An efficient approach for inverse kinematics and redundancy resolution of spatial redundant robots for cluttered environment

Abstract

This paper presents an optimization-based approach for solving the inverse kinematics problem of spatial redundant manipulators in cluttered workspaces. To provide better flexibility and manipulability in the narrow regions, the joints were modeled with multiple degrees of freedom (DOF) and are considered as universal joints. Each universal joint has two orthogonal DOF, which are made by pitch axis and yaw axis. The inverse kinematic (IK) problem is multimodal by nature, and it has multiple solutions. A global search and multi-start framework have been implemented to determine the multiple kinematic configurations for a given task location. The characteristic feature of the IK problem has multiple configurations for a given task space location. The process of determining the best from multiple solutions is called redundancy resolution. Secondary criteria such as joint distance minimization and collision avoidance have been chosen to perform the task of redundancy resolution. A classical constrained optimization technique has been implemented to perform the tasks of inverse kinematics and redundancy resolution. The collision avoidance scheme was implemented with a collision detection algorithm by using a bounding box approach. Simulations were performed for 9-DOF spatial manipulators with 3D obstacles in the workspace. Results are reported on IK, multiple IK solutions, and redundancy resolution of the robot in an unconstrained and cluttered environment. Results show that the proposed method is accurate and computationally efficient in determining the IK solution of spatial redundant manipulators in a multi-obstacle and restricted environment.