WO2022142858A9 - 机器人移动路径规划方法、确定规划的路径点偏离历史路径程度的方法、装置、机器人及计算机可读存储介质 - Google Patents
机器人移动路径规划方法、确定规划的路径点偏离历史路径程度的方法、装置、机器人及计算机可读存储介质 Download PDFInfo
- Publication number
- WO2022142858A9 WO2022142858A9 PCT/CN2021/132198 CN2021132198W WO2022142858A9 WO 2022142858 A9 WO2022142858 A9 WO 2022142858A9 CN 2021132198 W CN2021132198 W CN 2021132198W WO 2022142858 A9 WO2022142858 A9 WO 2022142858A9
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- path
- point
- historical
- planned
- global
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 60
- 238000001514 detection method Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 5
- 238000012790 confirmation Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 description 14
- 238000010586 diagram Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 8
- 230000009191 jumping Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
Definitions
- the present application relates to the field of robotics, and in particular, to a method for planning a moving path of a robot, a method, an apparatus, a robot, and a computer-readable storage medium for determining the degree to which a planned path point deviates from a historical path.
- the path planning of mobile robots is usually divided into global planning and local planning.
- the global planning mainly plans an infinite path from the starting point to the target point. Collision global path.
- Local planning usually selects forward-looking points as "stage goals" on the global path according to the set forward-looking distance, while avoiding obstacles encountered during the movement. Therefore, the effect of global planning will affect the performance of local planning, which in turn affects the performance of mobile robots.
- a method for planning a movement path of a robot a method, an apparatus, a robot, and a computer-readable storage medium for determining the degree of deviation of a planned path point from a historical path are provided.
- a method for planning a movement path of a robot comprising:
- a method and device for planning a moving path of a robot comprising:
- the judgment module is used to judge whether there is an obstacle in the planned historical path when the global path of the robot is not planned for the first time; the selection module is used to select the historical path and the obstacle if the obstacle occurs.
- the coincident illegal path point is used as the starting point of the global path of this planning; the confirmation module is used to confirm the starting point of the global path from the current planning to The path points between the target points make the deviation of the global path of the current planning and the historical path less than the preset range; the control module is used to obtain the planned path points according to the confirmed path points and the target points.
- a global path controlling the robot to move along the current planned global path.
- a method for determining the degree to which a planned path point deviates from a historical path comprising: during this path planning, determining the distance p dist between the currently to-be-calculated path point and the closest point of the historical path; The deviation weight value w is determined according to the distance p dist and the deviation weight value w to determine the degree to which the current path point to be calculated deviates from the historical path.
- a device for determining the degree to which a planned waypoint deviates from a historical route comprising:
- the closest distance determination module is used to determine the distance p dist between the current path point to be calculated and the closest point of the historical path during this path planning;
- the deviation weight acquisition module is used to obtain the preconfigured deviation weight value corresponding to the current moving scene w;
- a deviation degree determination module configured to determine the degree to which the currently to-be-calculated path point deviates from the historical path according to the distance p dist and the deviation weight value w.
- a robot comprising:
- a memory and a processor stores executable program codes; the processor coupled with the memory invokes the executable program codes stored in the memory to execute the above-mentioned robot movement path planning method .
- a robot comprising: a memory and a processor
- the memory stores executable program codes; the processor coupled with the memory calls the executable program codes stored in the memory to execute the above-mentioned method of determining the degree of deviation of the planned path point from the historical path. method.
- a computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, the steps of the above method are implemented.
- FIG. 1 is a schematic flowchart of a method for planning a movement path of a robot according to an embodiment of the present application
- FIG. 2 is a schematic diagram of the principle of jumping of the planned moving path when there is a remote obstacle of the robot in the embodiment of the application;
- FIG. 3 is a schematic flowchart of a method for planning a movement path of a robot provided by another embodiment of the present application.
- FIG. 4 is a schematic diagram of the comparison of the fit between the global path and the historical path planned by using different algorithms in the embodiment of the present application;
- FIG. 5 is a schematic structural diagram of a robot movement path planning device provided by an embodiment of the present application.
- FIG. 6 is a schematic structural diagram of a robot movement path planning device provided by another embodiment of the present application.
- [Correction 27.01.2022 under Rule 91] 8 is a schematic structural diagram of an apparatus for determining the degree of deviation of a planned path point from a historical path according to an embodiment of the present application;
- FIG. 9 is a schematic structural diagram of a robot according to an embodiment of the present application.
- the embodiment of the present application provides a method for planning a moving path of a robot.
- a preset path planning algorithm is used to globally plan the moving path of the robot from the starting point to the target point (ie, the end point), and repeating
- the optimal moving path can be planned by performing global planning multiple times.
- the preset path planning algorithm is specifically a search-based path planning algorithm, which may include: A*(A-star) algorithm, Dijkstra algorithm, etc.
- A* algorithm is used as an example when describing the solution in this embodiment, and other search-based path planning algorithms such as the Dijkstra algorithm can be similarly deduced.
- the traditional A* algorithm is used for path planning, the current position of the robot is set as the starting point, and the global target is the target point for planning, and the planned historical path is obtained.
- the non-first global planning that is, in the second and subsequent global planning, first determine whether the historical path is legal, that is, determine whether the robot will encounter obstacles when moving on the historical path. History path is invalid. If the historical path is valid, the historical path is maintained. If the historical path is illegal, first find the illegal path point closest to the robot along the historical path, take the illegal path point as the starting point of this global planning, and use the optimized A* algorithm to control the robot's The movement path is globally planned.
- the optimized A* algorithm based on the A* algorithm is to modify the calculation formula of the mobile cost value f n .
- the rest of the pathfinding steps are the same as the A* algorithm.
- the global path obtained by the optimized A* algorithm is as close as possible to the above
- the historical path is fitted once to avoid large jumps.
- pn represents the degree to which the path point n to be calculated deviates from the historical path
- pn w ⁇ p dist
- p dist is the distance between the current path point to be calculated and the closest point of the historical path
- w is the weight value , which can be updated in the process of calculating the planned path, and can be specifically determined according to the effect of the actually generated path. If the global path planned this time coincides or almost coincides with the historical path, it means that w is too large, and it may be too close to the obstacle when the environment changes, so the value of w needs to be reduced, but if w is too small, it is easy to happen Jump, you need to increase the value of w.
- the optimized A* algorithm is used to calculate the global path, detect whether the global path has a path hop, and after it is determined that the global path has hopped, the global planning will not be performed within a certain period of time, and the historical path will be used as the current global path. path.
- the above-mentioned method for planning a moving path of a robot can be applied to a robot or an intelligent terminal.
- the intelligent terminal and the robot are connected to the same wireless network and send the calculated global path to the robot.
- the smart terminal can be a computer, a mobile phone or a wearable smart device, and usually has structures such as an arithmetic unit, a controller, a memory, an input device, and an output device.
- the robot movement path planning method is described in detail below.
- FIG. 1 a schematic flowchart of a method for planning a movement path of a robot provided by an embodiment of the present application.
- the method can be applied to robots or smart terminals, as shown in Figure 1, the method specifically includes:
- Non-first planning refers to planning after the first planning.
- the historical path refers to the global path obtained by the previous global planning of the current global path planning.
- the historical path is an illegal path
- the path points that coincide with the obstacle in the historical path are illegal path points. These illegal path points will hinder the robot from moving on the path point. To move, the robot needs to avoid.
- the robot when the robot performs the first global path planning, the current position of the robot is used as the starting point of the planning.
- the robot performs global path planning again while moving according to the previous global path (that is, the historical path), and stops the global path planning when it moves to the target point.
- obstacles may be added to the planned moving path, making the planned historical path an illegal path.
- the second and subsequent global paths first determine whether the previously planned historical path is an illegal path, that is, whether there are obstacles on the historical path, and can judge whether there are obstacles on the historical path through the collected sensor data Obstacles, the sensor can be ranging sensor or vision sensor, etc.
- an illegal path point on the historical path that coincides with the obstacle is selected as the starting point of the global path of this planning, and using it as the starting point can shorten the overall planning time.
- the path length is reduced, the amount of calculation is reduced, and the global path jumps in the near distance due to noise in the observation of obstacles in the distance, which affects the smoothness of the robot's movement.
- Fig. 2 is a schematic diagram of the principle of the planned movement path jumping when there is a remote obstacle between the starting point and the target point.
- There are obstacles for example, more than 2 meters), which will cause the cost of moving path 1 and moving path 2 to be almost the same in global planning.
- This will cause the forward-looking point of the robot's local planning to jump on path 1 and path 2, causing the machine to run unsmoothly.
- the starting point of the global planning is set as an illegal path point on the historical path, the obstacle will become a near-end obstacle, and the situation of two paths will not occur during the global planning, which can avoid the occurrence of the global planning. jump.
- Deviation from the historical path cost item indicating the degree to which the path point to be calculated deviates from the historical path.
- the distance information between the path point to be calculated and the starting point of the current planning global path is used to measure this time. How far the planned global path deviates from this historical path.
- Adding the deviating historical path cost item to the path algorithm can make the deviation between the global path planned this time and the historical path less than the preset range, that is, keep the global path planned this time close to the historical path, to avoid the current calculation
- the global path deviates too much from the historical path, resulting in a large jump of the robot.
- the value of the preset range is related to the robot's task scene, the robot's task content, and the robot's own volume. For example, in this task scene, there are many obstacles, and the path that the robot needs to move is complex, and the task to be performed requires the path accuracy of the robot's movement. If it is higher, or the robot is large, the preset range needs to be smaller to improve the smoothness of the robot's movement.
- the moving path of the robot is simple, and the tasks performed require the robot to move with a low path accuracy, or the robot is small in size.
- the preset range is 0.3 to 0.5 meters.
- the above steps S101 to S103 are repeatedly executed until the robot reaches the target point.
- the starting point of the global path shortens the path length of the global planning, reduces the amount of calculation, and avoids the global path jumping in the vicinity due to the noise of the distant obstacle observation, and through the preset including the deviation from the historical path cost item
- the path algorithm is used to confirm the path points from the starting point of the current planning global path to the target point, so that the deviation between the current planning global path and the historical path is less than the preset range, so as to avoid the global path caused by this calculation.
- the path deviates too much from the historical path, resulting in a large jump of the robot.
- the global path planned for this time can be obtained by the above method, which can reduce the problems caused by environmental changes, sensor noise effects and changes in obstacles when planning the global path. There are frequent jumps, reducing the impact on local planning and improving the fluency of robot movement.
- FIG. 3 a schematic flowchart of a method for planning a movement path of a robot provided by another embodiment of the present application.
- the method can be applied to robots or smart terminals, as shown in Figure 3, the method specifically includes:
- the first illegal waypoint is the illegal waypoint closest to the robot among all illegal waypoints in the historical path that coincide with the obstacle.
- the first illegal point that coincides with the obstacle is selected as the starting point of the global path of this planning, and the position where the obstacle is observed at the first time can be used as the starting point of the global path of this planning, and the maximum reduction is achieved. Because of the impact of the appearance of the obstacle on the global path planning.
- the existing A* algorithm is to calculate a path with the least cost from the starting point of the plan to the goal point.
- the path-finding area between the starting point and the target point of the robot is divided into several grids, each grid can be regarded as a path point to be calculated, f n represents the moving cost of the path point, after determining the path-finding
- each path point to be calculated has a moving cost value f n
- the preset path algorithm in this embodiment is an algorithm optimized on the basis of the A* algorithm. Specifically, the calculation formula of the moving cost value f n is modified, and the cost item p n deviating from the historical path is added, that is, the calculation formula of f n is added.
- p n represents the degree to which the path point n to be calculated deviates from the historical path
- p n w ⁇ p dist
- p dist is the distance between the current path point to be calculated and the closest point of the historical path
- w is the deviation weight
- the value of w can be determined according to the effect of the planned path and the actually generated path before the robot performs global path planning. If the planned global path coincides or almost coincides with the historical path, it means that w is too large, and it may be too close to the obstacle when the environment changes, so the value of w needs to be reduced, but if the global path planning is easy to occur If it jumps, it means that w is too small, and the value of w needs to be increased.
- the w value can be set at one time before planning the path, and the w value is fixed in the process of planning the global path
- the value of w is the optimal value of the current scene
- the value of w can be adjusted accordingly.
- the w value corresponding to the changed scene is obtained from the local database or cloud server, and the configuration is updated to the path algorithm. , to improve the accuracy of planning the global path.
- the three figures in Figure 4 are (a), (b) and (c) from left to right.
- (a) represents the last planned historical path at time 1; due to the influence of sensor observation errors or robot positioning errors, the obstacles in (a) occur at the positions of obstacles in (b) and (c)
- the path marked by the solid line in (b) represents the global path of this planning calculated by the existing traditional A* algorithm at time 2, and the path marked by the dotted line represents the historical path in (a).
- the path marked by the solid line in (c) represents the global path of this planning calculated by the preset path algorithm provided in this embodiment at time 2
- the path marked by the dotted line represents (a) The historical path in , because the preset path algorithm considers the historical path cost, the planned global path will be more suitable for the historical path.
- the preset path algorithm is also provided with an open list and a closed list, respectively saving the path points that can be considered to be calculated and the path points that are no longer considered, and each time from the open list to select a value to be calculated with the smallest f n value.
- the pathfinding path it is judged whether the pathpoint is the target point, if so, the pathfinding is successful, and the algorithm ends; otherwise, the pathfinding continues.
- the preset path algorithm refer to the traditional A* algorithm, which will not be repeated here.
- the detection method includes: calculating the distance between the path point in the current planned global path and the closest path point on the historical path, if in the current planned global path, the path whose distance is greater than the first preset distance If the ratio of points exceeds the preset ratio, it is confirmed that the planned global path has a path jump, that is, there are too many path points on the current global path, and these path points are all distant from the path points on the previous historical path. If it is too far, in this case, it can be determined that the global path planned for this time has a path hopping.
- the first preset distance is preferably 0.5 meters, and the preset ratio is preferably 50%.
- the detection method further includes: when the robot continues to re-plan an illegal global path within the first preset time period within the preset moving distance range, confirming that the current planned global path has jumped.
- the preset moving distance range is preferably a radius of 0.5 meters, and the first preset duration is preferably 20 seconds.
- step S204 is executed; if a jumping occurs, step S205 is executed.
- S205 Stop planning the global path within the second preset time period, take the historical path planned last time as the global path planned this time, and control the robot to move along the historical path.
- the second preset duration is preferably 5 seconds. Stop planning the global path, you can perform local path planning, or you can clear obstacles first, and then start the global path planning.
- the starting point of the global path planned this time shortens the path length of the global planning, reduces the amount of calculation, and avoids the global path jumping in the vicinity due to noise in the observation of distant obstacles, and by including the cost of deviating from the historical path
- the preset path algorithm of the item confirms the path points from the starting point of the current planning global path to the target point, so that the deviation between the current planning global path and the historical path is less than the preset range, avoiding the The global path calculated this time deviates too much from the historical path, resulting in a large jump of the robot. After the global path planned for this time is obtained, the jump detection is performed on it.
- the historical path calculated last time is used as the current global path, and the planned global path is obtained by the above method, which can reduce the existence of existing global paths caused by environmental changes, sensor noise effects and obstacles changes.
- the frequent jumps of the robot reduce the impact on local planning and improve the fluency of robot movement.
- FIG. 5 a schematic structural diagram of a robot movement path planning device provided by an embodiment of the present application. For the convenience of description, only the parts related to the embodiments of the present application are shown.
- the device can be installed in a robot or an intelligent terminal.
- the device includes:
- the judgment module 301 is used for judging whether there are obstacles in the planned historical path when the global path of the robot is not planned for the first time;
- the selection module 302 is configured to select an illegal path point in the historical path that coincides with the obstacle as the starting point of the global path planned this time if an obstacle appears;
- the confirmation module 303 is used to confirm the path point from the starting point of the current planned global path to the target point by deviating from the preset path algorithm of the historical path cost item, so that the difference between the current planned global path and the historical path is confirmed.
- the deviation is less than the preset range;
- the control module 304 is configured to obtain the current planned global path according to the confirmed path point and the target point, and control the robot to move along the current planned global path.
- the selection module 302 is also used to select the first illegal path point that coincides with the obstacle in the historical path, as the starting point of the global path planned this time, and the first illegal path point is the closest to the robot. Invalid waypoint.
- the confirmation module 303 is further configured to obtain a preset path algorithm after the A* algorithm is optimized as follows;
- the moving cost formula of the path point to be calculated in the A* algorithm is set as:
- g n represents the actual cost from the starting point to the path point n to be calculated
- h n represents the estimated cost from the path point n to be calculated to the target point
- p n represents the deviation of the path point n to be calculated from the historical path
- the cost of the historical path p dist is the distance between the path point n to be calculated and the closest point of the historical path
- w is the pre-configured deviation weight value
- the robot movement path planning device provided in another embodiment further includes: a configuration module 401;
- the configuration module 401 is configured to acquire a deviation weight value corresponding to the changed movement scene when it is detected that the movement scene of the robot changes, and update and configure the deviation weight value into a preset path algorithm.
- the apparatus further includes: a detection module 402;
- the detection module 402 is configured to detect, according to a preset detection rule, whether a path jump occurs in the current planned global path;
- the trigger control module 304 uses the current planned global path as the movement path of the robot; if the path jump occurs, the trigger control module 304 uses the historical path as the movement path of the robot.
- the detection module 402 is also used to calculate the distance between the path point in the current planned global path and the closest path point on the historical path; if in the current planned global path, the path whose distance is greater than the first preset distance If the ratio of points exceeds the preset ratio, it is confirmed that the global path planned for this time has a path jump.
- the detection module 402 is further configured to confirm that the current planned global path has a path jump when the robot continues to re-plan an illegal global path within a preset moving distance range within a first preset time period.
- the control module 304 is further configured to stop planning the global path within the second preset time period, and use the historical path planned last time as the global path planned this time.
- the preset path algorithm of the cost item confirms the path points from the starting point of the current planning global path to the target point, so that the deviation between the current planning global path and the historical path is less than the preset range, avoiding the The global path calculated this time deviates too much from the historical path, resulting in a large jump of the robot. After the global path planned for this time is obtained, the jump detection is performed on it, and the detected global path is confirmed as the current global path.
- the historical path calculated last time is used as the current global path, and the planned global path is obtained by the above method, which can reduce the problems caused by environmental changes, sensor noise effects, and changes in obstacles when planning the global path. There are frequent jumps, reducing the impact on local planning and improving the fluency of robot movement.
- an embodiment of the present application also provides a method for determining the degree to which a planned path point deviates from a historical path, and the method specifically includes:
- Step S701 during this path planning, determine the distance p dist between the current path point to be calculated and the closest point of the historical path;
- Step S702 obtaining a preconfigured deviation weight value w corresponding to the current mobile scene
- the type, size, location and other information of obstacles are consistent, and the w value can be set at one time before planning the path.
- the w value is a fixed value
- the w value is the optimal value of the current scene
- the value of w can be adjusted accordingly.
- the w value corresponding to the changed scene is obtained from the local database or cloud server, and the configuration is updated to the path algorithm. , to improve the accuracy of planning the global path.
- Step S703 according to the distance p dist and the deviation weight value w, determine the degree to which the current path point to be calculated deviates from the historical path.
- the degree p n that the path point to be calculated currently deviates from the historical path is calculated according to the following formula:
- steps S701-S703 please refer to the above embodiments, which will not be repeated here.
- the embodiment of the present application also provides a device for determining the degree of deviation of the planned path point from the historical path, and the device specifically includes:
- the closest distance determination module 81 is used to determine the distance p dist between the current path point to be calculated and the closest point of the historical path in this path planning;
- the deviation weight obtaining module 82 obtains the preconfigured deviation weight value w corresponding to the current mobile scene
- the deviation degree determination module 83 is configured to determine the degree to which the path point to be calculated currently deviates from the historical path according to the distance p dist and the deviation weight value w.
- the deviation degree determination module 83 calculates the degree p n that the path point to be calculated currently deviates from the historical path according to the following formula:
- an embodiment of the present application further provides a robot, including a memory 100 and a processor 200, and the processor 200 may be a central processing unit of the robot.
- Storage 100 such as hard drive memory, non-volatile memory (such as flash memory or other electronically programmable limit erasure memory used to form solid state drives, etc.), volatile memory (such as static or dynamic random access memory, etc.), etc.,
- the embodiments of the present application are not limited.
- the memory 100 stores executable program codes; the processor 200 coupled with the memory 100 invokes the executable program codes stored in the memory 100 to execute the above-mentioned robot movement path planning method.
- the embodiment of the present application also provides a robot, which also has the structure shown in FIG. 7 , the difference is that when the processor 200 calls the executable program code stored in the memory 100 to execute the determined plan as described above method to measure the degree to which the path points deviate from the historical path.
- an embodiment of the present application also provides a computer-readable storage medium, which may be provided in the robot in the above-mentioned embodiments, and the computer-readable storage medium may be the one shown in FIG. 9 above.
- Memory 100 in an embodiment.
- a computer program is stored on the computer-readable storage medium, and when the program is executed by the processor, it realizes the robot movement path planning method described in the embodiment shown in FIG. 1 and FIG. A method for determining the extent to which a planned waypoint deviates from a historical route.
- the computer-storable medium can also be a USB flash drive, a removable hard disk, a read-only memory (ROM, Read-Only Memory), a RAM, a magnetic disk or an optical disk and other media that can store program codes.
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Manipulator (AREA)
Abstract
一种机器人移动路径规划方法、装置和机器人，其中机器人移动路径规划方法包括：对机器人的全局路径进行非首次规划时，判断规划的历史路径是否出现障碍物，若出现，则将历史路径中与障碍物重合的不合法路径点作为本次规划的全局路径的起始点（S101），通过包括偏离历史路径代价项的预设的路径算法，确认从本次规划的全局路径的起始点到目标点之间的路径点，使得本次规划的全局路径与历史路径的偏离度小于预设范围（S102），根据确认的路径点和目标点得到本次规划的全局路径，控制机器人沿本次规划的全局路径移动（S103）。
Description
本申请要求于2020年12月28日提交中国国家知识产权局专利局、申请号为202011581919.6、申请名称为“机器人移动路径规划方法、装置及机器人”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
本申请涉及机器人技术领域，特别是一种机器人移动路径规划方法、确定规划的路径点偏离历史路径程度的方法、装置、机器人及计算机可读存储介质。
随着人工智能技术的发展，各种智能自主移动机器人应用越来越广泛，目前移动机器人的路径规划通常分为全局规划和局部规划，其中，全局规划主要规划一条从起始点到目标点的无碰撞全局路径。局部规划则通常是在全局路径上根据设定的前瞻距离选取前瞻点作为一个个“阶段目标”进行运动，同时避开运动过程中遇到的障碍物。因此，全局规划的效果会影响局部规划的性能，进而影响移动机器人的运行表现。
现有技术中，由于环境变化、传感器噪声、动态障碍物等的影响，在规划全局路径时会存在频繁跳变的情况。全局路径的跳变会影响局部规划的效果，进而导致机器人运行过程中的运动不连续，左右摆动等异常现象，影响机器人的运行效果。
发明内容
根据本申请的各种实施例，提供一种机器人移动路径规划方法、确定规划的路径点偏离历史路径程度的方法、装置、机器人及计算机可读存储介质。
一种机器人移动路径规划方法，包括：
对机器人的全局路径进行非首次规划时，判断规划的历史路径是否出现障 碍物，若出现，则选择所述历史路径中与所述障碍物重合的不合法路径点作为本次规划的全局路径的起始点；通过包括偏离历史路径代价项的预设的路径算法，确认从所述本次规划的全局路径的起始点到所述目标点之间的路径点，使得所述本次规划的全局路径与所述历史路径的偏离度小于预设范围；根据确认的路径点和所述目标点得到本次规划的全局路径，控制所述机器人沿所述本次规划的全局路径移动。
一种机器人移动路径规划方法装置，包括：
判断模块，用于对机器人的全局路径进行非首次规划时，判断规划的历史路径是否出现障碍物；选择模块，用于若出现所述障碍物，则选择所述历史路径中与所述障碍物重合的不合法路径点作为本次规划的全局路径的起始点；确认模块，用于通过包括偏离历史路径代价项的预设的路径算法，确认从所述本次规划的全局路径的起始点到目标点之间的路径点，使得所述本次规划的全局路径与所述历史路径的偏离度小于预设范围；控制模块，用于根据确认的路径点和所述目标点得到本次规划的全局路径，控制所述机器人沿所述本次规划的全局路径移动。
一种确定规划的路径点偏离历史路径程度的方法，包括：在本次路径规划时，确定当前待计算的路径点距离历史路径最近点的距离p
dist；获取与当前移动场景所对应的预先配置的偏离权重值w；根据所述距离p
dist与偏离权重值w确定当前待计算的路径点偏离历史路径的程度。
一种确定规划的路径点偏离历史路径程度的装置，包括：
最近距离确定模块，用于在本次路径规划时，确定当前待计算的路径点距离历史路径最近点的距离p
dist；偏离权重获取模块，获取与当前移动场景所对应的预先配置的偏离权重值w；偏离程度确定模块，用于根据所述距离p
dist与偏离权重值w确定当前待计算的路径点偏离历史路径的程度。
一种机器人，包括：
存储器和处理器；所述存储器存储有可执行程序代码；与所述存储器耦合的所述处理器，调用所述存储器中存储的所述可执行程序代码，执行如上所述的机器人移动路径规划方法。
一种机器人，包括：存储器和处理器；
所述存储器存储有可执行程序代码；与所述存储器耦合的所述处理器，调用所述存储器中存储的所述可执行程序代码，执行如上所述的确定规划的路径点偏离历史路径程度的方法。
一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如上所述方法的步骤。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和优点将从说明书、附图以及权利要求书变得明显。
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他实施例的附图。
图1为本申请一实施例提供的机器人移动路径规划方法的流程示意图；
图2为本申请实施例中存在机器人的远端障碍物时规划的移动路径发生跳变的原理示意图；
图3为本申请另一实施例提供的机器人移动路径规划方法的流程示意图；
图4为本申请实施例中使用不同算法规划的全局路径与历史路径贴合度对比示意图；
图5为本申请一实施例提供的机器人移动路径规划装置的结构示意图；
图6为本申请另一实施例提供的机器人移动路径规划装置的结构示意图；
[根据细则91更正 27.01.2022]
图7为本申请一实施例提供的确定规划的路径点偏离历史路径程度的方法的流程示意图；
图7为本申请一实施例提供的确定规划的路径点偏离历史路径程度的方法的流程示意图；
[根据细则91更正 27.01.2022]
图8为本申请一实施例提供的确定规划的路径点偏离历史路径程度的装置的结构示意图；
图8为本申请一实施例提供的确定规划的路径点偏离历史路径程度的装置的结构示意图；
[根据细则91更正 27.01.2022]
图9为本申请一实施例提供的机器人的结构示意图。
图9为本申请一实施例提供的机器人的结构示意图。
为了便于理解本申请，下面将参照相关附图对本申请进行更全面的描述。附图中给出了本申请的较佳实施例。但是，本申请可以以许多不同的形式来实现，并不限于本文所描述的实施例。相反地，提供这些实施例的目的是使对本申请的公开内容的理解更加透彻全面。
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发明的技术领域的技术人员通常理解的含义相同。本文中在发明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，不是旨在限制本申请。本文所使用的术语“和/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。
本申请实施例提供了一种机器人移动路径规划方法，在规划机器人的移动路径时，通过预设的路径规划算法对机器人的移动路径进行从起始点到目标点(即终点)的全局规划，反复多次进行全局规划，可以规划出最优的移动路径，预设的路径规划算法具体为基于搜索的路径规划算法，可包括：A*(A-star)算法、Dijkstra算法等。为便于描述，本实施例具体阐述方案时均以A*算法为例，同理可推及Dijkstra算法等其他基于搜索的路径规划算法。
具体地，对于该机器人的首次全局规划，采用传统的A*算法进行路径规划，将机器人的当前位置设为起始点，全局目标为目标点进行规划，得到规划的历史路径。在非首次全局规划时，即，在第二次及以后的全局规划时，首先判断历史路径是否合法，即判断机器人在该历史路径移动时是否会碰到障碍物，会碰到则判定为该历史路径不合法。如果历史路径合法，则维持该条历史路径。如果该历史路径不合法，则沿着该历史路径首先找到离机器人最近的不合法的路径点，把该不合法的路径点作为本次全局规划的起始点，采用优化的A*算法对机器人的移动路径进行全局规划。
具体地，将传统的A*算法中移动代价值f
n的计算公式修改为：f
n＝g
n+h
n+p
n，其中，g
n表示从起始点到待计算的路径点n的实际代价，h
n表示从待计算的路径点n到目标点的估算代价，p
n表示偏离历史路径的代价项。基于A*算法的优化的A*算法是修改移动代价值f
n的计算公式，其余寻路步骤与A*算法相同，通过优化后的A*算法得到的本次计算的全局路径尽可能与上一次的历史路径贴合，避免出现大的跳变。本申请在进行全局规划时，将距离历史路径更近的路径点确认更优选，从而选取一条最靠近历史路径的路径，从而尽量避免机器人发生跳变。
具体地，p
n表示待计算的路径点n偏离该历史路径的程度，p
n＝w×p
dist，其中p
dist为当前待计算的路径点距离该历史路径最近点的距离，w为权重值，可在计算规划路径的过程中更新，具体可根据实际生成的路径的效果来确定。若本次规划的全局路径与该历史路径重合或几乎重合，表示w过大，可能在环境发生变化的时候离障碍物太近，则需要调小w的值，但是如果w过小，容易发生跳变，则需要调大w的值。
利用优化后的A*算法计算得到全局路径，检测该全局路径是否发生路径跳变，判定全局路径发生跳变之后，在一定时间内不再进行全局规划，并将该历史路径作为本次的全局路径。
上述机器人移动路径规划方法可以应用在机器人上，也可以应用在智能终端上，该智能终端与该机器人接入同一无线网络，将计算得到的全局路径发送给机器人。该智能终端可以是计算机、手机或可穿戴智能设备，通常具有运算器、控制器、存储器、输入设备和输出设备等结构。下面详细描述该机器人移动路径规划方法。
参见图1，本申请一实施例提供的机器人移动路径规划方法的流程示意图。该方法可应用于机器人也可以应用于智能终端，如图1所示，该方法具体包括：
S101、对机器人的全局路径进行非首次规划时，判断规划的历史路径是否出现障碍物，若出现，则将该历史路径中与该障碍物重合的不合法路径点作为本次规划的全局路径的起始点；
非首次规划是指在第一次规划之后的规划。
历史路径是指本次全局路径规划的前一次全局规划得到的全局路径。
历史路径中出现了障碍物，则该历史路径为不合法路径，在该历史路径中与该障碍物重合的路径点为不合法路径点，这些不合法路径点会阻碍机器人在该路径点上的移动，需要机器人避行。
具体地，机器人在进行第一次全局路径规划时，将机器人的当前位置作为规划的起始点。机器人边按照前一次全局路径(即历史路径)移动边再次进行全局路径规划，直到移动到目标点时停止全局路径规划。
在每次全局路径规划后，可能会在规划的移动路径上新增障碍物，使得规划的历史路径成为不合法的路径。在第二次及以后的全局路径时，首先判断前一次规划出的历史路径是否为不合法路径，即在该历史路径上是否存在障碍物，可以通过收集的传感器数据判断该历史路径上是否有障碍物，传感器可以是测距传感器或视觉传感器等。
若历史路径为不合法路径，出现了障碍物，则选择该历史路径上与障碍物重合的一个不合法路径点作为本次规划的全局路径的起始点，将其作为起始点 可以缩短全局规划的路径长度，减少计算量，并且避免因为远处的障碍物观测有噪声而导致全局路径在近处发生跳变，进而影响机器人的移动流畅性。
具体参见图2，图2为起始点到目标点之间存在远端障碍物时规划的移动路径发生跳变的原理示意图，当起始点A到目标点B的方向上，距离A点的远端(例如超过2米)有障碍物，会导致全局规划时移动路径1和移动路径2的代价相差无几，此时容易因为观测噪声等影响而引发全局路径在路径1和路径2之间跳变，这将导致机器人的局部规划的前瞻点在路径1和路径2上跳变，造成机器运行不流畅。此时如果将全局规划的起始点设为历史路径上的一个不合法的路径点，障碍物将成为近端障碍物，在全局规划时不会发生出现两条路径的情况，可避免全局规划发生跳变。
S102、通过包括偏离历史路径代价项的预设的路径算法，确认从本次规划的全局路径的起始点到目标点之间的路径点，使得本次规划的全局路径与该历史路径的偏离度小于预设范围；
偏离历史路径代价项，表示待计算的路径点偏离该历史路径的程度，在该偏离历史路径代价项通过待计算的路径点与本次规划的全局路径的起始点的距离信息，来衡量本次规划的全局路径与该历史路径的偏离程度。
将偏离历史路径代价项加入到路径算法中，可使得本次规划的全局路径与该历史路径的偏离度小于预设范围，即保持本次规划的全局路径靠近该历史路径，避免因本次计算的全局路径与该历史路径偏离太大，导致机器人因此发生大的跳变。
预设范围的数值大小与机器人的任务场景、机器人的任务内容以及机器人自身体积等相关联，例如在该任务场景中障碍物多，需要机器人移动的路径复杂，执行的任务要求机器人移动的路径精度较高，或者机器人体积大，都需要该预设范围的数值更小，提高机器人移动的顺畅度，优选地，该预设范围为0.1～0.2米；反之，该任务场景中障碍物少，需要机器人移动的路径简单，执行的任务要求机器人移动的路径精度较低，或者机器人体积小，该预设范围数值可以大一些，计算精度求低，且不影响机器人移动的顺畅度，优选地，该预设范围为0.3～0.5米。
S103、根据确认的路径点和该目标点得到本次规划的全局路径，控制机器 人沿本次规划的全局路径移动。
步骤S102确认的路径点以及该目标点共同构成本次规划的全局路径，控制机器人从当前位置，沿本次规划的移动路径移动。
按照预设的计算周期，重复执行上述步骤S101～S103，直到机器人抵达该目标点。
本实施例中，对机器人的全局路径进行非首次规划时，判断规划的历史路径是否出现障碍物，若出现，则将该历史路径中与该障碍物重合的不合法路径点作为本次规划的全局路径的起始点，缩短全局规划的路径长度，减少计算量，并且避免因为远处的障碍物观测有噪声而导致全局路径在近处发生跳变，并通过包括偏离历史路径代价项的预设的路径算法，确认从本次规划的全局路径的起始点到目标点之间的路径点，使得本次规划的全局路径与该历史路径的偏离度小于预设范围，避免因本次计算的全局路径与该历史路径偏离太大，导致机器人因此发生大的跳变，通过上述方法得到本次规划的全局路径，可降低因环境变化、传感器噪声影响及障碍物变化等导致的在规划全局路径时存在的频繁跳变，减少对局部规划的影响，提高机器人移动的流畅性。
参见图3，本申请另一实施例提供的机器人移动路径规划方法的流程示意图。该方法可应用于机器人也可以应用于智能终端，如图3所示，该方法具体包括：
S201、对机器人的全局路径进行非首次规划时，判断规划的历史路径是否出现障碍物，若出现，则将该历史路径中与该障碍物重合的第一个不合法路径点作为本次规划的全局路径的起始点；
该第一个不合法路径点是在该历史路径中与该障碍物重合的所有不合法路径点中，距离该机器人最近的不合法路径点。选用该与障碍物重合的第一个不合法点作为本次规划的全局路径的起始点，可以将在第一时间观测到障碍物的位置作为本次规划的全局路径的起始点，最大程度降低因该障碍物的出现对全局路径规划的影响。
S202、通过偏离历史路径代价项的预设的路径算法，确认从本次规划的全局路径的起始点到该目标点之间的路径点；
现有的A*算法的是计算出一条从规划的起始点到目标点代价最小的路径。将机器人从起始点到目标点之间的寻路区域划分为若干栅格，每个栅格可看作 是待计算的路径点，f
n表示该路径点的移动代价，在确定了寻路的起始点和目标点之后，每个待计算的路径点都有一个移动代价值f
n，该移动代价值的计算公式为：f
n＝g
n+h
n，其中，g
n表示从起始点到待计算的路径点n的实际代价，h
n表示从待计算的路径点n到目标点的估算代价。
本实施例中的预设的路径算法为在A*算法基础上优化的算法，具体地，修改移动代价值f
n的计算公式，增加偏离历史路径的代价项p
n，即将f
n的计算公式优化为f
n＝g
n+h
n+p
n，将距离历史路径更近的待计算的路径点确认为规划的路径点，从而选取一条最靠近历史路径的路径，使得本次计算的全局路径尽可能与上一次的历史路径贴合，避免出现大的跳变。
具体地，p
n表示待计算的路径点n偏离该历史路径的程度，p
n＝w×p
dist，其中p
dist为当前待计算的路径点距离该历史路径最近点的距离，w为偏离权重值，可在该机器人进行全局路径规划之前，根据规划得到的路径和实际生成的路径的效果来确定w的值。若规划得到全局路径与该历史路径重合或几乎重合，表示w过大，可能在环境发生变化的时候离障碍物太近，则需要调小w的值，但是如果在进行全局路径规划时容易发生跳变，则表示w过小，需要调大w的值。
在相同或类似的应用场景中，障碍物的类型、大小、位置等信息一致，该w值可在规划路径前一次性设定好，在规划全局路径的过程中该w值为一固定不变的值，该w值为当前场景的最优值，将p
n＝w×p
dist加入算法中，每一次计算得到的全局路径与该全局路径的历史路径的贴合度最优。进一步地，在计算规划路径的过程中因场景变化导致障碍物也变化后，该w的值可随之进行调整。当根据传感器获取的数据判断当前移动场景发生变化，或收到移动场景发生变化的控制指令时，从本地数据库或云端服务器中获取与变化后的场景对应的w值，更新配置到该路径算法中，提高规划全局路径的精度。
上述路径算法可以让本次规划的全局路径尽可能与上一次规划的历史路径贴合，避免出现大的跳变。在图4所示的场景中比较有效：
图4中的三幅图由左至右分别为(a)、(b)和(c)。其中，(a)表示在时刻1的上一次规划的历史路径；由于传感器观测误差或者机器人定位误差等影响，导致(a)中的障碍物在(b)和(c)中障碍物的位置发生了变化，(b)中的实线标识的路径表示在时刻2现有的传统A*算法计算得到的本次规划的全 局路径，虚线标识的路径表示(a)中的历史路径，该两条路径存在比较大的变化；(c)中的实线标识的路径表示在时刻2通过本实施例提供的预设的路径算法计算得到的本次规划的全局路径，虚线标识的路径表示(a)中的历史路径，由于该预设的路径算法考虑了该历史路径代价，所以规划出来的全局路径会比较贴合该历史路径。
进一步地，该预设的路径算法也是设置有开放列表和封闭列表，分别保存可以考虑的待计算的路径点和不再考虑的路径点，每次从开放列表取一个f
n值最小的待计算的路径点，作为寻路路径的下一步，判断该路径点是否为目标点，若是则寻路成功，算法结束；若否则继续寻路。该预设的路径算法的其他细节参照传统的A*算法，此处不再赘述。
S203、根据确认的路径点和该目标点得到本次规划的全局路径，并根据预设的检测规则，检测本次规划的全局路径是否发生路径跳变；
具体地，检测方式包括：计算本次规划的全局路径中的路径点，与历史路径上距离最近的路径点的距离，若在本次规划的全局路径中，距离大于第一预设距离的路径点的比例超过预设比例，则确认本次规划的全局路径发生路径跳变，即本次全局路径上出现了数量过多的路径点，这些路径点都与上一次历史路径上的路径点距离过远，这种情况下可判定本次规划的全局路径发生了路径跳变。该第一预设距离优选0.5米，该预设比例优选50％。
检测方式还包括：当机器人在预设移动距离范围内，在第一预设时长内持续重新规划不合法的全局路径，则确认本次规划的全局路径发生跳变。该预设移动距离范围优选方圆0.5米，该第一预设时长优选为20秒。
若未发生跳变，则执行步骤S204；若发生跳变，则执行步骤S205。
S204、将本次规划的全局路径作为机器人的移动路径，并控制机器人沿本次规划的全局路径移动；
S205、在第二预设时长内停止规划全局路径，将上一次规划的历史路径作为本次规划的全局路径，并控制机器人沿该历史路径移动。
该第二预设时长优选5秒。停止规划全局路径，可以进行局部路径规划，或者可先清除障碍物，再启动全局路径规划。
本实施例中，对机器人的全局路径进行非首次规划时，判断规划的历史路径是否出现障碍物，若出现，则将该历史路径中与该障碍物重合的距离机器人最近的不合法路径点作为本次规划的全局路径的起始点，缩短全局规划的路径长度，减少计算量，并且避免因为远处的障碍物观测有噪声而导致全局路径在近处发生跳变，并通过包括偏离历史路径代价项的预设的路径算法，确认从本次规划的全局路径的起始点到目标点之间的路径点，使得本次规划的全局路径与该历史路径的偏离度小于预设范围，避免因本次计算的全局路径与该历史路径偏离太大，导致机器人因此发生大的跳变，在得到本次规划的全局路径后对其进行跳变检测，通过检测的则确认为本次全局路径，未通过检测的则将上一次计算的历史路径作为本次全局路径，通过上述方法得到本次规划的全局路径，可降低因环境变化、传感器噪声影响及障碍物变化等导致的在规划全局路径时存在的频繁跳变，减少对局部规划的影响，提高机器人移动的流畅性。
参见图5，本申请一实施例提供的机器人移动路径规划装置的结构示意图。为了便于说明，仅示出了与本申请实施例相关的部分。该装置可设置于机器人或智能终端中。该装置包括：
判断模块301，用于对机器人的全局路径进行非首次规划时，判断规划的历史路径是否出现障碍物；
选择模块302，用于若出现障碍物，则选择历史路径中与障碍物重合的不合法路径点作为本次规划的全局路径的起始点；
确认模块303，用于通过偏离历史路径代价项的预设的路径算法，确认从本次规划的全局路径的起始点到目标点之间的路径点，使得本次规划的全局路径与历史路径的偏离度小于预设范围；
控制模块304，用于根据确认的路径点和目标点得到本次规划的全局路径，控制机器人沿本次规划的全局路径移动。
进一步地，选择模块302，还用于选择历史路径中与障碍物重合的第一个不合法路径点，作为本次规划的全局路径的起始点，第一个不合法路径点为距离机器人最近的不合法路径点。
确认模块303，还用于将A*算法进行如下优化后得到预设的路径算法；
将A*算法中的待计算的路径点的移动代价公式设置为：
f
n＝g
n+h
n+p
n；
p
n＝w×p
dist；
其中，g
n表示从起始点到待计算的路径点n的实际代价，h
n表示从待计算的路径点n到目标点的估算代价，p
n表示待计算的路径点n偏离历史路径的偏离历史路径的代价，p
dist为待计算的路径点n距离历史路径最近点的距离，w为预先配置的偏离权重值；
通过上述优化后A*算法，确认从本次规划的全局路径的起始点到目标点之间的路径点。
进一步地，参见图6，在另一个实施例提供的机器人移动路径规划装置中还包括：配置模块401；
配置模块401，用于当检测到机器人的移动场景发生改变，获取与改变后的移动场景对应的偏离权重值，并将偏离权重值更新配置到预设的路径算法中。
该装置还进一步地的包括：检测模块402；
检测模块402，用于根据预设的检测规则，检测本次规划的全局路径是否发生路径跳变；
若未发生路径跳变，则触发控制模块304将本次规划的全局路径作为机器人的移动路径；若发生路径跳变，则触发控制模块304将历史路径作为机器人的移动路径。
检测模块402，还用于计算本次规划的全局路径中的路径点，与历史路径上距离最近的路径点的距离；若在本次规划的全局路径中，距离大于第一预设距离的路径点的比例超过预设比例，则确认本次规划的全局路径发生路径跳变。
检测模块402，还用于当机器人在预设移动距离范围内，在第一预设时长内持续重新规划不合法的全局路径，则确认本次规划的全局路径发生路径跳变。
控制模块304还用于在第二预设时长内停止规划全局路径，并将上一次规划的历史路径作为本次规划的全局路径。
上述图5和图6所示实施例中的技术细节参见前述各实施例的描述，此处不再赘述。
本申请实施例中，对机器人的全局路径进行非首次规划时，判断规划的历史路径是否出现障碍物，若出现，则将该历史路径中与该障碍物重合的距离机器人最近的不合法路径点作为本次规划的全局路径的起始点，缩短全局规划的路径长度，减少计算量，并且避免因为远处的障碍物观测有噪声而导致全局路 径在近处发生跳变，并通过包括偏离历史路径代价项的预设的路径算法，确认从本次规划的全局路径的起始点到目标点之间的路径点，使得本次规划的全局路径与该历史路径的偏离度小于预设范围，避免因本次计算的全局路径与该历史路径偏离太大，导致机器人因此发生大的跳变，在得到本次规划的全局路径后对其进行跳变检测，通过检测的则确认为本次全局路径，未通过检测的则将上一次计算的历史路径作为本次全局路径，通过上述方法得到本次规划的全局路径，可降低因环境变化、传感器噪声影响及障碍物变化等导致的在规划全局路径时存在的频繁跳变，减少对局部规划的影响，提高机器人移动的流畅性。
如图7所示，本申请实施例还提供了一种确定规划的路径点偏离历史路径程度的方法，该方法具体包括：
步骤S701，在本次路径规划时，确定当前待计算的路径点距离历史路径最近点的距离p
dist；
步骤S702，获取与当前移动场景所对应的预先配置的偏离权重值w；
如上文中实施例所描述的，在相同或类似的应用场景中，障碍物的类型、大小、位置等信息一致，该w值可在规划路径前一次性设定好，在规划全局路径的过程中该w值为一固定不变的值，该w值为当前场景的最优值，将p
n＝w×p
dist加入A*算法中，每一次计算得到的全局路径与该全局路径的历史路径的贴合度最优。
进一步地，在计算规划路径的过程中因场景变化导致障碍物也变化后，该w的值可随之进行调整。当根据传感器获取的数据判断当前移动场景发生变化，或收到移动场景发生变化的控制指令时，从本地数据库或云端服务器中获取与变化后的场景对应的w值，更新配置到该路径算法中，提高规划全局路径的精度。
步骤S703，根据距离p
dist与偏离权重值w确定当前待计算的路径点偏离历史路径的程度。
具体地，当前待计算的路径点偏离历史路径的程度p
n根据下述公式计算得到：
p
n＝w×p
dist。
步骤S701-S703的其他细节请参阅上文中的实施例，此处不再赘述。
如图8所示，本申请实施例还提供了一种确定规划的路径点偏离历史路径 程度的装置，该装置具体包括：
最近距离确定模块81，用于在本次路径规划时，确定当前待计算的路径点距离历史路径最近点的距离p
dist；
偏离权重获取模块82，获取与当前移动场景所对应的预先配置的偏离权重值w；
偏离程度确定模块83，用于根据所述距离p
dist与偏离权重值w确定当前待计算的路径点偏离历史路径的程度。
具体地，偏离程度确定模块83根据下述公式计算当前待计算的路径点偏离历史路径的程度p
n：
p
n＝w×p
dist。
本装置的其他细节请参阅上文中的实施例，此处不再赘述。
如图9所示，本申请实施例还提供了一种机器人，包括存储器100和处理器200，处理器200可以是机器人的中央处理器。存储100例如硬盘驱动存储器，非易失性存储器(例如闪存或用于形成固态驱动器的其它电子可编程限制删除的存储器等)，易失性存储器(例如静态或动态随机存取存储器等)等，本申请实施例不作限制。
存储器100存储有可执行程序代码；与存储器100耦合的处理器200，调用所述存储器100中存储的所述可执行程序代码，执行如上所述的机器人移动路径规划方法。
本申请实施例还提供了一种机器人，此机器人同样具有如图7所示的结构，不同的是，当处理器200调用存储器100存储的所述可执行程序代码，执行如上所述的确定规划的路径点偏离历史路径程度的方法。
进一步的，本申请实施例还提供了一种计算机可读存储介质，该计算机可读存储介质可以是设置于上述各实施例中的机器人中，该计算机可读存储介质可以是前述图9所示实施例中的存储器100。该计算机可读存储介质上存储有计算机程序，该程序被处理器执行时实现前述图1和图2所示实施例中描述的机器人移动路径规划方法，或者实现前述如图7实施例中描述的确定规划的路径点偏离历史路径程度的方法。进一步的，该计算机可存储介质还可以是U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、RAM、磁碟或者光盘等 各种可以存储程序代码的介质。
以上所述实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对申请专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。
Claims (20)
- 一种机器人移动路径规划方法，用于规划机器人从起始点到目标点的全局路径，包括：对机器人的全局路径进行非首次规划时，判断规划的历史路径是否出现障碍物，若出现，则选择所述历史路径中与所述障碍物重合的不合法路径点作为本次规划的全局路径的起始点；通过包括偏离历史路径代价项的预设的路径算法，确认从所述本次规划的全局路径的起始点到所述目标点之间的路径点，使得所述本次规划的全局路径与所述历史路径的偏离度小于预设范围；根据确认的路径点和所述目标点得到本次规划的全局路径，控制所述机器人沿所述本次规划的全局路径移动。
- 根据权利要求1所述的方法，其特征在于，所述选择所述历史路径中与所述障碍物重合的不合法路径点作为本次规划的全局路径的起始点包括：选择所述历史路径中与所述障碍物重合的第一个不合法路径点，作为本次规划的全局路径的起始点，所述第一个不合法路径点为距离所述机器人最近的不合法路径点。
- 根据权利要求1所述的方法，其特征在于，所述通过包括偏离历史路径代价项的预设的路径算法，确认从所述本次规划的全局路径的起始点到所述目标点之间的路径点包括：将A*算法进行如下优化后得到所述预设的路径算法：将A*算法中的待计算的路径点的移动代价公式设置为：f n＝g n+h n+p n；p n＝w×p dist；其中，g n表示从所述起始点到待计算的路径点n的实际代价，h n表示从待计算的路径点n到所述目标点的估算代价，p n表示待计算的路径点n偏离所述历史路径的偏离历史路径的代价，p dist为待计算的路径点n距离所述历史路径最近点的距离，w为预先配置的偏离权重值；通过上述优化后的A*算法，确认从所述本次规划的全局路径的起始点到所述目标点之间的路径点。
- 根据权利要求3所述的方法，其特征在于，所述方法还包括：当检测到所述机器人的移动场景发生改变，获取与改变后的移动场景对应的所述偏离权重值，并将所述偏离权重值更新配置到所述预设的路径算法中。
- 根据权利要求4所述的方法，其特征在于，所述根据确认的路径点和所述目标点得到本次规划的全局路径之后还包括：根据预设的检测规则，检测所述本次规划的全局路径是否发生路径跳变；若未发生路径跳变，则将所述本次规划的全局路径作为所述机器人的移动路径；若发生路径跳变，则将所述历史路径作为所述机器人的移动路径。
- 根据权利要求5所述的方法，其特征在于，所述根据预设的检测规则，检测所述本次规划的全局路径是否发生路径跳变包括：计算所述本次规划的全局路径中的路径点，与所述历史路径上距离最近的路径点的距离；若在所述本次规划的全局路径中，所述距离大于第一预设距离的路径点的比例超过预设比例，则确认所述本次规划的全局路径发生路径跳变。
- 根据权利要求6所述的方法，其特征在于，所述根据预设的检测规则，检测所述本次规划的全局路径是否发生路径跳变还包括：当所述机器人在预设移动距离范围内，在第一预设时长内持续重新规划不合法的全局路径，则确认所述本次规划的全局路径发生路径跳变。
- 根据权利要求7所述的方法，其特征在于，所述确认所述本次规划的全局路径发生路径跳变之后包括：在第二预设时长内停止规划全局路径，并将上一次规划的历史路径作为本次规划的全局路径。
- 一种机器人移动路径规划装置，包括：判断模块，用于对机器人的全局路径进行非首次规划时，判断规划的历史 路径是否出现障碍物；选择模块，用于若出现所述障碍物，则选择所述历史路径中与所述障碍物重合的不合法路径点作为本次规划的全局路径的起始点；确认模块，用于通过包括偏离历史路径代价项的预设的路径算法，确认从所述本次规划的全局路径的起始点到目标点之间的路径点，使得所述本次规划的全局路径与所述历史路径的偏离度小于预设范围；控制模块，用于根据确认的路径点和所述目标点得到本次规划的全局路径，控制所述机器人沿所述本次规划的全局路径移动。
- 一种确定规划的路径点偏离历史路径程度的方法，包括：在本次路径规划时，确定当前待计算的路径点距离历史路径最近点的距离p dist；获取与当前移动场景所对应的预先配置的偏离权重值w；根据所述距离p dist与偏离权重值w确定当前待计算的路径点偏离历史路径的程度。
- 根据权利要求10所述的方法，其特征在于，所述获取与当前移动场景所对应的预先配置的偏离权重值w包括：当判断出当前移动场景发生变化时，从本地数据库或云端服务器中获取与变化后的场景对应的偏离权重值w。
- 根据权利要求10所述的方法，其特征在于，所述获取与当前移动场景所对应的预先配置的偏离权重值w包括：当收到移动场景发生变化的控制指令时，从本地数据库或云端服务器中获取与变化后的场景对应的偏离权重值w。
- 根据权利要求10所述的方法，其特征在于，所述根据所述距离p dist与偏离权重值w确定当前待计算的路径点偏离历史路径的程度，包括：当前待计算的路径点偏离历史路径的程度p n根据下述公式计算得到：p n＝w×p dist。
- 一种确定规划的路径点偏离历史路径程度的装置，包括：最近距离确定模块，用于在本次路径规划时，确定当前待计算的路径点距离历史路径最近点的距离p dist；偏离权重获取模块，获取与当前移动场景所对应的预先配置的偏离权重值w；偏离程度确定模块，用于根据所述距离p dist与偏离权重值w确定当前待计算的路径点偏离历史路径的程度。
- 一种机器人，包括：存储器和处理器；所述存储器存储有可执行程序代码；与所述存储器耦合的所述处理器，调用所述存储器中存储的所述可执行程序代码，执行如权利要求1所述的机器人移动路径规划方法。
- 根据权利要求15所述的机器人，其特征在于，所述选择所述历史路径中与所述障碍物重合的不合法路径点作为本次规划的全局路径的起始点包括：选择所述历史路径中与所述障碍物重合的第一个不合法路径点，作为本次规划的全局路径的起始点，所述第一个不合法路径点为距离所述机器人最近的不合法路径点。
- 根据权利要求15所述的机器人，其特征在于，所述通过包括偏离历史路径代价项的预设的路径算法，确认从所述本次规划的全局路径的起始点到所述目标点之间的路径点包括：将A*算法进行如下优化后得到所述预设的路径算法：将A*算法中的待计算的路径点的移动代价公式设置为：f n＝g n+h n+p n；p n＝w×p dist；其中，g n表示从所述起始点到待计算的路径点n的实际代价，h n表示从待计算的路径点n到所述目标点的估算代价，p n表示待计算的路径点n偏离所述历史路径的偏离历史路径的代价，p dist为待计算的路径点n距离所述历史路径最近点的距离，w为预先配置的偏离权重值；通过上述优化后的A*算法，确认从所述本次规划的全局路径的起始点到所 述目标点之间的路径点。
- 一种机器人，包括：存储器和处理器；所述存储器存储有可执行程序代码；与所述存储器耦合的所述处理器，调用所述存储器中存储的所述可执行程序代码，执行如权利要求10所述的确定规划的路径点偏离历史路径程度的方法。
- 根据权利要求18所述的机器人，其特征在于，所述根据所述距离p dist与偏离权重值w确定当前待计算的路径点偏离历史路径的程度，包括：当前待计算的路径点偏离历史路径的程度p n根据下述公式计算得到：p n＝w×p dist。
- 一种计算机可读存储介质，所述计算机可读存储介质存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1所述方法的步骤。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011581919.6 | 2020-12-28 | ||
| CN202011581919.6A CN112631306B (zh) | 2020-12-28 | 2020-12-28 | 机器人移动路径规划方法、装置及机器人 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| WO2022142858A1 WO2022142858A1 (zh) | 2022-07-07 |
| WO2022142858A9 true WO2022142858A9 (zh) | 2022-10-13 |
Family
ID=75325617
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2021/132198 WO2022142858A1 (zh) | 2020-12-28 | 2021-11-22 | 机器人移动路径规划方法、确定规划的路径点偏离历史路径程度的方法、装置、机器人及计算机可读存储介质 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112631306B (zh) |
| WO (1) | WO2022142858A1 (zh) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112631306B (zh) * | 2020-12-28 | 2021-12-14 | 深圳市普渡科技有限公司 | 机器人移动路径规划方法、装置及机器人 |
| CN114228708B (zh) * | 2021-12-31 | 2023-08-25 | 上海仙途智能科技有限公司 | 车辆控制方法和系统 |
| CN115922688A (zh) * | 2022-06-21 | 2023-04-07 | 北京小米移动软件有限公司 | 机械手的操作方法、装置及存储介质 |
| CN115016546B (zh) * | 2022-08-10 | 2022-10-28 | 中国科学院自动化研究所 | 无人机三维路径规划方法、装置、电子设备和存储介质 |
| CN115326058B (zh) * | 2022-10-17 | 2023-02-07 | 杭州华橙软件技术有限公司 | 机器人地图的生成方法、设备、存储介质和移动机器人 |
| CN116151036B (zh) * | 2023-04-17 | 2023-07-07 | 中铁九局集团有限公司 | 一种钢筋自动绑扎的路径规划方法及装置 |
Family Cites Families (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR200277538Y1 (ko) * | 1999-04-02 | 2002-06-14 | 김성국 | 케이블 선통장치용 송풍어댑터 |
| DE102011000250A1 (de) * | 2011-01-21 | 2012-07-26 | Vorwerk & Co. Interholding Gmbh | Verfahren zur Bestimmung der Position eines selbsttätig verfahrbaren Geräts |
| DE102011015775A1 (de) * | 2011-04-01 | 2012-10-04 | Volkswagen Aktiengesellschaft | Verfahren und Vorrichtung zum Durchführen einer Reiseroutenplanung für ein Fahrzeug |
| KR101350224B1 (ko) * | 2011-12-07 | 2014-01-14 | 에스케이플래닛 주식회사 | 차량 경로 탐색 방법 및 장치 |
| CN103576686B (zh) * | 2013-11-21 | 2017-01-18 | 中国科学技术大学 | 一种机器人自主导引及避障的方法 |
| JP2018124103A (ja) * | 2017-01-31 | 2018-08-09 | パイオニア株式会社 | 情報処理装置 |
| CN106774347A (zh) * | 2017-02-24 | 2017-05-31 | 安科智慧城市技术（中国）有限公司 | 室内动态环境下的机器人路径规划方法、装置和机器人 |
| JP6901331B2 (ja) * | 2017-06-20 | 2021-07-14 | 株式会社東芝 | 情報処理装置、移動体、情報処理方法、およびプログラム |
| CN107345815A (zh) * | 2017-07-24 | 2017-11-14 | 东北大学 | 一种基于改进a*算法的家庭服务机器人路径规划方法 |
| CN108040319B (zh) * | 2017-11-29 | 2020-05-22 | 新华三技术有限公司 | 一种终端历史轨迹的确定方法及装置 |
| CN109976148B (zh) * | 2017-12-28 | 2022-02-22 | 深圳市优必选科技有限公司 | 机器人运动路径规划方法、装置、存储介质及终端设备 |
| JP6927090B2 (ja) * | 2018-03-05 | 2021-08-25 | トヨタ自動車株式会社 | 経路生成装置 |
| KR102552814B1 (ko) * | 2018-06-29 | 2023-07-06 | 현대오토에버 주식회사 | 차량용 경로 탐색과 안내 장치 및 방법 |
| CN114879695B (zh) * | 2019-05-15 | 2023-03-24 | 百度在线网络技术（北京）有限公司 | 轨迹匹配方法、装置、设备和介质 |
| CN110426056A (zh) * | 2019-07-30 | 2019-11-08 | 深圳市普渡科技有限公司 | 路径规划方法 |
| CN110632921B (zh) * | 2019-09-05 | 2022-11-18 | 北京百度网讯科技有限公司 | 机器人路径规划方法、装置、电子设备和存储介质 |
| CN111177934B (zh) * | 2019-12-31 | 2021-06-22 | 福瑞泰克智能系统有限公司 | 参考路径规划的方法、设备和存储介质 |
| CN111208838B (zh) * | 2020-04-20 | 2020-11-03 | 北京三快在线科技有限公司 | 一种无人驾驶设备的控制方法及装置 |
| CN111582566B (zh) * | 2020-04-26 | 2024-01-26 | 上海高仙自动化科技发展有限公司 | 路径规划方法及规划装置、智能机器人及存储介质 |
| CN111504325B (zh) * | 2020-04-29 | 2023-09-26 | 南京大学 | 一种基于扩大搜索邻域的加权a*算法的全局路径规划方法 |
| CN111938513B (zh) * | 2020-06-30 | 2021-11-09 | 珠海市一微半导体有限公司 | 一种机器人越障的沿边路径选择方法、芯片及机器人 |
| CN112631306B (zh) * | 2020-12-28 | 2021-12-14 | 深圳市普渡科技有限公司 | 机器人移动路径规划方法、装置及机器人 |
-
2020
- 2020-12-28 CN CN202011581919.6A patent/CN112631306B/zh active Active
-
2021
- 2021-11-22 WO PCT/CN2021/132198 patent/WO2022142858A1/zh unknown
Also Published As
| Publication number | Publication date |
|---|---|
| CN112631306B (zh) | 2021-12-14 |
| WO2022142858A1 (zh) | 2022-07-07 |
| CN112631306A (zh) | 2021-04-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022142858A9 (zh) | 机器人移动路径规划方法、确定规划的路径点偏离历史路径程度的方法、装置、机器人及计算机可读存储介质 | |
| US11878427B2 (en) | Robot navigation using 2D and 3D path planning | |
| US20210103286A1 (en) | Systems and methods for adaptive path planning | |
| WO2020187134A1 (zh) | 移动机器人的返回充电方法、装置、移动机器人、系统及存储介质 | |
| US8195331B2 (en) | Method, medium, and apparatus for performing path planning of mobile robot | |
| US20170004406A1 (en) | Parallel belief space motion planner | |
| Zhu et al. | A new hybrid navigation algorithm for mobile robots in environments with incomplete knowledge | |
| KR101585504B1 (ko) | 자율 이동 차량의 경로 생성 방법 및 경로 생성 장치 | |
| CN112082554A (zh) | 机器人导航方法、装置、终端设备及存储介质 | |
| CN107843252B (zh) | 导航路径优化方法、装置及电子设备 | |
| JP5614202B2 (ja) | ロボット | |
| CN113534818B (zh) | 路径导航规划方法、装置、存储介质及电子设备 | |
| CN110702117B (zh) | 基于地图的路径规划方法、终端设备及计算机存储介质 | |
| CN111123934A (zh) | 轨迹评估方法、轨迹评估装置及移动机器人 | |
| CN111090284A (zh) | 自行走设备返回基站的方法及自行走设备 | |
| CN109933072A (zh) | 机器人回座时检测到充电座护栏信号的控制方法 | |
| KR101965044B1 (ko) | 대상체를 추종하는 방법 및 디바이스 | |
| JP7058761B2 (ja) | 移動体制御装置、移動体制御学習装置、及び移動体制御方法 | |
| KR102183830B1 (ko) | 무인 자율주행차량의 경로제어장치 및 경로제어방법 | |
| WO2019049664A1 (ja) | 自走装置、自走装置の走行制御方法及び走行制御プログラム | |
| Burgard et al. | Introduction to mobile robotics | |
| CN114578821A (zh) | 移动机器人的脱困方法、移动机器人及存储介质 | |
| Luo et al. | Autonomous mobile robot localization based on multisensor fusion approach | |
| CN110196054B (zh) | 一种导航方法及系统 | |
| JP2018120482A (ja) | ロボットおよびその制御方法 |
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: 21913600 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 02.11.2023) |