Page 113 - Mechatronics for Safety, Security and Dependability in a New Era

Ch21-I044963.fm Page 97 Tuesday, August 1, 2006 3:30 PM
Tuesday, August
3:30 PM
1, 2006
Page 97
Ch21-I044963.fm
97 97
METHOD FOR SOLVING INVERSE KINEMATICS OF
REDUNDANT ROBOT UNDER RESTRAINT BY OBSTACLES

J. Kawamoto, K. Tashiro, M. Takano and S. Aoyagi

Systems Mangement Engineering, Kansai University
3-3-35, Yamate-cho, Suita, Osaka 564-8680, Japan

ABSTRACT

For the purpose of collision avoidance, a new efficient method for solving inverse kinematics of a
redundant robot (all joints are rotational) is proposed. This method is divided into three procedures,
which are position synthesis by 3 joints, orientation synthesis by 3 wrist joints, and collision avoidance
by other remaining joints. Each of these three procedures can be calculated analytically. After series of
three procedures, the positioning error is occurred, since they are not independent from each other.
Therefore, the calculation is carried out iteratively until sufficient convergence is obtained. A simulator
based on this method is developed. A robot of 14 DOF can successfully pass through two cylindrical
holes in two thick walls, while realizing high positioning and orientating accuracy of its end effector.

KEYWORDS

Redundant robot, Path planning, Simulator, Collision avoidance

INTRODUCTION
For the purpose of collision avoidance, a redundant robot with many rotational joints is necessary. The
transition of configuration of this redundant robot, which is called simply as "path" in this paper,
should be planned so as that it can avoid obstacles and realize desired goal position and orientation of
its end effector. Several researches are carried out to investigate how to decide a path of a redundant
robot. In these researches, among many possible paths, one solution path is selected generally by
minimizing some evaluation function [1, 2\. Here, it is focused how to use redundant DOF (Degree of
freedom) effectively. On the other hand, this research assumes the situation that the robot
configuration is firmly restrained by obstacles, and all DOF are used for realizing goal pose (position
and orientation) of the end effector and avoiding these obstacles, namely there are no excess DOF.
Under this condition, the path planning is not important issue, instead, the method how to decide the
joint angles which satisfy the goal pose of the end effector and the restrained (specified) configuration
for avoiding the obstacles. Namely, it is necessary to solve inverse kinematics (synthesis) of the robot.
In this research, an effective calculation method to solve this synthesis problem under the restraint by
obstacles is newly proposed.