Flocking of Multi-Agent Systems with Limited Rotating Field of Views

In this paper, the flocking problem in a network of double-integrator agents with limited angular fields of view (FOV) is investigated. The angular FOV represents the limitation imposed on the velocity sensor mounted on every agent in the network, while the position sensors are assumed to be omnidirectional. To increase the sensing capabilities of agents and preserve network connectivity, it is assumed that the FOV of all agents rotate with constant angular velocity. The problem is formulated in the framework of switched nonlinear systems to address the switching topology of the network. The control inputs are subsequently designed as a combination of alignment and attractive/repulsive forces, such that the velocity vectors reach consensus and the inter-agent collision is avoided. The asymptotic convergence of the network configuration is guaranteed under the proposed controller. Simulations confirm the efficacy of the controller.