Sliding mode control with gain-scheduled and improved boundary layer for nonholonomic multi-robot formation

In this paper, we have proposed a sliding mode control law with gain-scheduled and improved boundary layer for leader-follower based formation control. The switching gains of sliding mode controller are required to change with the absolute value of sliding mode function. Specifically, a larger switching gain should be adopted in order to ensure rapid convergence, while a smaller one should be adopted to suppress the chattering. In addition, an improved boundary layer method is applied such that the chattering can be further eliminated. Both of the simulation and experimental results show that the proposed control law weakens the chattering of the system, reduces the system steady-state error and guarantees the response speed of the system.