Feedforward Formation Control based on Self-Organized Body-Schema

Abstract

In this paper we propose a new architecture of the self-organized body schema (SO-BoS) system capable of learning the configuration space of three rotorcrafts relative to a single sensor space. The SO-BoS cortical map architecture represents the posterior parietal cortex where sensory-motor of the human posture is synthesized (circular-reaction); we inspired on such process to control the displacement in formation of three rotorcraft UAVs. The SO-BoS-based feedforward control drives the aerial robots formation towards the desired position while avoiding obstacles and inter-agent collisions. The proposed strategy is a promising approach for aerial vehicles systems due to the plasticity resulting from the learning-babbling stage. A numerical result section is given to provide the inherent discussions to assess the effectiveness of the proposed intelligent navigation scheme.