An Optimization Based Inverse Kinematics of Redundant Robots Avoiding Obstacles and Singularities

Abstract

Redundant manipulators are characterized by a high number of degrees of freedom (DOF) than the required number to perform a given task. This additional DOF of the robot enhances it to work in the cluttered environment by avoiding obstacles and provides improved dexterity while performing a given task. Inverse Kinematics (IK) of redundant manipulators has infinite solutions. Among these infinite solutions, only those solutions are preferred which fulfill the criteria such as joint distance minimization, singularity avoidance, and joint torque minimization. This paper focuses on the IK of redundant manipulators for a given path with secondary objectives as performance criteria. The IK problem is formulated as an optimization problem by choosing the joint distance and singularity avoidance as objectives and collision with obstacles in the workspace as constraints. Simulations have been performed on serial redundant manipulators by varying different types of obstacles and their positions in the workspace. Results are also reported on redundancy resolution of serial manipulators based on singularity avoidance criterion.