Behavioral control based adaptive bandwidth allocation in a system of Unmanned Ground Vehicles

As the number of nodes in large scale network control systems increases, allocating constant bandwidth to all the nodes is neither practical nor efficient in terms of quality of control because of the limited amount of available bandwidth. In order to resolve this issue of bandwidth management in a networked control system consisting of several control loops, dynamic methods for resource allocation should be used. Bandwidth allocation techniques like Large Error First (LEF) and some auction based techniques exist which have approached this problem from the perspective of efficient bandwidth usage, and from the perspective of performance improvement in the control loop. In this paper, we present a novel Behavioral Control (BC) based adaptive bandwidth allocation method in a fleet of Unmanned Ground Vehicles (UGV). The BC algorithm uses the UGV's speed and trajectory relative to the future path to allocate the available bandwidth. Simulation results have shown that the performance, in terms of the trajectory tracking error and tracking time, has improved up to 22% and 9% resp. when using the BC algorithm compared to existing bandwidth allocation algorithms.