Robust Collaborative Team Formation Control of Hybrid Teams of Biped Robots and Wheeled Vehicles Under External Disturbance and Communication Interactions

In this study, a collaborative team formation system of hybrid teams of biped robots and tractor-trailers is designed for some common task. The collaborative team formation system consists of a number of hybrid team formation subsystems. Each hybrid team formation subsystem is composed of a tractor-trailer as the leader and a group of biped robots as followers. For the convenience of collaborative team formation design, novel reference models are proposed to generate the desired collaborative time-varying leader-follower team formation for the cooperative tractor-trailers and biped robots in each hybrid team. Accordingly, the collaborative team formation design problem is simplified to a set of independent robust model reference tracking design problems for each agent under external disturbances and communication couplings from other agents. Subsequently, the robust
$H_{\infty }$
decentralized collaborative team formation tracking control strategy is proposed to achieve the independent team formation tracking control for each agent and efficiently attenuate the effect of external disturbance and communication coupling from other agents on the team formation tracking performance of each agent. In order to avoid solving the difficult Hamilton-Jacobi inequality (HJI) of the robust
$H_{\infty }$
decentralized collaborative team formation tracking control for each agent, the numerical linear parameter-varying (LPV) modeling method is employed to approximate the nonlinear team formation tracking error dynamic system of each agent such that the HJI can be equivalently transformed to a set of linear-matrix inequalities (LMIs) which can be efficiently solved by MATLAB LMI TOOLBOX. Since the robust
$H_{\infty }$
decentralized collaborative team formation tracking control scheme is employed based on local information, the collaborative team formation system can be extended to very large-scale biped robots and tractor-trailers. Finally, a simulation example of robust
$H_{\infty }$
decentralized collaborative team formation tracking control design of three collaborative teams, in which each hybrid team consists of one tractor-trailer as the leader and six biped robots as followers, is given to compare with the optimal
$H_{2}$
decentralized method to verify the effectiveness of the proposed scheme.