Department of Computer, Control and Management Engineering

Omnidirectional robots can be realized using Mecanum wheels or using a suitable arrangement of conventional steerable wheels. The latter group, known as omnidirectional steerable wheeled mobile robots (SWMRs), are known to have a lower cost with respect to the former, and to be more robust thanks to the presence of conventional wheels. Nevertheless, their modeling and control is complex, due to the presence of singularities in their representation. This paper proposes a framework for trajectory tracking of SWMRs using Nonlinear Model Predictive Control (NMPC) based on a real-time iteration scheme. The NMPC generates feasible motions for the robot, avoiding model singularities of the mobile base, together with bounds on driving and steering velocities on the wheels. Our NMPC works alongside a finite state machine, responsible for singularity avoidance during starting and stopping motion, and a state trajectory generation scheme based on dynamic feedback linearization, which makes our framework capable of tracking any trajectory. Our approach is validated on a Neobotix MPO-700 on various trajectories.