• 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.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.25 no.3
• /
• pp.339-352
• /
• 2001

The inverse kinematics problem of Stewart platform is straightforward, but no closed form solution of the forward kinematic problem has been presented. Since we need the real-time forward kinematic solution in MIMO control and the motion monitoring of the platform, it is important to acquire the 6 DOF displacements of the platform from measured lengths of six cylinders in small sampling period. Newton-Raphson method a simple algorithm and good convergence, but it takes too long calculation time. So we reduce 6 nonlinear kinematic equations to 3 polynomials using Nairs method and 3 polynomials to 2 polynomials. Then Newton-Raphson method is used to solve 3 polynomials and 2 polynomials respectively. We investigate operation counts and performance of three methods which come from the equation reduction and Newton-Raphson method, and choose the best method.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.1944-1946
• /
• 2008

This paper analyzed orientation simulation of Stewart platform which is a parallel manipulator of 6-DOF. When platform shape had been given, inverse kinematics as the problem about length of actuator could get an answer using a vector function simply, and forward kinematics as the problem solving shape of platform through the length of actuator could get answer using repetitive and manual explaining Newton-Raphson method because it is expressed a high nonlinear polynomial expression. In addition, for control the Stewart platform it could drive simply and it could confirm the state of orientation in real-time.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.9
• /
• pp.823-829
• /
• 2012

An excitation table is commonly used for vibration and ride tests for parts or assemblies of automobiles, aircrafts, or other heavy systems. The authors have analyzed several kinematic properties of an excitation table that is under development for heavy transport vehicles. It consists of one table and 7 linear hydraulic actuators. The authors have performed mobility analysis, inverse kinematics, forward kinematics, and singularity analysis. Especially, we have proposed a fast forward kinematic solution considering the limited motion of the excitation table. On the assumption that the motion variables such as rotation angles and displacements are small, the forward kinematic problem is converted to the observer problem of a linear system. This provides a fast solution. Also we have verified that there are no singularity points in the working range by numerical analysis.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.301-303
• /
• 2018

This paper describes a robot control system and control method of a human arm type redundant manipulator. The control of a redundant manipulator suffer from computational complexity and singularity problem because of numerical inverse kinematics. To deal with such problems, analytical methods for a redundant robot arm have been researched to enhance the performance of inverse kinematics. In this research, we propose a numerical control method and weighted pseudo inverse kinematics algorithm. Using this algorithm, it is possible to generate a trajectory passing through the singular points and intuitively move the elbow without regard to the end-effector pose. Performance of the proposed algorithm was verified by various simulations. It is shown that the trajectory planning and using this algorithm provides correct results near the singular points and can utilize redundancy intuitively. We proved this system's validity through field test.

• The Journal of Korea Robotics Society
• /
• v.12 no.1
• /
• pp.42-54
• /
• 2017

Using an inverse of the geometric Jacobian matrix is one of the most popular ways to control robot manipulators, because the Jacobian matrix contains the relationship between joint space velocities and operational space velocities. However, the control algorithm based on Jacobian matrix has algorithmic singularities: The robot manipulator becomes unstable when the Jacobian matrix loses rank. To solve this problem, various methods such as damped and filtered inverse have been proposed, but comparative studies to evaluate the performance of these algorithms are insufficient. Thus, this paper deals with a comparative analysis of six representative singularity avoidance algorithms: Damped Pseudo Inverse, Error Damped Pseudo Inverse, Scaled Jacobian Transpose, Selectively Damped Inverse, Filtered Inverse, and Task Transition Method. Especially, these algorithms are verified through computer simulations with a virtual model of a humanoid robot, THORMANG, in order to evaluate tracking error, computational time, and multiple task performance. With the experimental results, this paper contains a deep discussion about the effectiveness and limitations of each algorithm.

• Korean Journal of Computational Design and Engineering
• /
• v.14 no.1
• /
• pp.18-24
• /
• 2009

This paper proposes an analytical method of evaluating the maximum error by modeling the exact tool path for the tool traverse singular region in five-axis machining. It is known that the NC data from the inverse kinematics transformation of 5-axis machining can generate singular positions where incoherent movements of the rotary axes can appear. These lead to unexpected errors and abrupt operations, resulting in scoring on the machined surface. To resolve this problem, previous methods have calculated several tool positions during a singular operation, using inverse kinematics equations to predict tool trajectory and approximate the maximum error. This type of numerical approach, configuring the tool trajectory, requires much computation time to obtain a sufficient number of tool positions in a region. We have derived an analytical equation for the tool trajectory in a singular area by modeling the tool operation into a linear and a nonlinear part that is a general form of the tool trajectory in the singular area and that is suitable for all types of five-axis machine tools. In addition, we have evaluated the maximum tool-path error exactly, using our analytical model. Our algorithm can be used to modify NC data, making the operation smoother and bringing any errors to within tolerance.

• The KIPS Transactions:PartB
• /
• v.10B no.6
• /
• pp.657-664
• /
• 2003

Reliable tracking of moving humans is essential to motion estimation, video surveillance and human-computer interface. This paper presents a new approach to human motion tracking that combines appearance-based and model-based techniques. Monocular color video is processed at both pixel level and object level. At the pixel level, a Gaussian mixture model is used to train and classily individual pixel colors. At the object level, a 3D human body model projected on a 2D image plane is used to fit the image data. Our method does not use inverse kinematics due to the singularity problem. While many others use stochastic sampling for model-based motion tracking, our method is purely dependent on nonlinear programming. We convert the human motion tracking problem into a nonlinear programming problem. A cost function for parameter optimization is used to estimate the degree of the overlapping between the foreground input image silhouette and a projected 3D model body silhouette. The overlapping is computed using computational geometry by converting a set of pixels from the image domain to a polygon in the real projection plane domain. Our method is used to recognize various human motions. Motion tracking results from video sequences are very encouraging.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1995.04a
• /
• pp.787-796
• /
• 1995

A man model is a useful design tool for the evaluation of man machine systems and products. An arm reach trajectory prediction for such a model will be specifically useful to present human activities and, consequently, could increase the accuracy and reality of the evaluation. In this study, a three-dimensional reach trajectory prediction model was developed using an inverse kinematics technique. The upper body was modeled as a four link open kinematic chain with seven degrees of freedom. The Resolved Motion Method used for the robot kinematics problem was used to predict the joint movements. The cost function of the perceived discomfort developed using the central composite design was also used as a performance function. This model predicts the posture by moving the joints to minimize the discomfort on the constraint of the end effector velocity directed to a target point. The results of the pairwise t-test showed that all the joint coordinates except the shoulder joint's showed statistically no differences at .alpha. = 0.01. The reach trajectory prediction model developed in this study was found to accurately simulate human arm reach trajectory and the model will help understand the human arm reach movement.

• Journal of Korean Institute of Industrial Engineers
• /
• v.25 no.3
• /
• pp.382-392
• /
• 1999

Industrial robots have increased in both the number and applications in today's material handling systems. However, traditional approaches to robot controling have had limited success in complicated environment, especially for real time applications. One of the main reasons for this is that most traditional methods use a set of kinematic equations to figure out the physical environment of the robot. In this paper, a neural network model to solve robot manipulator's inverse kinematics problem is suggested. It is composed of two Self-Organizing Feature Maps by which the workspace of robot environment and the joint space of robot manipulator is inter-linked to enable the learning of the inverse kinematic relationship between workspace and joint space. The proposed model has been simulated with two robot manipulators, one, consisting of 2 links in 2-dimensional workspace and the other, consisting of 3 links in 2-dimensional workspace, and the performance has been tested by accuracy of the manipulator's positioning and the response time.