Augmented Feasibility Maps: A Simultaneous Approach to Redundancy Resolution and Path Planning

Abstract

Redundant robotic manipulators are capable of performing complex tasks with an unprecedented level of dexterity and precision. However, their redundancy also introduces significant computational challenges, particularly in the realms of redundancy resolution and path planning. This paper introduces a novel approach to simultaneously address these challenges through the concept of Augmented Feasibility Maps, by integrating task coordinates as decision variables into the traditional feasibility maps. We validate the AFM concept by using Rapidly-Exploring Random Trees to explore the maps, demonstrating its efficacy in simulations of various dimensionalities. The method is capable of incorporating kinematic constraints, such as obstacle avoidance while adhering to joint limits.