Enhanced decentralized robust adaptive control of robots with arbitrarily-switched unknown constraints

In this article, we propose an Enhanced Decentralized Robust Adaptive Control (EDRAC) strategy that guarantees global stable performance for robots under arbitrarily-switched and unknown constraints. The EDRAC strategy is a synergy of the concepts of the decentralized control, finding a Common Lyapunov Function (CLF), and the Sliding Mode Control (SMC). The decentralization in the control action is realized through decomposing the joints dynamics into two parts; a function of the specific joint variables and another one in term of the all joints variables. The first term is assumed to be known and the latter is compensated through adapting its upper bound in the control loop hence relaxing the need for knowing part of the robot dynamics and simplifying the control strategy. The CLF guarantees global stable performance under arbitrary constraints switching, and the SMC makes the EDRAC strategy robust against possible parameters drift. Simulation is performed for a two link robot interacting with two switching constraints and the efficiency of the EDRAC strategy is shown.