Distributed Formation Control for Multiagent Systems in the Presence of External Disturbances

Xiaowu Yang,Xiaoping Fan

doi:10.1109/access.2019.2923852

Abstract

This paper analyzes the distributed formation control algorithms that rely on local measurements for second-order multiagent systems under the leader-follower control structure. Here, the reference states are only available to the leader agents, the follower agents have access to the information from the leaders to achieve the control objective, and the target formation is defined by their relative positions. The formation control problem with constant and time-varying reference states is studied, and two classes of practical control system problems in the presence external disturbances are also considered. In the former case, a non-saturated Proportional-integral (PI) controller is proposed to attenuate the constant disturbances, namely, the static errors are eliminated. In the latter case, a saturated distributed control algorithm is first proposed to address the formation control of multiagent systems, which, in many practical applications, is subject to bounded control inputs. To eliminate the effect of chattering caused by external bounded time-varying disturbances, the distributed formation control algorithm is then proposed. The analysis for the proposed control algorithms according to Lyapunov stability theory is provided. A sufficient collision-free condition is also presented. Finally, to demonstrate the effectiveness of the obtained theoretical results, the numerical simulation results of several illustrative examples are presented.

Highlights

Formation control for multiagent systems (MASs) has been extensively studied because its prominent advantages with respect to redundancy, flexibility, robustness and fault tolerance, which can support feasible control algorithms by employing distributed sensing and actuation to achieve the specific control objectives [1]–[8]
Motivated by the above discussion, we mainly focus on the distributed formation control algorithms for MASs in the presence of bounded unknown constant and time-varying disturbances, and a saturated formation control algorithm is analyzed
With the above results in hand, we investigate the formation control algorithm for multiagent systems in the presence of constant disturbances

Summary

INTRODUCTION

Formation control for multiagent systems (MASs) has been extensively studied because its prominent advantages with respect to redundancy, flexibility, robustness and fault tolerance, which can support feasible control algorithms by employing distributed sensing and actuation to achieve the specific control objectives [1]–[8]. In [35], the authors studied the time-varying formation tracking control problem for high-order multiagent systems in the presence of heterogeneous nonlinearity and disturbance. The study of distributed robust formation control is meaningful and interesting for multiagent systems under the directed interaction information topology. This problem is practical and further required to investigated extensively. Motivated by the above discussion, we mainly focus on the distributed formation control algorithms for MASs in the presence of bounded unknown constant and time-varying disturbances, and a saturated formation control algorithm is analyzed.

BACKGROUND

COLLISION AVOIDANCE CONSTRAINTS

SIMULATION EXAMPLES

CONCLUSION