• International Journal of Precision Engineering and Manufacturing
• /
• v.2 no.1
• /
• pp.76-80
• /
• 2001

A binary parallel robot manipulator uses actuators that have only two stable states being built by stacking variable geometry trusses on top of each other in a long serial chain. Discrete characteristics of the binary manipulator make it impossible to analyze an inverse kinematic problem in conventional ways. We therefore introduce new definitions of workspace and inverse kinematic solution, and the apply a genetic algorithm to the newly defied inverse kinematic problem. Numerical examples show that our genetic algorithm is very efficient to solve the inverse kinematic problem of binary robot manipulators.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.39 no.9
• /
• pp.985-996
• /
• 1990

An inverse kinematic solution of a robot manipulator using multilayer perceptrons is proposed. Neural networks allow the solution of some complex nonlinear equations such as the inverse kinematics of a robot manipulator without the need for its model. However, the back-propagation (BP) learning rule for multilayer perceptrons has the major limitation of being too slow in learning to be practical. In this paper, a new algorithm named Dynamically Reconfiguring BP is proposed to improve its learning speed. It uses a modified version of Kohonen's Self-Organizing Feature Map (SOFM) to partition the input space and for each input point, select a subset of the hidden processing elements or neurons. A subset of the original network results from these selected neuron which learns the desired mapping for this small input region. It is this selective property that accelerates convergence as well as enhances resolution. This network was used to learn the parity function and further, to solve the inverse kinematic problem of a robot manipulator. The results demonstrate faster learning than the BP network.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.551-556
• /
• 1991

This paper deal with the solution of kinematics and inverse kinematics of industrial FANUC robot by the bisection method with IBM PC 386. The inverse kinematics of FANUC robot cannot be solved by the algebraical method, because arm matrix T$_{6}$ is very complex and 6-joint angles are associated with the position and the approach of end-effector. Instead we found other 5-joint angle by an algebraical method after finding .theta.$_{4}$ value by a bisection method.d.

• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.12
• /
• pp.1026-1032
• /
• 2003

A biped walking robot which is developed as a platform for researching walking algorithm is briefly introduced. The developed walking robot has 6 degrees of freedom per one leg. The origins of the last three axis do not intersect at a point, so the kinematic analysis is cubmersome with the conventional method. In the former version of the robot, Jacobian-based inverse kinematics method is used. However, the Jacobian-based inverse kinematics method has drawbacks for the application in which knee is fully extended such as stair-case walking. The reason far that is the Jacobian becomes ill-conditioned near the singular points and the method is not able to give adequate solutions. So, a method for giving a closed-form inverse kinematics solution is proposed. The proposed method is based on careful consideration of the kinematic structure of the biped walking robot.

• Journal of Mechanical Science and Technology
• /
• v.20 no.7
• /
• pp.917-928
• /
• 2006

In recent decades, Artificial Neural Networks (ANNs) have become the focus of considerable attention in many disciplines, including robot control, where they can be used to solve nonlinear control problems. One of these ANNs applications is that of the inverse kinematic problem, which is important in robot path planning. In this paper, a neural network is employed to analyse of inverse kinematics of PUMA 560 type robot. The neural network is designed to find exact kinematics of the robot. The neural network is a feedforward neural network (FNN). The FNN is trained with different types of learning algorithm for designing exact inverse model of the robot. The Unimation PUMA 560 is a robot with six degrees of freedom and rotational joints. Inverse neural network model of the robot is trained with different learning algorithms for finding exact model of the robot. From the simulation results, the proposed neural network has superior performance for modelling complex robot's kinematics.

• 제어로봇시스템학회:학술대회논문집
• /
• 1989.10a
• /
• pp.124-126
• /
• 1989

Inverse kinematic problem is a crucial point for robot manipulator control. In this paper, to implement the Jacobian control technique we used the Hopfield(Tank)'s neural network. The states of neurons represent joint veocities, and the connection weights are determined from the current value of the Jacobian matrix. The network energy function is constructed so that its minimum corresponds to the minimum least square error. At each sampling time, connection weights and neuron states are updated according to current joint position. Inverse kinematic solution to the planar redundant manipulator is solved by computer simulation.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.36-40
• /
• 1990

Conventional back propagation learning is generally characterized by slow and rather inaccurate learning which makes it difficult to use in control applications. A new multilayer perception architecture and its learning algorithm is proposed that consists of a Kohonen front layer followed by a back propagation network. The Kohonen layer selects a subset of the hidden layer neurons for local tuning. This architecture has been tested on the inverse kinematic solution of robot manipulator while demonstrating its fast and accurate learning capabilities.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.953-958
• /
• 1992

This paper deals with the solution of inverse kinematics of the industrial FANUC robot with IBM PC386. The inverse kinematics of FANUC robot cannot be solved by the algebraical method, because arm matirix T$_{6}$ is very complex and 6-joint angles are associated with the position and the approach of end-effector. Instead we fuund otehr 5-joint angle by and algebraical method after finding .THETA.$_{1}$ value by a numerical method.d.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.18 no.6
• /
• pp.1421-1429
• /
• 1994

An algorithm is developed for solving the inverse kinematic problem of a 6-degree-of-freedom robot with a wrist offset for which the closed form inverse solutions are not obtainable, but knowledge of one joint variable allows closed form solutions of the remaining joint variables. The algorithm does not require Forward Kinematics nor Jacobian but uses the implicit kinematic relationships between joint variables and the given hand position. An iterative back substitution method is used to solve the inversion and the optimal conditions of the convergence are incoporated. An example is given to illustrate the concepts, the solution procedure and its convergency.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.28B no.10
• /
• pp.761-770
• /
• 1991

This paper suggests a measure constraint locus for characterization of the performance of a subtask for a redundant robot. The measure constraint locus are the loci of points satisfying the necessary constraint for optimality of measure in the joint configuration space. To uniquely obtain an inverse kinematic solution, one must consider both measure constraint locus and self-motion manifolds which are set of homogeneous solutions. Using measure constraint locus for maniqulability measure, the invertible workspace without singularities and the topological property of the configuration space for linding equilibrium configurations are analyzed. We discuss some limitations based on the topological arguments of measure constraint locus, of the inverse kinematic algorithm for a cyclic task. And the inverse kinematic algorithm using global maxima on self-motion manifolds is proposed and its property is studied.