Abstract

The problem of designing bounded distributed connectivity preserving control strategies for multi-agent systems is studied in this work. In distributed control of multi-agent systems, each agent is required to measure some variables of other agents, or a subset of them. Such variables include, for example, relative positions, relative velocities, and headings of the neighboring agents. One of the main assumptions in this type of systems is the connectivity of the corresponding network. Therefore, regardless of the overall objective, the designed control laws should preserve the network connectivity, which is usually a distance-dependent condition. The designed controllers should also be bounded because in practice the actuators of the agents can only handle finite forces or torques. This problem is investigated for two cases of single-integrator agents and unicycles, using a novel class of distributed potential functions. The proposed controllers maintain the connectivity of the agents that are initially in the connectivity range. Therefore, if the network is initially connected, it will remain connected at all times. The results are first developed for a static information flow graph, and then extended to the case of dynamic edge addition. Connectivity preservation for problems involving static leaders is covered as well. The potential functions are chosen to be smooth, resulting in bounded control inputs. These functions are subsequently used to develop connectivity preserving controllers for the consensus and containment problems. Collision avoidance is investigated as another relevant problem, where a bounded distributed swarm aggregation strategy with both connectivity preservation and collision avoidance properties is presented. Simulations are provided throughout the work to support the theoretical findings.