Adaptability of a Decentralized Kinematic Control Algorithm to Reactive Motion under Microgravity

Abstract

Adaptation to partial failure is one of the most important topics in space robotics, since space robots cannot be repaired after they have been launched. We recently proposed a decentralized autonomous control algorithm to realize adaptation to partial failure using parallel processing with very low-performance processors. Another important consideration in space robotics is the control of dynamical motion under micro-gravity conditions. Our control algorithm mainly focuses on kinematic-based control, since dynamic-based control is considered too complex for our limited CPU system to calculate in real time. It is important to assess the ability of our simple algorithm to adapt to such dynamic phenomenon not only to determine the fault-adaptability of the algorithm, but also to extend the algorithm to a flexible arm. In this study, we assess our decentralized autonomous control algorithm in terms of its adaptability to dynamic phenomenon, such as the reaction to the motion of the manipulator itself caused by failure in the braking system.