WO2021042519A1 - Procédé de planification de trajet d'assemblage et dispositif associé - Google Patents

Procédé de planification de trajet d'assemblage et dispositif associé Download PDF

Info

Publication number
WO2021042519A1
WO2021042519A1 PCT/CN2019/117030 CN2019117030W WO2021042519A1 WO 2021042519 A1 WO2021042519 A1 WO 2021042519A1 CN 2019117030 W CN2019117030 W CN 2019117030W WO 2021042519 A1 WO2021042519 A1 WO 2021042519A1
Authority
WO
WIPO (PCT)
Prior art keywords
sampling
path
target
center
new
Prior art date
Application number
PCT/CN2019/117030
Other languages
English (en)
Chinese (zh)
Inventor
王义文
王健宗
Original Assignee
平安科技（深圳）有限公司
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 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2021042519A1 publication Critical patent/WO2021042519A1/fr

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
    • G01C21/3407—Route searching; Route guidance specially adapted for specific applications
    • G01C21/343—Calculating itineraries, i.e. routes leading from a starting point to a series of categorical destinations using a global route restraint, round trips, touristic trips

Definitions

  • This application relates to the field of path planning, and specifically relates to an assembly path planning method and related devices.
  • RRT algorithm is a fast random path planning method. Because it uses random sampling planning method, it does not need scene preprocessing, so it can solve the problem of multi-dimensional non-convex space. The path planning problem is widely used to solve the optimal path in the static road network problem.
  • the traditional RRT algorithm adopts a global uniform random sampling strategy, the calculation amount of the algorithm will increase sharply and the efficiency of the algorithm will be reduced; and it cannot get rid of the trouble of local optimal solution, so that the RRT algorithm only has a probability optimal solution.
  • the embodiments of the present application provide an assembly path planning method and related devices, which can accelerate the sampling convergence speed and solve the problem of being easily trapped in a local optimum during the path planning process.
  • the first aspect of the embodiments of the present application discloses an assembly path planning method, the method includes:
  • a second aspect of the present application discloses an assembly path planning device, the assembly path planning device includes:
  • the sampling unit is used to select a sampling area that includes the starting node, and generate T new pose nodes by random numbers in the sampling area, where T is a positive integer, and select the first from the T new pose nodes
  • a sampling center selecting a new sampling area that includes the first sampling center, and repeating the method of obtaining the first sampling center until a preset condition is met, and n second sampling centers are obtained, and the preset condition includes
  • the path obtaining unit is configured to connect the starting node, the first sampling center and the n second sampling centers to obtain a first path, and obtain a target path according to the first path.
  • the third aspect of the present application discloses an electronic device, including a processor, a memory, a communication interface, and one or more programs.
  • the one or more programs are stored in the memory and configured to be processed by the processor.
  • the program is executed by a device, and the program includes a method for executing any one of the first aspect.
  • the fourth aspect of the present application discloses a computer non-volatile readable storage medium, the computer storage medium stores a computer program, and the computer program is executed by a processor to implement any one of the first aspect method.
  • the sampling area including the starting node is selected; T new pose nodes are generated by random numbers in the sampling area; the first sampling center is selected from the T new pose nodes, Select a new sampling area that includes the first sampling center; repeat the method of obtaining the first sampling center until a preset condition is met, and n second sampling centers are obtained; connect the start node and the first sampling center The sampling center and the n second sampling centers obtain the first path, and the target path is obtained according to the first path; this assembly path planning method can greatly accelerate the sampling convergence speed and solve the problem that the path planning of the traditional RRT algorithm tends to be trapped locally Optimal problem.
  • FIG. 1 is a schematic flowchart of an assembly path planning method provided by an embodiment of the application
  • Figure 2 is a schematic diagram of partial sampling provided by an embodiment of the application.
  • FIG. 3 is a schematic diagram of an expansion strategy based on accurate collision calculation provided by an embodiment of the application
  • FIG. 4 is a schematic structural diagram of an electronic device provided by an embodiment of this application.
  • Fig. 5 is a schematic structural diagram of an assembly path planning device provided by an embodiment of the application.
  • the electronic devices involved in the embodiments of the present application may include various handheld devices with wireless communication functions, in-vehicle devices, wireless headsets, computing devices or other processing devices connected to wireless modems, as well as various forms of user equipment (user equipment). , UE), mobile station (mobile station, MS), terminal device (terminal device), etc.
  • the electronic device may be, for example, a smart phone, a tablet computer, a headset box, and so on.
  • the devices mentioned above are collectively referred to as electronic devices.
  • FIG. 1 is a schematic flowchart of an assembly path planning method provided by an embodiment of the application, including:
  • S101 Select a sampling area that includes a starting node.
  • this application improves the RRT algorithm and proposes an assembly path planning method based on the improved RRT algorithm, which can solve the problem of traditional RRT Algorithm problem.
  • Knowing a feasible solution space containing the starting node q s and the target node q e (that is, the pre-selected end point), it is necessary to find the closest feasible path from the starting node to the target node in the feasible solution space to
  • the starting node is the sampling center, and an area containing the starting node is selected as the sampling area in the feasible solution space for local sampling, so that a significant local sampling effect can be achieved without falling into the local optimal solution.
  • S102 Generate T new pose nodes through random numbers in the sampling area, where T is a positive integer.
  • T different new pose nodes can be obtained to form an extended random tree, and T can be predetermined Set, for example, 5, 10, 15, 20 and so on.
  • S103 Select a first sampling center from the T new pose nodes, and select a new sampling area that includes the first sampling center.
  • Figure 2 is a schematic diagram of local sampling.
  • the local sampling strategy is adopted to determine the next sampling area through the current sampling center.
  • the current sampling center and the next The distance between the sampling centers therefore, the node q near that is closest to the current sampling center is selected from the T new pose nodes as the first sampling center, and the first sampling center is selected according to the method described in S101 Of the new sampling area.
  • the method of acquiring the sampling center and the sampling area is the same each time.
  • the sampling area is determined according to the current sampling center, the next sampling center is determined in the sampling area, and the method of obtaining the first sampling center is repeated, until Reach the preset maximum sampling times or reach the target node.
  • the first path is obtained by connecting all sampling centers including the start node and the target node.
  • the traditional RRT algorithm uses a global uniform random sampling strategy, the algorithm calculation is huge, the algorithm efficiency is low, and it cannot get rid of the local maximum.
  • the problem of the optimal solution is that only the optimal solution can be obtained.
  • by combining random sampling, partial sampling and end-point sampling the success rate of sampling center expansion of the expanded random tree during iterative sampling is improved, and the success rate of sampling center expansion is improved.
  • the convergence speed of the traditional RRT algorithm is solved, and the problem of local optimal solution is solved.
  • the possible paths are obtained through random sampling, local sampling and end-point sampling at the same time, and the first path, the second path and the third path are obtained. The above three paths are fused to obtain the final target path.
  • the assembly path planning method is applied to a six-degree-of-freedom intelligent robot.
  • the six-degree-of-freedom intelligent robot is a core component that relieves industrial production pressure and improves industrial production efficiency. , Forearm, wrist arm and terminal controller, with compact structure, complex parts and other related characteristics.
  • the embodiment of this application uses a six-degree-of-freedom robot as a carrier to verify whether the improved RRT algorithm can successfully complete the path planning task in a virtual environment through simulation.
  • the simulation result shows that the improved RRT algorithm can be successfully used for complex assemblies (six-degrees of freedom robot).
  • the optimal assembly path of the assembly is obtained.
  • the successful completion of the six-degree-of-freedom robot assembly path planning proves that the improved RRT algorithm is scientific and effective.
  • the sampling area containing the starting node is selected; T new pose nodes are generated by random numbers in the sampling area; the first one is selected from the T new pose nodes.
  • Sampling center select a new sampling area that includes the first sampling center; repeat the method of obtaining the first sampling center until a preset condition is met, and n second sampling centers are obtained; connect the start node and the The first sampling center and the n second sampling centers obtain the first path, and obtain the target path according to the first path; this assembly path planning method can greatly accelerate the sampling convergence speed and solve the traditional RRT algorithm path planning It is easy to fall into the problem of local optimum.
  • the generating T new pose nodes by random numbers in the sampling area is obtained by the following formula:
  • T new pose nodes Repeat the formula T times to obtain T new pose nodes, and record T new pose nodes in the form of a list, where f z is a random number, f z ⁇ (-1, 0) ⁇ (0, 1), f y is the range factor, f y ⁇ (0, 1), q s is the starting node, q e is the target node, q d is the current sampling center, and q rand is the new pose node.
  • Each node parameter in the above formula is a six-dimensional variable, and the six-dimensional includes the X, Y, Z three-dimensional and posture angle coordinate systems in the rectangular space coordinate system
  • Yaw is the yaw angle, which rotates around the Z axis
  • Pitch is the pitch angle
  • Y axis the roll angle
  • the new pose node q rand is at The possible positions in the spatial rectangular coordinate system consist of the point O1 (x1-0.4x2, y1-0.4y2, z1-0.4z2) and the point O2 (x1+0.4x2, y1+0.4y2, z1+0.4z2) as the body pair Inside the rectangular parallelepiped formed by the corners, the rectangular parallelepiped is the sampling area. Obviously, q d is at the center of the rectangular parallelepiped.
  • the posture information (orientation) of the T new posture nodes in the cuboid can be determined according to the coordinates of q d and q s -q e in the posture angle coordinate system.
  • the method further includes:
  • the next sampling center is determined according to the sampling direction and the sampling step length.
  • the obtaining a new sampling direction according to the target node, the first sampling center and the collision point includes:
  • Figure 3 is a schematic diagram of the expansion strategy based on accurate collision calculation, where q s is the starting node and q e is the target node.
  • q rand is the current sampling point
  • q is obtained by calculation.
  • Near is the point closest to q rand in the expansion tree, which is used as the point to be expanded in the next step.
  • the method further includes:
  • the third path is obtained by sampling at the end point.
  • random sampling is to randomly select sampling points in the feasible solution space, which is the basic sampling strategy of the traditional RRT algorithm.
  • End-point sampling is to sample the target node as the sampling point in the feasible solution space.
  • the end-point sampling strategy can expand the random tree Converge quickly to the target node.
  • the obtaining the target path according to the first path includes:
  • performing mathematical processing on the first path, the second path, and the third path to obtain the target path includes: selecting k equidistant abscissas on the abscissa of the planar rectangular coordinate system;
  • k target ordinates corresponding to the k abscissas are obtained, and according to the k target ordinates
  • the coordinates obtain k target sampling centers, and the k target sampling centers are connected to obtain the target path.
  • the final target path can be obtained by averaging the weighting of the first path, the second path, and the third path.
  • k equidistant abscissas ( k1, k2,,, kn)
  • the target path can also be obtained by processing the above three paths in other ways, which is not limited in this application.
  • FIG. 4 is a schematic structural diagram of an electronic device provided by an embodiment of the application. As shown in the figure, it includes a processor, a memory, a communication interface, and one or more programs. In the memory, and configured to be executed by the processor.
  • the program includes instructions for executing the following steps:
  • the program includes instructions for executing the following steps:
  • the next sampling center is determined according to the sampling direction and the sampling step length.
  • the program includes instructions for executing the following steps:
  • the program includes instructions for executing the following steps:
  • the third path is obtained by sampling at the end point.
  • the program includes instructions for executing the following steps:
  • the program includes instructions for executing the following steps:
  • k target ordinates corresponding to the k abscissas are obtained, and according to the k target ordinates
  • the coordinates obtain k target sampling centers, and the k target sampling centers are connected to obtain the target path.
  • the terminal includes hardware structures and/or software modules corresponding to each function.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software-driven hardware depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.
  • the embodiments of the present application may divide the terminal into functional units according to the foregoing method examples.
  • each functional unit may be divided corresponding to each function, or two or more functions may be integrated into one processing unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit. It should be noted that the division of units in the embodiments of the present application is illustrative, and is only a logical function division, and there may be other division methods in actual implementation.
  • Fig. 5 is a schematic structural diagram of an assembly path planning device 500 provided by an embodiment of the application.
  • the assembly route planning device 500 includes a sampling unit 501 and a route acquisition unit 502, wherein:
  • the path obtaining unit 502 is configured to connect the start node, the first sampling center, and the n second sampling centers to obtain a first path, and obtain a target path according to the first path.
  • the sampling unit 501 generates T new pose nodes in the sampling area through the following formula:
  • T new pose nodes where f z is a random number, f z ⁇ (-1,0) ⁇ (0,1), f y is a range factor, and f y ⁇ ( 0, 1), q s is the starting node, q e is the target node, q d is the current sampling center, and q rand is the new pose node.
  • the assembly path planning device further includes an obstacle avoidance unit 503 for confirming whether there is a collision point on the line between the starting node and the first sampling center, and if the collision point exists, according to The target node, the first sampling center, and the collision point obtain a new sampling direction, determine a sampling step, and determine the next sampling center according to the sampling direction and the sampling step.
  • an obstacle avoidance unit 503 for confirming whether there is a collision point on the line between the starting node and the first sampling center, and if the collision point exists, according to The target node, the first sampling center, and the collision point obtain a new sampling direction, determine a sampling step, and determine the next sampling center according to the sampling direction and the sampling step.
  • the obstacle avoidance unit 503 is specifically configured to:
  • the path obtaining unit 502 is specifically configured to:
  • the third path is obtained by sampling at the end point.
  • the path obtaining unit 502 is specifically configured to:
  • the path obtaining unit 502 is specifically configured to:
  • k target ordinates corresponding to the k abscissas are obtained, and according to the k target ordinates
  • the coordinates obtain k target sampling centers, and the k target sampling centers are connected to obtain the target path.
  • the above-mentioned unit may be used to execute the method described in the above-mentioned embodiment.
  • a sampling area containing the starting node is selected; T new pose nodes are generated by random numbers in the sampling area; the first sampling center is selected from the T new pose nodes, and A new sampling area including the first sampling center; repeat the method of obtaining the first sampling center until a preset condition is met, and n second sampling centers are obtained; connect the start node and the first sampling The center and the n second sampling centers obtain the first path, and the target path is obtained according to the first path; through this assembly path planning method, the sampling convergence speed can be greatly accelerated, and the path planning of the traditional RRT algorithm is easily trapped in the local maximum. Excellent question.
  • the embodiment of the present application also provides a computer non-volatile readable storage medium that stores a computer program for electronic data exchange.
  • the computer program enables the computer to execute any of the assembly path planning methods described in the above method embodiments. Part or all of the steps.
  • the embodiments of the present application also provide a computer program product.
  • the computer program product includes a non-transitory computer non-volatile readable storage medium storing a computer program.
  • the computer program causes a computer to execute the method described in the foregoing method embodiment. Part or all of the steps of any assembly path planning method.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Complex Calculations (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

La présente invention concerne un procédé de planification de trajet d'assemblage et un dispositif associé, le procédé comprenant : la sélection d'une région d'échantillonnage incluant un nœud de départ (101) ; la génération de T nouveaux nœuds de pose dans la région d'échantillonnage au moyen de nombres aléatoires (102) ; la sélection d'un premier centre d'échantillonnage parmi les T nouveaux nœuds de pose, et la sélection d'une nouvelle région d'échantillonnage incluant le premier centre d'échantillonnage (103) ; la répétition de l'étape d'acquisition du premier centre d'échantillonnage jusqu'à ce qu'une condition prédéfinie soit satisfaite, de manière à obtenir n deuxièmes centres d'échantillonnage (104) ; et la connexion du nœud de départ, du premier centre d'échantillonnage et des n deuxièmes centres d'échantillonnage de façon à obtenir un premier trajet, et l'obtention d'un trajet cible en fonction du premier trajet (105). Le procédé de planification de trajet d'assemblage accélère considérablement le taux de convergence d'échantillonnage, et résout le problème selon lequel une planification de trajet classique à l'aide d'un algorithme d'arbre aléatoire à exploration rapide (RRT) peut facilement tomber dans le piège de solutions localement optimales.
PCT/CN2019/117030 2019-09-02 2019-11-11 Procédé de planification de trajet d'assemblage et dispositif associé WO2021042519A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910824569.2A CN110646006A (zh) 2019-09-02 2019-09-02 装配路径规划方法及相关装置
CN201910824569.2 2019-09-02

Publications (1)

Publication Number Publication Date
WO2021042519A1 true WO2021042519A1 (fr) 2021-03-11

Family

ID=69009998

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/117030 WO2021042519A1 (fr) 2019-09-02 2019-11-11 Procédé de planification de trajet d'assemblage et dispositif associé

Country Status (2)

Country Link
CN (1) CN110646006A (fr)
WO (1) WO2021042519A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113985884A (zh) * 2021-10-29 2022-01-28 国网山东省电力公司经济技术研究院 一种电力巡检机器人路径规划方法、系统和机器人

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111220786A (zh) * 2020-03-09 2020-06-02 生态环境部华南环境科学研究所 一种深水沉积物有机污染快速监测方法
CN112344938B (zh) * 2020-10-31 2022-07-19 安徽中科源起科技有限公司 基于指向和势场参数的空间环境路径生成及规划方法
CN112987799B (zh) * 2021-04-16 2022-04-05 电子科技大学 一种基于改进rrt算法的无人机路径规划方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102375416A (zh) * 2010-08-13 2012-03-14 同济大学 基于快速搜索树的人型机器人踢球动作信息处理方法
CN102708242A (zh) * 2012-05-07 2012-10-03 上海飞机制造有限公司 产品管路件的拆卸路径求解方法及装置
CN106695802A (zh) * 2017-03-19 2017-05-24 北京工业大学 基于多自由度机械臂的改进式rrt*的避障运动规划方法
US20170241790A1 (en) * 2016-02-24 2017-08-24 Honda Motor Co., Ltd. Path plan generating apparatus for mobile body
CN110083165A (zh) * 2019-05-21 2019-08-02 大连大学 一种机器人在复杂狭窄环境下路径规划方法

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108983780A (zh) * 2018-07-24 2018-12-11 武汉理工大学 一种基于改进rrt*算法的移动机器人路径规划方法
CN109886279B (zh) * 2019-01-24 2023-09-29 平安科技（深圳）有限公司 图像处理方法、装置、计算机设备及存储介质
CN109877836B (zh) * 2019-03-13 2021-06-08 浙江大华技术股份有限公司 路径规划方法、装置、机械臂控制器和可读存储介质
CN110181515B (zh) * 2019-06-10 2020-12-01 浙江工业大学 一种双机械臂协同装配作业路径规划方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102375416A (zh) * 2010-08-13 2012-03-14 同济大学 基于快速搜索树的人型机器人踢球动作信息处理方法
CN102708242A (zh) * 2012-05-07 2012-10-03 上海飞机制造有限公司 产品管路件的拆卸路径求解方法及装置
US20170241790A1 (en) * 2016-02-24 2017-08-24 Honda Motor Co., Ltd. Path plan generating apparatus for mobile body
CN106695802A (zh) * 2017-03-19 2017-05-24 北京工业大学 基于多自由度机械臂的改进式rrt*的避障运动规划方法
CN110083165A (zh) * 2019-05-21 2019-08-02 大连大学 一种机器人在复杂狭窄环境下路径规划方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113985884A (zh) * 2021-10-29 2022-01-28 国网山东省电力公司经济技术研究院 一种电力巡检机器人路径规划方法、系统和机器人

Also Published As

Publication number Publication date
CN110646006A (zh) 2020-01-03

Similar Documents

Publication Publication Date Title
WO2021042519A1 (fr) Procédé de planification de trajet d'assemblage et dispositif associé
TWI748409B (zh) 資料處理方法、處理器、電子設備和電腦可讀媒體
CN110751039B (zh) 多视图3d人体姿态估计方法及相关装置
CN112784739B (zh) 模型的训练方法、关键点定位方法、装置、设备和介质
CN107563295B (zh) 基于多Kinect的全方位人体追踪方法及处理设备
CN110991526A (zh) 一种非迭代式点云匹配方法、介质、终端和装置
WO2024011792A1 (fr) Procédé et appareil de traitement d'image, dispositif électronique et support d'enregistrement
CN114202632A (zh) 网格线性结构恢复方法、装置、电子设备及存储介质
WO2024139520A1 (fr) Procédé et appareil de traitement de vidéo, dispositif, support de stockage et produit
CN113850904A (zh) 发丝模型的确定方法、装置、电子设备及可读存储介质
CN110082713A (zh) 定位方法、终端设备及存储介质
CN118067127A (zh) 采摘机器人的路径规划方法、系统、存储介质及设备
CN109844774A (zh) 一种并行反卷积计算方法、单引擎计算方法及相关产品
JP2021122736A5 (fr)
CN110414663A (zh) 神经网络的卷积实现方法及相关产品
CN104346797A (zh) 关键像素点匹配方法和装置、图像匹配方法和装置
CN109523634A (zh) 优化网格的方法及装置
CN112884869A (zh) 运动重定向策略选择方法、装置、设备及存储介质
CN113218399A (zh) 一种基于多智能体分层强化学习的迷宫导航方法及装置
Matsushita et al. Behavior of independent-minded particle swarm optimization
CN118060981B (zh) 曲面打磨轨迹规划方法、装置及电子设备
CN116901073B (zh) 多机械臂协同轨迹规划方法、装置、电子设备及存储介质
JP6492779B2 (ja) マッピング情報生成プログラム、マッピング情報生成方法、及びマッピング情報生成装置
Ni et al. An optimized POSIT algorithm based on mean convergence
CN111144212B (zh) 一种深度图像目标分割方法及装置

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: 19944378

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19944378

Country of ref document: EP

Kind code of ref document: A1