CN112847358B - Path planning method and industrial robot - Google Patents

Path planning method and industrial robot Download PDF

Info

Publication number
CN112847358B
CN112847358B CN202011640924.XA CN202011640924A CN112847358B CN 112847358 B CN112847358 B CN 112847358B CN 202011640924 A CN202011640924 A CN 202011640924A CN 112847358 B CN112847358 B CN 112847358B
Authority
CN
China
Prior art keywords
path planning
industrial robot
control point
weight matrix
category
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.)
Active
Application number
CN202011640924.XA
Other languages
Chinese (zh)
Other versions
CN112847358A (en
Inventor
陈先开
李攀
张焱
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.)
Shenzhen Cosmosvision Intelligent Technology Co ltd
Original Assignee
Shenzhen Cosmosvision Intelligent 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 Shenzhen Cosmosvision Intelligent Technology Co ltd filed Critical Shenzhen Cosmosvision Intelligent Technology Co ltd
Priority to CN202011640924.XA priority Critical patent/CN112847358B/en
Publication of CN112847358A publication Critical patent/CN112847358A/en
Application granted granted Critical
Publication of CN112847358B publication Critical patent/CN112847358B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
    • B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning

Landscapes

  • Engineering & Computer Science (AREA)
  • Robotics (AREA)
  • Mechanical Engineering (AREA)
  • Manipulator (AREA)

Abstract

The invention provides a path planning method and an industrial robot, wherein the path planning method comprises the following steps: obtaining a control point sample set; performing model training on the control point sample set to obtain a path planning model; receiving a plurality of capturing point data of the industrial robot input by a user; inputting all the grabbing point data into a path planning model; obtaining the category corresponding to each captured point data output by the path planning model; and outputting the feasible path according to the capture point data of which the category is the non-collision control point. According to the path planning method, model training is carried out by acquiring the conditions of a plurality of historical grabbing points of the industrial robot, and mechanical limitation and mistaken grabbing of the industrial robot during grabbing are avoided. In addition, model training provides an obstacle avoidance algorithm based on the SVM algorithm model, a mechanical arm path planning algorithm is accurately set, redundant calculation in the traditional path planning process is avoided, and the obstacle avoidance and path planning efficiency of the industrial robot is effectively improved.

Description

Path planning method and industrial robot
Technical Field
The invention belongs to the technical field of intelligent control, and particularly relates to a path planning method and an industrial robot.
Background
The path planning method of the industrial robot at present comprises the steps of firstly calculating the coordinates of a barrier relative to the industrial robot, then determining the angle of each axis of the industrial robot by inverse kinematics, finally planning the rotation of the axis of the industrial robot by a function interpolation method,
however, the method needs to perform multiple calculations and multiple planning, and is low in efficiency, and the calculation method is often complicated, cannot automatically identify the obstacle, and is low in precision.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a path planning method and an industrial robot, which effectively improve the efficiency of obstacle avoidance and path planning of the industrial robot.
In a first aspect, a method for path planning includes the following steps:
s1: obtaining a control point sample set; samples in the control point sample set comprise historical grasping point data of the industrial robot and categories of the historical grasping point data; the category is a collision control point or a non-collision control point;
s2: performing model training on the control point sample set to obtain a path planning model;
s3: receiving a plurality of capturing point data of the industrial robot input by a user;
s4: inputting all the grabbing point data into a path planning model;
s5: obtaining the category corresponding to each captured point data output by the path planning model;
s6: and outputting the feasible path according to the capture point data of which the category is the non-collision control point.
Preferably, the historical grasping point data is the position x of the end shaft of the industrial robotnWherein x isn∈RDD is the dimension of a vector space R of a real number domain, N takes values of 1-N, and N is the number of tail end shafts of the industrial robot;
the category is ynWherein y isnE {1, -1 }; when y isnWhen 1, the category is a non-collision control point, when ynWhen-1, the category is a collision control point.
Preferably, the step S2 specifically includes:
s11: initializing a weight matrix d of an SVM kernel function set and a support vector weight matrix alpha of a control point sample set, and enabling K to be 1;
s12: if K is less than the preset maximum iteration times and | | | d | | | calculation2If < T, executing step S13, otherwise executing step S15; k is iteration times, d is a weight matrix of the SVM kernel function set, and T is a preset weight matrix threshold value;
s13: fixing the weight matrix d, and updating the support vector weight matrix alpha;
s14: fixing the support vector weight matrix alpha and updating the weight matrix d;
s15: and outputting the optimized result to obtain the path planning model.
Preferably, the step S11 specifically includes:
initializing a weight matrix d ═ 1/(M +1),.... 1/(M +1) ];
initializing a support vector weight matrix α ═ 1/(M +1),...., 1/(M +1) ];
and M is the number of kernel functions in the adopted SVM kernel function set.
Preferably, the step S13 specifically includes:
1) calculating the amount of deviation of the number
Figure BDA0002881002320000021
Wherein KmFor the mth kernel in the SVM kernel set, dmIs the mth value of the weight matrix d, representing the weight value of the mth kernel function, alphanThe nth value of the support vector weight matrix alpha represents the support vector weight of the nth sample in the control point sample set;
2) when the minimum value of the numerical deviation quantity Q is less than or equal to 0, defining the subscript of the minimum value as i;
3) denote the ith column of the kernel matrix K as
Figure BDA0002881002320000022
kmIs the mth kernel function;
4) if α isiNot equal to 0, then Δ αi＝biyi-Qi/yi,αi＝αi+Δαi，QiIs the ith value of the deviation Q, biIs a constant; the number offset Q is updated according to:
Figure BDA0002881002320000031
wherein q ═ 1, 2...., N };
5) set of calculations U ═ Q1-α1,Q2-α2,......,QN-αN}；
6) If there are more than 0 elements in the set U, and αnNot equal to 0, the index with the largest value of the element in the set U is denoted i, and the number deviation Q is updated according to:
Figure BDA0002881002320000032
after the update, let alphaq＝0。
Preferably, the step S14 specifically includes:
calculating hyperplane linear weight W ═ W1,...wi,...wm},
Figure BDA0002881002320000033
Updating
Figure BDA0002881002320000034
p is 0-100;
let K be K +1, return to step S12.
Preferably, the step S15 specifically includes:
outputting the optimized result { alpha*，d*,K*Therein of
Figure BDA0002881002320000035
Figure BDA0002881002320000036
Preferably, the step S3 specifically includes:
receiving angle values of D joints at the end of the industrial robot input by a user: theta is ═ theta1,θ2,...,θD]；
Calculating the position point Pos of the h-th jointhAnd a transformation matrix Jh；
Figure BDA0002881002320000037
H is the number of joints of the industrial robot, PoshIs a transformation matrix JhThe amount of translation of;
computing matrix KM，
Figure BDA0002881002320000038
Wherein r ismAre real numbers.
Preferably, the step S5 specifically includes:
calculating a nuclear function value of an end joint of the industrial robot:
Figure BDA0002881002320000039
if F (x) > 0, the category of the grabbing point is a non-collision control point, and if F (x) < 0, the category of the grabbing point is a collision control point.
In a second aspect, an industrial robot comprises a processor and a memory, the processor and the memory being connected to each other, wherein the memory is used for storing a computer program, the computer program comprising program instructions, the processor being configured to invoke the program instructions to execute the path planning method according to the first aspect.
According to the technical scheme, the path planning method and the industrial robot provided by the invention have the advantages that the condition of multiple historical grabbing points of the industrial robot is obtained for model training, and the industrial robot is prevented from mechanical limitation and mistaken grabbing during grabbing. In addition, model training provides an obstacle avoidance algorithm based on the SVM algorithm model, a mechanical arm path planning algorithm is accurately set, redundant calculation in the traditional path planning process is avoided, and the obstacle avoidance and path planning efficiency of the industrial robot is effectively improved.
Drawings
In order to more clearly illustrate the detailed description of the invention or the technical solutions in the prior art, the drawings that are needed in the detailed description of the invention or the prior art will be briefly described below. Throughout the drawings, like elements or portions are generally identified by like reference numerals. In the drawings, elements or portions are not necessarily drawn to scale.
Fig. 1 is a flowchart of a path planning method according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a three-dimensional pose of a visual recognition obstacle.
FIG. 3 is a schematic of a kernel function of a training model.
Fig. 4 is a schematic diagram of the collision determination position applied by the model.
Detailed Description
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and therefore are only examples, and the protection scope of the present invention is not limited thereby. It is to be noted that, unless otherwise specified, technical or scientific terms used herein shall have the ordinary meaning as understood by those skilled in the art to which the invention pertains.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
As used in this specification and the appended claims, the term "if" may be interpreted contextually as "when", "upon" or "in response to a determination" or "in response to a detection". Similarly, the phrase "if it is determined" or "if a [ described condition or event ] is detected" may be interpreted contextually to mean "upon determining" or "in response to determining" or "upon detecting [ described condition or event ]" or "in response to detecting [ described condition or event ]".
The first embodiment is as follows:
a path planning method, see fig. 1, comprising the steps of:
s1: obtaining a control point sample set; samples in the control point sample set comprise historical grasping point data of the industrial robot and categories of the historical grasping point data; the category is a collision control point or a non-collision control point, as shown in fig. 2;
specifically, the historical grasping point data is the position x of the tail end shaft of the industrial robotnWherein x isn∈RDD is the dimension of a vector space R of a real number domain, N takes values of 1-N, and N is the number of tail end shafts of the industrial robot. Position x of end shaft of industrial robotnThe pose of the robot end flange and the pose of the workpiece can be obtained through image data and point cloud data by using a visual identification technology.
The category is ynWherein y isnE {1, -1 }; when y isnWhen the number is 1, the category is a non-collision control point, namely an obstacle control point; when y isnWhen the category is-1, the category is a collision control point, i.e., a feasible control point.
S2: performing model training on the control point sample set to obtain a path planning model, which specifically comprises:
s11: initializing a weight matrix d of an SVM kernel function set and a support vector weight matrix alpha of a control point sample set, and enabling K to be 1; the method specifically comprises the following steps:
initializing a weight matrix d ═ 1/(M +1),.... 1/(M +1) ];
initializing a support vector weight matrix α ═ 1/(M +1),...., 1/(M +1) ];
and M is the number of kernel functions in the adopted SVM kernel function set.
S12: if K is less than the preset maximum iteration times and | | | d | | | calculation2If < T, executing step S13, otherwise executing step S15; k is iteration times, d is a weight matrix of the SVM kernel function set, and T is a preset weight matrix threshold value;
s13: fixing the weight matrix d, and updating the support vector weight matrix alpha, specifically comprising:
1) calculating the amount of deviation of the number
Figure BDA0002881002320000061
Wherein KmFor the mth kernel in the SVM kernel set, dmIs the mth value of the weight matrix d, representing the weight value of the mth kernel function, alphanThe nth value of the support vector weight matrix alpha represents the support vector weight of the nth sample in the control point sample set;
2) when the minimum value of the numerical deviation quantity Q is less than or equal to 0, defining the subscript of the minimum value as i;
3) denote the ith column of the kernel matrix K as
Figure BDA0002881002320000062
kmFor the mth kernel function, common kernel functions include Gaussian kernel function and polynomial kernel functionNumbers, linear kernel functions, and the like;
4) if α isiNot equal to 0, then Δ αi＝biyi-Qi/yi,αi＝αi+Δαi，QiIs the ith value of the deviation Q, biIs a constant, usually 10, and has a value range of real numbers greater than 0; the number offset Q is updated according to:
Figure BDA0002881002320000063
wherein q ═ 1, 2...., N };
5) set of calculations U ═ Q1-α1,Q2-α2,......,QN-αN}；
6) If there are more than 0 elements in the set U, and αnNot equal to 0, the index with the largest value of the element in the set U is denoted i, and the number deviation Q is updated according to:
Figure BDA0002881002320000064
after the update, let alphaq＝0。
S14: fixing the support vector weight matrix alpha, and updating the weight matrix d, specifically comprising:
calculating hyperplane linear weight W ═ W1,...wi,...wm},
Figure BDA0002881002320000065
Updating
Figure BDA0002881002320000071
p is 0-100, and the default is 2;
let K be K +1, return to step S12.
S15: outputting an optimization result to obtain the path planning model, which specifically comprises the following steps:
outputting the optimized result { alpha*，d*,K*Therein of
Figure BDA0002881002320000072
Figure BDA0002881002320000073
Wherein alpha is*For the optimized support vector weight matrix, d*For the optimized weight matrix, K*Is the optimized kernel matrix.
S3: receiving a plurality of points of grabbing data of an industrial robot input by a user, specifically comprising:
receiving angle values of D joints at the end of the industrial robot input by a user: theta is ═ theta1,θ2,...,θD](ii) a Calculating the position point Pos of the h-th jointhAnd a transformation matrix Jh；
Figure BDA0002881002320000074
H is the joint number of the industrial robot, and default is D, PoshIs a transformation matrix JhThe amount of translation of;
computing matrix KM，
Figure BDA0002881002320000075
Wherein r ismThe value range is a real number between preset intervals (a, b).
Specifically, when the path planning method is used for path planning, the pose of the workpiece, the position of the obstacle and the pose of the workpiece coordinate system in the industrial robot base coordinate system need to be determined.
S4: inputting all the grabbing point data into a path planning model;
s5: obtaining the category corresponding to each captured point data output by the path planning model, specifically comprising:
as shown in fig. 3, a kernel function value of the end joint of the industrial robot is calculated:
Figure BDA0002881002320000076
if F (x) > 0, the category of the grasping point is the non-collision control point, and if F (x) < 0, the category of the grasping point is the collision control point, as shown in FIG. 4.
S6: and outputting the feasible path according to the capture point data of which the category is the non-collision control point.
Specifically, the method calculates different feasible kinematic inverse solution postures of the robot at the same position, and one workpiece can have multiple postures. According to the method, the grabbing points meeting the requirements are selected from the multiple grabbing points, and the grabbing points can guarantee that the industrial robot can effectively grab the workpiece and can also guarantee the realization of rapid obstacle avoidance and optimal path planning.
The method obtains the types of a plurality of grasping point data of the industrial robot through steps S3-S6, and outputs a feasible path according to the grasping points of which the types are non-collision control points, so that each point of the feasible path cannot encounter an obstacle and cannot collide with the obstacle, and the functions of obstacle avoidance and path planning of the robot are realized.
According to the method, model training is carried out by acquiring the conditions of a plurality of historical grabbing points of the industrial robot, and mechanical limitation and mistaken grabbing of the industrial robot during grabbing are avoided. In addition, the model training provides an obstacle avoidance algorithm based on the SVM algorithm model, the mechanical arm path planning algorithm is accurately set, redundant calculation in the traditional path planning process is avoided, whether the robot collides with an obstacle or not is observed and calculated by naked eyes, and the obstacle avoidance and path planning efficiency of the industrial robot is effectively improved.
Example two:
an industrial robot comprising a processor and a memory, said processor and memory being interconnected, wherein said memory is adapted to store a computer program, said computer program comprising program instructions, said processor being configured to invoke said program instructions to perform the path planning method of embodiment one.
It should be understood that in the embodiments of the present invention, the Processor may be a Central Processing Unit (CPU), and the Processor may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may include both read-only memory and random access memory, and provides instructions and data to the processor. The portion of memory may also include non-volatile random access memory. For example, the memory may also store device type information.
For a brief description, the embodiments of the present invention do not refer to the corresponding contents in the foregoing embodiments of the method.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the present invention, and they should be construed as being included in the following claims and description.

Claims (8)

1. A method of path planning, comprising the steps of:
s1: obtaining a control point sample set; samples in the control point sample set comprise historical grasping point data of the industrial robot and categories of the historical grasping point data; the category is a collision control point or a non-collision control point;
s2: performing model training on the control point sample set to obtain a path planning model;
s3: receiving a plurality of capturing point data of the industrial robot input by a user;
s4: inputting all the grabbing point data into a path planning model;
s5: obtaining the category corresponding to each captured point data output by the path planning model;
s6: outputting a feasible path according to the captured point data of which the category is the non-collision control point;
the historical grasping point data is the position x of the tail end shaft of the industrial robotnWherein x isn∈RDD is the dimension of a vector space R of a real number domain, N takes values of 1-N, and N is the number of tail end shafts of the industrial robot;
the category is ynWherein y isnE {1, -1 }; when y isnWhen 1, the category is a non-collision control point, when ynWhen the value is equal to-1, the category is a collision control point;
the step S2 specifically includes:
s11: initializing a weight matrix d of an SVM kernel function set and a support vector weight matrix alpha of a control point sample set, and enabling K to be 1;
s12: if K is less than the preset maximum iteration times and | | | d | | | calculation2If < T, executing step S13, otherwise executing step S15; k is iteration times, d is a weight matrix of the SVM kernel function set, and T is a preset weight matrix threshold value;
s13: fixing the weight matrix d, and updating the support vector weight matrix alpha;
s14: fixing the support vector weight matrix alpha and updating the weight matrix d;
s15: and outputting the optimized result to obtain the path planning model.
2. The path planning method according to claim 1, wherein the step S11 specifically includes:
initializing a weight matrix d ═ 1/(M +1),.... 1/(M +1) ];
initializing a support vector weight matrix α ═ 1/(M +1),...., 1/(M +1) ];
and M is the number of kernel functions in the adopted SVM kernel function set.
3. The path planning method according to claim 2, wherein the step S13 specifically includes:
1) calculating the amount of deviation of the number
Figure FDA0003492694440000021
Wherein KmFor the mth kernel in the SVM kernel set, dmIs the mth value of the weight matrix d, representing the weight value of the mth kernel function, alphanThe nth value of the support vector weight matrix alpha represents the support vector weight of the nth sample in the control point sample set;
2) when the minimum value of the numerical deviation quantity Q is less than or equal to 0, defining the subscript of the minimum value as i;
3) denote the ith column of the kernel matrix K as
Figure FDA0003492694440000022
kmIs the mth kernel function;
4) if α isiNot equal to 0, then Δ αi＝biyi-Qi/yi,αi＝αi+Δαi，QiIs the ith value of the deviation Q, biIs a constant; the number offset Q is updated according to:
Figure FDA0003492694440000023
wherein q ═ 1, 2...., N };
5) set of calculations U ═ Q1-α1,Q2-α2,......,QN-αN}；
6) If there are more than 0 elements in the set U, and αnNot equal to 0, the index with the largest value of the element in the set U is denoted i, and the number deviation Q is updated according to:
Figure FDA0003492694440000024
after the update, let alphaq＝0。
4. The path planning method according to claim 3, wherein the step S14 specifically includes:
calculating hyperplane linear weight W ═ W1,...wi,...wm},
Figure FDA0003492694440000025
Updating
Figure FDA0003492694440000026
p is 0-100;
let K be K +1, return to step S12.
5. The path planning method according to claim 4, wherein the step S15 specifically includes:
outputting the optimized result { alpha*，d*,K*Therein of
Figure FDA0003492694440000031
Figure FDA0003492694440000032
6. The path planning method according to claim 5, wherein the step S3 specifically includes:
receiving angle values of D joints at the end of the industrial robot input by a user: theta is ═ theta1,θ2,...,θD]；
Calculating the position point Pos of the h-th jointhAnd a transformation matrix Jh；
Figure FDA0003492694440000033
H is the number of joints of the industrial robot, PoshIs a transformation matrix JhThe amount of translation of;
computing matrix KM，
Figure FDA0003492694440000034
Wherein r ismAre real numbers.
7. The path planning method according to claim 6, wherein the step S5 specifically includes:
calculating a nuclear function value of an end joint of the industrial robot:
Figure FDA0003492694440000035
if F (x) > 0, the category of the grabbing point is a non-collision control point, and if F (x) < 0, the category of the grabbing point is a collision control point.
8. An industrial robot comprising a processor and a memory, said processor and memory being interconnected, wherein said memory is adapted to store a computer program comprising program instructions, said processor being configured to invoke said program instructions to perform a path planning method according to any of claims 1 to 7.
CN202011640924.XA 2020-12-31 2020-12-31 Path planning method and industrial robot Active CN112847358B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011640924.XA CN112847358B (en) 2020-12-31 2020-12-31 Path planning method and industrial robot

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011640924.XA CN112847358B (en) 2020-12-31 2020-12-31 Path planning method and industrial robot

Publications (2)

Publication Number Publication Date
CN112847358A CN112847358A (en) 2021-05-28
CN112847358B true CN112847358B (en) 2022-04-26

Family

ID=76000712

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011640924.XA Active CN112847358B (en) 2020-12-31 2020-12-31 Path planning method and industrial robot

Country Status (1)

Country Link
CN (1) CN112847358B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113752265B (en) * 2021-10-13 2024-01-05 国网山西省电力公司超高压变电分公司 Method, system and device for planning obstacle avoidance path of mechanical arm
CN115091470B (en) * 2022-08-18 2022-11-18 季华实验室 Friction compensation model training method and device, electronic equipment and storage medium

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104062902B (en) * 2014-05-15 2017-01-04 江南大学 Delta robot time optimal trajectory planning method
CN105184019A (en) * 2015-10-12 2015-12-23 中国科学院自动化研究所 Robot grabbing method and system
CN205734906U (en) * 2016-05-18 2016-11-30 天津畅乐电子科技股份有限公司 A kind of multi-functional quick dexterous manipulation hands structure
CN107283428A (en) * 2017-08-22 2017-10-24 北京京东尚科信息技术有限公司 robot control method, device and robot
CN107729953B (en) * 2017-09-18 2019-09-27 清华大学 Robot plume method for tracing based on continuous state behavior domain intensified learning
CN108196453B (en) * 2018-01-24 2020-11-20 中南大学 Intelligent calculation method for mechanical arm motion planning group

Also Published As

Publication number Publication date
CN112847358A (en) 2021-05-28

Similar Documents

Publication Publication Date Title
US11878433B2 (en) Method for detecting grasping position of robot in grasping object
CN112847358B (en) Path planning method and industrial robot
US20210166418A1 (en) Object posture estimation method and apparatus
JP5647155B2 (en) Body feature detection and human pose estimation using inner distance shape relation
Shademan et al. Robust jacobian estimation for uncalibrated visual servoing
US11161244B2 (en) System and method for automatic error recovery in robotic assembly
US11458619B2 (en) Method for controlling the motion of a robot arm, computing device and robot
WO2021082229A1 (en) Data processing method and related device
US20230077638A1 (en) Deterministic robot path planning method for obstacle avoidance
La Anh et al. Robotic grasping based on efficient tracking and visual servoing using local feature descriptors
CN113269835A (en) Industrial part pose identification method and device based on contour features and electronic equipment
Zhang et al. QR code-based self-calibration for a fault-tolerant industrial robot arm
JP2020077231A (en) Position detection program, position detection method and position detection device
Sanz et al. Planar grasping characterization based on curvature-symmetry fusion
US20210114206A1 (en) Techniques for generating controllers for robots
CN114049318A (en) Multi-mode fusion feature-based grabbing pose detection method
McGarry et al. Machine learning methods to improve the accuracy of industrial robots
Martínez et al. Visual identification and similarity measures used for on-line motion planning of autonomous robots in unknown environments
Liu et al. Application of deep reinforcement learning in control of drawing robots
CN111687829A (en) Anti-collision control method, device, medium and terminal based on depth vision
JP7307776B2 (en) Obstacle avoidance method for robot arm and obstacle avoidance system for robot arm
US20240135481A1 (en) Method and apparatus for 6dof object pose estimation using self-supervision learning
US11491650B2 (en) Distributed inference multi-models for industrial applications
US20220108046A1 (en) Generative design techniques for soft robot manipulators
Suo et al. Dynamic Optimization Data Association based on JCBB Algorithm

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant