Bipartite Time-Varying Prescribed Range Formation Control for a Class of Nonlinear Multiagent Systems With Unknown Disturbances

Yang Yang,Didi Chen,Qidong Liu,Sergey Gorbachev,Victor Kuzin
DOI: https://doi.org/10.1109/jsyst.2022.3215853
IF: 4.802
2022-01-01
IEEE Systems Journal
Abstract:Formation control is a hot topic in multiagent systems (MASs). In this article, a bipartite time-varying prescribed range formation is developed for a class of nonlinear MASs with unknown disturbances. The time-varying prescribed range formation means that agents achieve a formation shape within a predefined region relative to their neighbors, and the region changes along with the time-varying formation distance. An error transformation is introduced to transform this bipartite prescribed range formation problem into a stabilization control one. A surface-error-based predictor is developed to generate prediction errors for update of neural networks (NNs), and this form decouples approximation of system dynamics and control signal. With prescribed range information and signed graph, a disturbance observer with predictor-based NNs is constructed to compensate external disturbances and NNs' approximation errors. Also, a normalization learning method is employed to reduce the number of NN's learning parameters. With Lyapunov-based stability analysis, it is proven that a bipartite time-varying formation is achieved within a prescribed range. Simulation results with a group of quadrotors verify the effectiveness of the proposed bipartite control scheme.
computer science, information systems,telecommunications,engineering, electrical & electronic,operations research & management science
What problem does this paper attempt to address?