用机器学习的方式，将所述可移动平台的运动结果作为输入，确定所述参数阈值；
    根据所述可移动平台检测到的环境信息，确定所述参数阈值。
  41. 根据权利要求24-37任一项所述的装置，其特征在于，所述处理器，还用于：
    获取多种参考指标的权重；
    所述处理器在从多条所述航线中，确定目标航线时，具体用于：
    根据所述多种参考指标的权重，从所述多条航线中，确定所述目标航线。
  42. 根据权利要求41所述的装置，其特征在于，所述处理器，还用于执行如下一项或多项：
    根据用户选定的目标参考指标，生成所述多种参考指标的权重；
    获取用户输入的所述多种参考指标的权重；
    根据所述可移动平台沿所述目标路径运动的情景需求，确定所述多种参考指标的权重；
    采用机器学习的方式，将所述可移动平台的运动结果作为输入，确定所述多种参考指标的权重；
    根据所述可移动平台检测到的环境信息，确定所述多种参考指标的权重。
  43. 根据权利要求41或42所述的装置，其特征在于，所述处理器，具体用于：
    根据所述多种参考指标和所述多条航线，获取每条航线在每种参考指标上的评分；
    根据每条航线分别在所述多种参考指标上的评分以及所述多种参考指标的权重，获得每条航线对应的总评分；
    根据多条所述航线分别对应的总评分，从所述多条航线中确定目标航线。
  44. 根据权利要求24-43任一项所述的装置，其特征在于，所述处理器，具体用于：
    检测所述可移动平台所处的环境信息；
    根据所述环境信息，从所述多条航线，确定所述目标航线。
  45. 根据权利要求24-44任一项所述的装置，其特征在于，所述处理器，还用于：
    在根据所述目标航线控制所述可移动平台运动的过程中，检测所述可移动平台所处的环境信息；
    根据所述环境信息，从所述多条航线中，再次确定目标航线；
    根据再次确定的目标航线，控制所述可移动平台继续沿所述目标路径运动。
  46. 根据权利要求45所述的装置，其特征在于，所述处理器，具体用于：
    若根据所述环境信息确定所述可移动平台运动需经过障碍物所在的位置，则从所述多条航线，再次确定目标航线，以规避所述障碍物。
  47. 一种可移动平台，其特征在于，包括如权利要求24-46任一项所述的可移动平台的控制装置。
  48. 根据权利要求47所述的可移动平台，其特征在于，所述可移动平台为自动驾驶车辆。
  49. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质中存储有指令，当所述指令在计算机上运行时，使得计算机执行权利要求1-23中任一项所述的方法。
  50. 一种计算机程序产品，包括计算机指令，其特征在于，所述计算机指令被处理器执行时实现权利要求1-23任一项所述的方法。
PCT/CN2021/073832 2021-01-26 2021-01-26 可移动平台的控制方法和装置 Ceased WO2022160100A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN202180083247.2A CN116745581A (zh) 2021-01-26 2021-01-26 可移动平台的控制方法和装置
PCT/CN2021/073832 WO2022160100A1 (zh) 2021-01-26 2021-01-26 可移动平台的控制方法和装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/073832 WO2022160100A1 (zh) 2021-01-26 2021-01-26 可移动平台的控制方法和装置

Publications (1)

Publication Number Publication Date
WO2022160100A1 true WO2022160100A1 (zh) 2022-08-04

Family

ID=82652848

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/073832 Ceased WO2022160100A1 (zh) 2021-01-26 2021-01-26 可移动平台的控制方法和装置

Country Status (2)

Country Link
CN (1) CN116745581A (zh)
WO (1) WO2022160100A1 (zh)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106323305A (zh) * 2015-06-29 2017-01-11 北京四维图新科技股份有限公司 一种导航方法及装置
CN108663059A (zh) * 2017-03-29 2018-10-16 高德信息技术有限公司 一种导航路径规划方法和装置
CN109752016A (zh) * 2018-12-29 2019-05-14 同济大学 一种用于无人低速车的平行行驶航线轨迹生成系统
US20190202462A1 (en) * 2017-12-28 2019-07-04 Automotive Research & Testing Center Method of lane change decision-making and path planning
CN109976332A (zh) * 2018-12-29 2019-07-05 惠州市德赛西威汽车电子股份有限公司 一种用于无人驾驶的高精度地图模型及自主导航系统
CN110553660A (zh) * 2019-08-31 2019-12-10 武汉理工大学 一种基于a*算法和人工势场的无人车轨迹规划方法
CN111289006A (zh) * 2020-03-20 2020-06-16 上海商汤临港智能科技有限公司 车道导航路径生成方法及装置、驾驶控制方法及装置
CN111947678A (zh) * 2020-08-27 2020-11-17 重庆智行者信息科技有限公司 一种针对结构化道路的自动驾驶导航方法及系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106643747B (zh) * 2015-11-04 2020-09-08 中国移动通信集团公司 导航方法及装置
US11003183B2 (en) * 2017-09-13 2021-05-11 Baidu Usa Llc Driving scene based path planning for autonomous driving vehicles
CN111213031A (zh) * 2018-12-06 2020-05-29 深圳市大疆创新科技有限公司 一种导航处理方法、装置及导航处理设备
CN110361028B (zh) * 2019-07-26 2021-03-16 武汉中海庭数据技术有限公司 一种基于自动驾驶循迹的路径规划结果生成方法及系统
CN110794841B (zh) * 2019-11-12 2022-07-12 深圳创维数字技术有限公司 路径导航方法、装置及计算机可读存储介质

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106323305A (zh) * 2015-06-29 2017-01-11 北京四维图新科技股份有限公司 一种导航方法及装置
CN108663059A (zh) * 2017-03-29 2018-10-16 高德信息技术有限公司 一种导航路径规划方法和装置
US20190202462A1 (en) * 2017-12-28 2019-07-04 Automotive Research & Testing Center Method of lane change decision-making and path planning
CN109752016A (zh) * 2018-12-29 2019-05-14 同济大学 一种用于无人低速车的平行行驶航线轨迹生成系统
CN109976332A (zh) * 2018-12-29 2019-07-05 惠州市德赛西威汽车电子股份有限公司 一种用于无人驾驶的高精度地图模型及自主导航系统
CN110553660A (zh) * 2019-08-31 2019-12-10 武汉理工大学 一种基于a*算法和人工势场的无人车轨迹规划方法
CN111289006A (zh) * 2020-03-20 2020-06-16 上海商汤临港智能科技有限公司 车道导航路径生成方法及装置、驾驶控制方法及装置
CN111947678A (zh) * 2020-08-27 2020-11-17 重庆智行者信息科技有限公司 一种针对结构化道路的自动驾驶导航方法及系统

Also Published As

Publication number Publication date
CN116745581A (zh) 2023-09-12

Similar Documents

Publication Publication Date Title
JP7576633B2 (ja) 目標軌道計画方法
CN109583151B (zh) 车辆的行驶轨迹预测方法及装置
JP6809611B2 (ja) 走行支援方法及び走行支援装置
WO2021071810A1 (en) Methods and systems for topological planning in autonomous driving
JP6315107B2 (ja) 目標経路生成装置および走行制御装置
CN103935361A (zh) 用于自主的车道变换、经过和超越行为的有效数据流算法
CN110275524A (zh) 车辆控制装置
CN108693869A (zh) 车辆控制装置
JP2023506840A (ja) ブロックされる領域の案内
JP7716103B2 (ja) 自律走行車のための軌道形状生成方法及び装置
WO2022213373A1 (zh) 一种轨迹规划方法以及相关设备
JP2021115973A (ja) 車両制御装置
US20190056737A1 (en) Autonomous behavior control using policy triggering and execution
JP2012216069A (ja) 車両および車両制御プログラム
JP2009157502A (ja) 進路評価装置
CN114834451A (zh) 自主车辆的主动式车道变换
US11531865B2 (en) Systems and methods for parallel autonomy of a vehicle
JPWO2018173403A1 (ja) 車両制御装置及び車両制御方法
JP2021196874A (ja) 走行支援方法及び走行支援装置
JP7164640B2 (ja) 車両システム
JP7408793B2 (ja) 走行経路生成装置、走行経路生成方法、及び自動運転システム
JP6288305B2 (ja) 目標車速生成装置および走行制御装置
JP2023151307A (ja) 走路推定方法及び走路推定装置
CN116745581A (zh) 可移动平台的控制方法和装置
JP2022139058A (ja) 制御システム、移動体、制御方法、およびプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21921720

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 202180083247.2

Country of ref document: CN

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21921720

Country of ref document: EP

Kind code of ref document: A1