Task-Driven Formation of Nonholonomic Vehicles With Communication Constraints

Abstract

This work focuses on developing a task-driven formation framework in the presence of communication constraints for a group of nonholonomic mobile vehicles. The center of this framework is a cost function that characterizes both the task-related and the communication constraint related costs followed by a newly defined concept of compatible task and communication constraints. It is shown that, if the task and communication constraints are compatible as defined, then the formation problem can be solved as an optimization for the given cost function through a family of formation control laws obtained from off-the-shelf optimization algorithms. In particular, the gradient-like control laws can be developed to find non-unique local optimal solutions of this optimization problem under some sufficient conditions and, hence, achieve the desired formation for the underlying group of mobile vehicles with communication constraints. In order to demonstrate the effectiveness of this framework, an illustrative example is presented for the formation task of a group of vehicles under communication constraints from the field of view (FOV) of visual camera sensors. It can be observed that, although cameras are the only ways of communication, the proposed gradient-like control law can drive the group of vehicles to achieve the desired formation with guaranteed collision avoidance.