Cooperative control of multi-agent systems with limited angular field of view
Mohammad Mehdi Asadi, Amir Ajorlou, Amir G. Aghdam · 2012
This paper is concerned with the problems of consensus and containment in a network of directed single-integrator agents with some practical constraints in sensing other agents. Limitations in field of view (FOV), which are considered in this work, are very common in vision-based sensing/communication devices (such as cameras or directed antennas). The set of neighbors for each agent depends on its angular position, which varies by time. This poses new challenges to the cooperative control of agents, where it is desired to achieve objectives such as consensus and containment. The problem is formulated in the context of impulsive switched systems, and a controller is proposed with a switching law which ensures the connectivity of the network at all times, except for some limited time instants. A controller is subsequently designed to achieve consensus using the concept of Laplacian-based control. Finally, two subgroups of agents are defined as static leaders and moving followers, and the controller is modified properly to achieve containment in this case asymptotically.