CN108908340B - Redundant robot repetitive motion planning method adopting finite interval neural network - Google Patents

Redundant robot repetitive motion planning method adopting finite interval neural network Download PDF

Info

Publication number
CN108908340B
CN108908340B CN201810876449.2A CN201810876449A CN108908340B CN 108908340 B CN108908340 B CN 108908340B CN 201810876449 A CN201810876449 A CN 201810876449A CN 108908340 B CN108908340 B CN 108908340B
Authority
CN
China
Prior art keywords
neural network
robot
theta
convergence
end effector
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
CN201810876449.2A
Other languages
Chinese (zh)
Other versions
CN108908340A (en
Inventor
孙明轩
张钰
吴雨芯
翁丁恩
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.)
Zhejiang University of Technology ZJUT
Original Assignee
Zhejiang University of Technology ZJUT
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 Zhejiang University of Technology ZJUT filed Critical Zhejiang University of Technology ZJUT
Priority to CN201810876449.2A priority Critical patent/CN108908340B/en
Publication of CN108908340A publication Critical patent/CN108908340A/en
Application granted granted Critical
Publication of CN108908340B publication Critical patent/CN108908340B/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
    • 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)
  • Numerical Control (AREA)

Abstract

A redundant robot repetitive motion planning method based on finite interval neural network is disclosed, which is used in Cartesian space Given a desired trajectory r of a robotic end effector d(t) and giving the desired pull-in angle θ for each joint d(0) (ii) a For the repeated motion of the robot, asymptotic convergence performance indexes are adopted, a quadratic optimization problem of redundant robot trajectory planning is converted into a time-varying matrix equation solving problem, and a finite interval neural network is used as a solver. The invention realizes the repeated movement planning task of rapid limited time convergence of the redundant robot under the condition of initial position deviation.

Description

Redundant robot repetitive motion planning method adopting finite interval neural network
Technical Field
The invention relates to the motion planning technology of an industrial robot, and particularly provides a redundant robot repetitive motion control method based on a finite interval neural network, which can be converged in a finite time and under the condition that an initial position deviates from an expected track.
Background
Currently, industrial robots are widely used in machining, medical treatment, food industry, industrial manufacturing, logistics, military and other fields, and can perform heavy labor such as welding, spraying, and transportation. Due to the complex working environment, the moving object may collide with the obstacles in the environment. The redundant robot has good flexibility and fault tolerance, can utilize redundant degrees of freedom to enhance obstacle avoidance without influencing the operation of an end effector, and can complete variable operation tasks in a complex working environment.
The number of joints that a redundant robot can move independently is called the degree of freedom of motion. Redundant robots have more degrees of freedom than are required to perform a given end effector task, and redundant degrees of freedom make redundant solutions problems with an infinite number of solutions. The basic problem of real-time motion control of the robot is redundancy analysis, the corresponding values of each joint angle of the redundant robot are solved by knowing the position and the posture of an end effector, and the conventional method is an analysis scheme based on pseudo-inverse. Considering the relation between the angle of each joint of the n-degree-of-freedom robot and the displacement of the end effector
r(t)＝f(θ(t))
Wherein r (t) is a pose variable of the end effector under a Cartesian coordinate system, and theta (t) represents a joint angle. The relationship between the differential of the terminal Cartesian space and the joint space is
Figure BDA0001753385770000011
Wherein the content of the first and second substances,
Figure BDA0001753385770000014
And
Figure BDA0001753385770000015
Respectively, are the respective time derivatives of the same,
Figure BDA0001753385770000012
Is the robot jacobian matrix. Obtaining a minimum norm solution of the joint velocity variable by calculating the pseudo-inverse of J (theta)
Figure BDA0001753385770000013
Wherein, J +＝JT(JJT)-1Is the pseudo-inverse of the jacobian matrix.
Minimum speed norm performance indicators with equality constraints are the objective functions of motion planning (d.e. whole, Resolved motion control of manipulators and human prosthetics), IEEE trans. man-Machine syst.,1969,10(2):47-53)
Figure BDA0001753385770000021
Figure BDA0001753385770000022
In the formula, A is a positive definite weighting matrix. Solving the above-mentioned planning problem, the demand solves the following system of equations
Figure BDA0001753385770000023
It is solved into
Figure BDA0001753385770000024
It can be seen that formula (1) is a special case of formula (3) when a ═ I.
A redundancy resolution scheme based on Quadratic Optimization (QP) is concerned, and f.t. cheng, t. -h.cheng proposes a joint unbiased performance index in 1994 (f.t. cheng, t. -h.cheng, and y.sun, Resolving robot redundancy and integrity constraints (a redundant robot trajectory planning method under an inequality constraint), IEEE trans.robotics automation, 1994, 10 (1): 65-71):
Figure BDA0001753385770000025
the non-repeatability problem may generate an unexpected joint configuration, so that the repeated operation of the closed track of the end of the Redundant robot can cause unexpected situations, even accidents and dangerous situations, to happen, the most widely applied pseudo-Inverse Control method cannot ensure the repeatability of the movement, in order to complete the repeated movement, the method of self-movement is usually adopted for making up, and the adjustment of the self-movement is often not efficient (see the details of C A and Huang C, Review of pseudo-Inverse Control method for using with a kinetic Reandmanufacture manufacturers, IEEE, Systems, training, cyberneron, 1983,13(2): 245-250; Ja-K245. simulation Reductant managers (pseudo-Inverse Control method), Klns, Syst. Man, 1983,13, 2): pseudo-simulation, and pseudo-simulation of simulation system, and 856, and pseudo-simulation of simulation system, and simulation of simulation Systems, 849, and simulation of simulation, and simulation of simulation Systems, 13, respectively, and simulation of the like, namely simulation of simulation and simulation of the simulation system, model.
In order to execute the Repetitive motion task, a Repetitive motion index is introduced as an optimization criterion, and a Repetitive Motion Planning (RMP) scheme (Zhang Y, Wang J, Xia Y.A dual Neural network for a Repetitive motion resolution of a kinematic redundant robot trajectory to joint trajectories and joint velocities limits, IEEE Trans Neural network, 2003,14(3):658-667) is formed. A common repetitive motion index is the asymptotic convergence performance index AOC (asymptotic-convergent optimality criterion) as follows:
Figure BDA0001753385770000031
The Quadratic Programming (QP) here is solved using a Recurrent Neural Network (RNN) computation method. The general neural network solver has asymptotic convergence performance, and an effective solution can be obtained as long as the calculation time is long enough.
And describing a redundancy analysis problem based on quadratic programming, and solving published documents by adopting a recurrent neural network. Compared with a recurrent neural network with asymptotic convergence dynamic characteristics, the dynamic characteristics of the neural network with the limited interval have limited time convergence, and the neural network with the limited interval has high convergence speed and high calculation precision. It is worth pointing out that the redundant robot trajectory planning problem is resolved into a time-varying computation problem, and an effective method for solving the time-varying problem is to use a recurrent neural network computation method with limited time convergence.
Disclosure of Invention
The invention provides a neural network based on a limited interval, which has limited time convergence, high calculation precision and easy realization
Provided is a redundant robot motion planning method. The invention adopts asymptotic convergence performance indexes, converts the quadratic optimization problem of redundant robot trajectory planning into a time-varying matrix equation solving problem, takes a finite interval neural network as a solver, and realizes a repeated motion planning task of rapid finite time convergence of the redundant robot under the condition of initial position deviation.
In order to achieve the purpose, the invention provides the following technical scheme.
A redundant robot repetitive motion planning method based on a finite interval neural network comprises the following steps:
(1) Given a desired trajectory r of a robotic end effector in Cartesian space d(t) and giving the desired pull-in angle θ for each joint d(0)；
(2) For the robot with repeated motion, the initial joint angle is theta (0) ═ theta 0Initial joint angle θ 0May be different from the desired joint angle theta d(0) I.e. theta 0≠θd(0)；
(3) Describing the redundant robot repetitive motion planning as a quadratic planning problem, providing an asymptotic convergence performance index, and forming a repetitive motion planning scheme:
Figure BDA0001753385770000041
Wherein the content of the first and second substances,
g(θ)＝-ρ0(θ(t)-θd(0))
ρ0>0，θ(t)-θd(0) Indicating the deviation of the joint angle from the desired joint angle, since the initial position of the robot is not on the desired trajectory, a feedback deviation, r, must be added to the left of the end effector velocity vs. joint velocity relationship df (theta) representing the error between the desired trajectory and the actual trajectory, which error is continuously reduced to zero due to artificially converging relationships, and a parameter β for adjusting the rate at which the end effector moves to the desired trajectory, β >0; j (theta) is a Jacobian matrix obtained according to the DH parameters of the robot, and f (theta) is the actual track of the end effector of the robot;
(4) Constructing a dynamic property equation describing a finite interval neural network as follows
Figure BDA0001753385770000042
Wherein the content of the first and second substances,
Figure BDA0001753385770000043
E is an error matrix variable, E ijIs an element of the matrix E, p >0,κ>0, and q 1,q2,p1,p2Are all positive odd numbers, satisfy q 1<p1，q2<p2；
Equation (5) is stable in a limited interval, and its limited time convergence needs to be described in two cases:
1) When | E ij(0) | is not less than 1, from E ij(0) To E ij(t) varies by 0, and the convergence time satisfies
Figure BDA0001753385770000044
2) When | E ij(0)|<1, from E ij(0) To E ij(t) varies by 0, and the convergence time satisfies
Figure BDA0001753385770000045
Combining formula (6) and (7) and having a convergence time of
Figure BDA0001753385770000051
(5) Defining lagrange functions
Figure BDA0001753385770000052
Wherein λ is a Lagrange multiplier vector,
Figure BDA0001753385770000053
About
Figure BDA0001753385770000054
And lambda are respectively calculated to obtain the partial derivatives, and the partial derivatives are made to be zero and are obtained by sorting
W(t)Y(t)＝v (10)
Wherein the content of the first and second substances,
Figure BDA0001753385770000055
Figure BDA0001753385770000056
Figure BDA0001753385770000057
I is an identity matrix;
(6) To solve the quadratic programming problem in step (3), the error is defined by equation (10)
E(t)＝W(t)Y(t)-v (11)
Giving out the neural network equation of the limited interval according to the dynamic characteristic equation (5) of the neural network of the limited interval
Figure BDA0001753385770000059
And obtaining the angle of each joint of the robot through a neural network calculation process.
The technical conception of the invention is as follows: performance indicators with asymptotic convergence
Figure BDA0001753385770000058
Wherein the content of the first and second substances,
g(θ)＝-ρ0(θ(t)-θd(0))
ρ0>0，θ(t)-θd(0) representing the deviation of the joint angle from the desired joint angle, this error is continuously reduced to zero due to the artificially constructed convergence relationship >0) May be used to adjust the rate at which the end effector moves to the desired trajectory.
Method for constructing finite interval neural network dynamic characteristic equation for redundant robot repetitive motion planning
Figure BDA0001753385770000061
Wherein the content of the first and second substances,
Figure BDA0001753385770000062
E is an error matrix variable, E ijIs an element of the matrix E, p >0,κ>0, and q 1,q2,p1,p2Are all positive odd numbers, satisfy q 1<p1，q2<p2；
Equation (5) is stable in a limited interval, and its limited time convergence needs to be described in two cases:
1) When | E ij(0) | is not less than 1, from E ij(0) To E ij(t) varies by 0, and the convergence time satisfies
Figure BDA0001753385770000063
2) When | E ij(0)|<1, from E ij(0) To E ij(t) varies by 0, and the convergence time satisfies
Figure BDA0001753385770000064
Combining formula (6) and (7) and having a convergence time of
Figure BDA0001753385770000065
When E is ij(t) the finite time converges to zero, indicating that the end effector is moving along the desired trajectory.
The invention has the advantages that the finite interval neural network method is applied to the redundant robot repetitive motion planning, and the repetitive motion planning task of rapid finite time convergence of the redundant robot is realized under the condition of initial position deviation. Compared with the asymptotic neural network and the binomial attraction final state neural network repetitive motion planning method, the finite interval neural network method has the characteristic of faster convergence, and finally the precision of the convergence angle of each joint of the robot is higher. The method is suitable for solving the time-varying problem (the formula (10) is a time-varying matrix equation), is easy to realize in engineering application, has low cost and meets the actual engineering requirements.
Drawings
Fig. 1 is a flow chart of a re-planning scheme provided by the present invention.
FIG. 2 is E ij(0) Taking functions of different initial values
Figure BDA0001753385770000074
Fig. 3 is a diagram of a redundant robot PUMA560 of the present invention for a repetitive planning scheme. (extracted from Y Zhang, ZZhang. repeatable motion planning and control of redundant robot repeated motion planning and control.) Berlin: Springer,2013:23-24)
Fig. 4 is a motion trajectory of the redundant robot PUMA560 end effector.
Fig. 5 is a motion trajectory of each joint of the redundant robot PUMA560, wherein 1 is a mechanical arm of the PUMA560, 2 is an elbow trajectory, 3 is a shoulder trajectory, 4 is a wrist trajectory, and 5 is an end effector motion trajectory.
Fig. 6 shows the respective joint angles of the redundant robot PUMA 560.
Fig. 7 shows the joint speeds of the redundant robot PUMA 560.
Fig. 8 illustrates various position errors of the redundant robot PUMA560 end effector.
FIG. 9 shows the error of the finite interval neural network and the asymptotic neural network in the solution of the binomial attraction final state neural network.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
Referring to fig. 1 to 9, a redundant robot repetitive motion planning method based on a finite interval neural network, fig. 1 is a flow chart of a redundant robot repetitive motion planning scheme, which is composed of the following three steps: 1. determining an expected track of the redundant robot end effector and an expected angle of each joint to be folded; 2. adopting asymptotic convergence performance indexes and forming a redundant robot repetitive motion quadratic programming scheme; 3. solving a quadratic programming problem by using a finite interval neural network to obtain each joint angle track, which comprises the following specific steps:
1) Determining a desired trajectory
Setting desired closeout of redundant robot PUMA560
Figure BDA0001753385770000071
Determining the center coordinates of a circular trajectory
(x is 0.2m, y is 0, z is 0), the radius of the circle is 0.2m, and the included angle between the circular surface and the xy plane is 0
Figure BDA0001753385770000072
The end effector completes the circular trajectory time T equal to 10 s. Setting six joint angle initial values of the robot to be the initial values considering that the initial position of the redundant robot is not on the expected motion track
Figure BDA0001753385770000073
2) Secondary planning scheme for establishing repeated motion of redundant robot
Describing the redundant robot repetitive motion trajectory planning as the following quadratic planning problem, and providing an asymptotic convergence performance index
Figure BDA0001753385770000081
Wherein the content of the first and second substances,
g(θ)＝-ρ0(θ(t)-θd(0))
ρ0>0，θ(t)-θd(0) Indicating the deviation of the joint angle from the desired joint angle, since the initial position of the robot is not on the desired trajectory, a feedback deviation, r, must be added to the left of the end effector velocity vs. joint velocity relationship d-f (θ) representing the error existing between the desired trajectory and the actual trajectory; due to the artificially constructed convergence relationship, this error is continually reduced, until zero, parameter β >0, which is used to adjust the rate at which the end effector moves to the desired trajectory; j (theta) is a Jacobian matrix obtained according to the DH parameters of the robot, and f (theta) is the actual track of the end effector of the robot;
3) Constructing a finite interval neural network to solve the quadratic programming problem
Constructing a neural network model with a limited interval and a dynamic characteristic equation of
Figure BDA0001753385770000082
Wherein the content of the first and second substances,
Figure BDA0001753385770000083
E is an error matrix variable, E ijIs an element of the matrix E, p >0,κ>0, and q 1,q2,p1,p2Are all positive odd numbers, satisfy q 1<p1，q2<p2；
Equation (5) is stable in a limited interval, and its limited time convergence needs to be described in two cases:
1) When | E ij(0) | is not less than 1, from E ij(0) To E ij(t) varies by 0, and the convergence time satisfies
Figure BDA0001753385770000084
2) When | E ij(0)|<1, from E ij(0) To E ij(t) varies by 0, and the convergence time satisfies
Figure BDA0001753385770000085
Combining formula (6) and (7) and having a convergence time of
Figure BDA0001753385770000091
Defining the following Lagrangian function
Figure BDA0001753385770000092
To pair
Figure BDA0001753385770000093
About
Figure BDA0001753385770000094
And lambda are respectively calculated to obtain the partial derivatives, and the partial derivatives are made to be zero and are obtained by sorting
W(t)Y(t)＝v (10)
Wherein the content of the first and second substances,
Figure BDA0001753385770000095
Figure BDA0001753385770000096
Figure BDA0001753385770000097
I is an identity matrix;
Error is defined by equation (10)
E(t)＝W(t)Y(t)-v (11)
Giving out the neural network equation of the limited interval according to the dynamic characteristic equation (5) of the neural network of the limited interval
Figure BDA0001753385770000098
And obtaining the angle of each joint of the robot through a neural network calculation process.
FIG. 2 shows that when E ij(0) Function taking different values
Figure BDA0001753385770000099
It can be seen from the figure that when E ij(0) Greater than 1, equal to 1 and less than 1, finite interval neural network error The differential state equations can converge quickly.
The redundant robot PUMA560 shown in fig. 3 is used to implement the repeat planning scheme of the present invention. The robot consists of 1 base, 3 connecting rods and 6 joints, wherein the length of each connecting rod is
L＝[0.4318,0,0.4318,0.25625]Tm。
fig. 4 shows the motion trajectory of the robot end effector in space, the target circular trajectory and the motion trajectory of the robot end effector are shown, it can be seen that the initial position of the end effector is not on the desired trajectory, and as shown in fig. 8, as time increases (T10 s), the accuracy of the final position error of the end effector reaches 2 × 10 in the three directions of the XYZ axes -5m, the actual trajectory coincides with the desired trajectory.
Fig. 5 shows the motion trail of each joint of the redundant robot. As can be seen from the figure, the trajectories of the respective joints are closed after one cycle of the operation, and repetitive motion control is realized.
In order to verify the effectiveness of the finite interval neural network method in the repetitive motion planning, the joint angle transient trajectory and the joint angular velocity transient trajectory obtained in the process of completing the circular trajectory by the robot PUMA560 end effector are shown in fig. 6 and 7. It can be seen that the joint angles of the redundant robot eventually converge to the desired joint angle positions.
when T is 10s, the maximum deviation between the joint angle before and after the robot movement and the expected joint angle obtained by solving by adopting the finite interval neural network is 7.7363 × 10 -5rad, maximum deviation of final value error of each joint angle obtained by solving with an asymptotic neural network is 3.0485 × 10 -4rad, the maximum deviation of the final value error of each joint angle obtained by solving with a binomial attractor final state neural network is 2.1004 × 10 -4rad, detailed in table 1:
Figure BDA0001753385770000101
TABLE 1
Wherein the asymptotic neural network dynamics
Figure BDA0001753385770000102
Dynamic characteristic of two-term attractor final state neural network
Figure BDA0001753385770000103
Finite interval neural network dynamics
Figure BDA0001753385770000104
In order to compare convergence performance of the asymptotic neural network, the binomial attraction final state neural network and the finite interval neural network, error norm | | | | E (t) | | | | W (t) Y (t) -v (t) | is calculated 2. Fig. 9 shows the error norm convergence trajectories for solving the quadratic programming problem with an asymptotic neural network, a binomial attraction final state neural network, and a finite interval neural network, respectively. As can be seen from the figure, when the norm of E (T) converges to 0.01, it takes time T (E) to solve with the finite interval neural network ij(0) 0.48s, time T (E) is required to solve with a binomial attraction final state neural network ij(0) 0.56s, time T (E) required to solve with the asymptotic neural network ij(0))＝2.45s。

Claims (1)

1. A robot repetitive motion planning method based on a finite interval neural network is characterized in that: the method comprises the following steps:
(1) Given a desired trajectory r of a robotic end effector in Cartesian space d(t) and giving the desired joint angle θ of each joint d(0)；
(2) For the robot with repeated motion, the initial joint angle is theta (0) ═ theta 0Initial joint angle θ 0Different from the desired joint angle theta d(0) I.e. theta 0≠θd(0)；
(3) Describing the redundant robot repetitive motion planning as a quadratic planning problem, providing an asymptotic convergence performance index, and forming a repetitive motion planning scheme:
Figure FDA0002483499630000011
Wherein the content of the first and second substances,
g(θ)＝-ρ0(θ(t)-θd(0))
ρ0>0，θ(t)-θd(0) Indicating the deviation of the joint angle from the desired joint angle, since the initial position of the robot is not on the desired trajectory, a feedback deviation, r, must be added to the left of the end effector velocity vs. joint velocity relationship df (theta) representing the error between the desired trajectory and the actual trajectory, which error is continuously reduced to zero due to artificially converging relationships, and a parameter β for adjusting the rate at which the end effector moves to the desired trajectory, β >0; j (theta) is a Jacobian matrix obtained according to the DH parameters of the robot, and f (theta) is the actual track of the end effector of the robot;
(4) Constructing a dynamic property equation describing a finite interval neural network as follows
Figure FDA0002483499630000012
Wherein the content of the first and second substances,
Figure FDA0002483499630000013
E is an error matrix variable, E ijIs an element of the matrix E, p >0,κ>0, and q 1,q2,p1,p2Are all positive odd numbers, satisfy q 1<p1，q2<p2；
Equation (5) is stable in a limited interval, and its limited time convergence needs to be described in two cases:
1) When | E ij(0) | is not less than 1, from E ij(0) To E ij(t) varies by 0, and the convergence time satisfies
Figure FDA0002483499630000021
2) When | E ij(0)|<1, from E ij(0) To E ij(t) varies by 0, and the convergence time satisfies
Figure FDA0002483499630000022
Combining formula (6) and (7) and having a convergence time of
Figure FDA0002483499630000023
(5) Defining lagrange functions
Figure FDA0002483499630000024
Wherein λ is a Lagrange multiplier vector,
Figure FDA0002483499630000025
About
Figure FDA0002483499630000026
And lambda are respectively calculated to obtain the partial derivatives, and the partial derivatives are made to be zero and are obtained by sorting
W(t)Y(t)＝v (10)
Wherein the content of the first and second substances,
Figure FDA0002483499630000027
Figure FDA0002483499630000028
Figure FDA0002483499630000029
I is an identity matrix;
(6) To solve the quadratic programming problem in step (3), the error is defined by equation (10)
E(t)＝W(t)Y(t)-v (11)
Giving out the neural network equation of the limited interval according to the dynamic characteristic equation (5) of the neural network of the limited interval
Figure FDA00024834996300000210
And obtaining the angles of all joints of the robot through a neural network calculation process.
CN201810876449.2A 2018-08-03 2018-08-03 Redundant robot repetitive motion planning method adopting finite interval neural network Active CN108908340B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810876449.2A CN108908340B (en) 2018-08-03 2018-08-03 Redundant robot repetitive motion planning method adopting finite interval neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810876449.2A CN108908340B (en) 2018-08-03 2018-08-03 Redundant robot repetitive motion planning method adopting finite interval neural network

Publications (2)

Publication Number Publication Date
CN108908340A CN108908340A (en) 2018-11-30
CN108908340B true CN108908340B (en) 2020-08-04

Family

ID=64393258

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810876449.2A Active CN108908340B (en) 2018-08-03 2018-08-03 Redundant robot repetitive motion planning method adopting finite interval neural network

Country Status (1)

Country Link
CN (1) CN108908340B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110083160B (en) * 2019-05-16 2022-04-19 哈尔滨工业大学（深圳） Robot trajectory planning method based on deep learning
US11235459B2 (en) * 2019-08-15 2022-02-01 Intrinsic Innovation Llc Inverse kinematic solver for wrist offset robots
CN111890363B (en) * 2020-07-27 2022-12-30 四川大学 Mechanical arm motion planning method based on rapid self-adaptive gradient neural network algorithm

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013215839A (en) * 2012-04-09 2013-10-24 Jtekt Corp Method for controlling robot having redundant degree of freedom, device for controlling robot, and system for controlling robot
CN107127754A (en) * 2017-05-09 2017-09-05 浙江工业大学 A kind of redundant mechanical arm repetitive motion planning method based on final state attraction optimizing index
CN107891424A (en) * 2017-11-10 2018-04-10 浙江科技学院 A kind of finite time Neural network optimization for solving redundant mechanical arm inverse kinematics
CN107972030A (en) * 2017-11-10 2018-05-01 浙江科技学院 A kind of initial position modification method in redundant mechanical arm repeating motion
CN107972031A (en) * 2017-11-10 2018-05-01 浙江科技学院 A kind of redundant mechanical arm repeats the initial position fix method of movement

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013215839A (en) * 2012-04-09 2013-10-24 Jtekt Corp Method for controlling robot having redundant degree of freedom, device for controlling robot, and system for controlling robot
CN107127754A (en) * 2017-05-09 2017-09-05 浙江工业大学 A kind of redundant mechanical arm repetitive motion planning method based on final state attraction optimizing index
CN107891424A (en) * 2017-11-10 2018-04-10 浙江科技学院 A kind of finite time Neural network optimization for solving redundant mechanical arm inverse kinematics
CN107972030A (en) * 2017-11-10 2018-05-01 浙江科技学院 A kind of initial position modification method in redundant mechanical arm repeating motion
CN107972031A (en) * 2017-11-10 2018-05-01 浙江科技学院 A kind of redundant mechanical arm repeats the initial position fix method of movement

Also Published As

Publication number Publication date
CN108908340A (en) 2018-11-30

Similar Documents

Publication Publication Date Title
CN107962566B (en) Repetitive motion planning method for mobile mechanical arm
CN106985138B (en) Attract the redundant mechanical arm method for planning track of optimizing index based on final state
CN107972030B (en) Initial position correction method in redundant mechanical arm repeated movement
CN108908347B (en) Fault-tolerant repetitive motion planning method for redundant mobile mechanical arm
CN107972031B (en) Method for positioning repeatable-motion initial position of redundant mechanical arm
CN108908340B (en) Redundant robot repetitive motion planning method adopting finite interval neural network
CN109159122B (en) Redundant robot repetitive motion planning method adopting elliptical final state neural network
CN109159124B (en) Redundant robot repetitive motion planning method adopting rapid double-power final state neural network
Vasu Fuzzy PID based adaptive control on industrial robot system
CN109940615B (en) Terminal state network optimization method for synchronous repeated motion planning of double-arm manipulator
CN109159121B (en) Redundant robot repetitive motion planning method adopting parabolic final state neural network
Liu et al. Research on kinematic modeling and analysis methods of UR robot
CN108908341B (en) Secondary root type final state attraction redundant robot repetitive motion planning method
CN116175585A (en) UDE control method for multi-joint mechanical arm with input saturation and output constraint
Chaparro-Altamirano et al. Kinematic and workspace analysis of a parallel robot used in security applications
Jiang et al. Coordination control of dual-arm robot based on modeled predictive control
Li et al. Fault-tolerant motion planning of redundant manipulator with initial position error
Yang et al. An obstacle avoidance and trajectory tracking algorithm for redundant manipulator end
Yerlikaya et al. Obtaining high-dimensional configuration space for robotic systems operating in a common environment
Kuppan Chetty et al. A heuristic approach towards path planning and obstacle avoidance control of planar manipulator
JALADI et al. An inverse kinematics analysis of space station remote manipulator system (SSRMS) using genetic algorithms
Duan et al. The Optimal Solution for Base Frame Installation of Dual-arm Robot
Fang et al. An iteration method for inverse kinematics of redundancy robot
CN118123811A (en) Redundant robot repetitive motion planning method based on initial value correction
CN111993466B (en) Double-arm robot combined operation test method based on laser tracker

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