Decentralized second-order adaptive protocol of multi-agent systems with nonlinear disturbance based on only onboard partial measurement

Abstract

The decentralized second-order adaptive protocol of uncertain multi-agent systems with only onboard partial measurement is considered in this research work. It is assumed that only the onboard relative position measurement between agents is accessible and no agent has access to its absolute and relative velocity with respect to its neighbors. Moreover, we suppose that the wireless communication is not available and all relative positions are measured by onboard sensors. Therefore, each agent should estimate its own as well as the leader unknown dynamics. A new adaptive control algorithm is presented for all followers and afterwards, the uncertain closed-loop dynamics is derived. By utilizing the Lyapunov theorem, the asymptotic stability of the error dynamics will be proved under the proposed consensus algorithm in presence of unknown nonlinear dynamics. It will be confirmed that under the persistently exciting condition, the exact identification will achieve and the estimations of unknown parameters will converge to their actual values. The merits of the adaptive control method are demonstrated by providing different numerical and experimental results.

Keywords: Adaptive consensus; Decentralized leader-following network; Nonlinear unknown dynamics; Onboard relative position measurement; multi-agent system (MAS).