Distributed Formation Algorithm Based on Integrated Localization and Control

Collaborative formation is an important basis for multi-agent collaborative systems to complete advanced tasks. Previous studies show that the integrated framework of localization and control could achieve higher accuracy than the decoupled framework. We follow the integrated framework and design a feasible distributed formation algorithm based on relative measurements, which has near-optimal performance. Besides, we prove that if the measurement network topology is a connected regular graph, the proposed algorithm is asymptotically stable and an upper bound of the mean formation error is provided, which reveals how localization error and control error affect formation accuracy. Numerical results verify the performance of the proposed algorithm as well as the effects of the noise parameters and network topology on formation performance.