Nonlinear event-triggered tracking control of a mobile robot: design, analysis and experimental results

We consider the scenario where a controller communicates with a mobile robot via a network. Our objective is to design the control law such that the robot tracks a given reference trajectory while reducing the usage of the communication channel. For that purpose, we design a nonlinear event-triggered feedback law. We follow an emulation-like approach in the sense that we first synthesize the controller while ignoring the communication constraints and we then derive an appropriate triggering condition. We prove that the states of the robot's model practically converge towards the states of the reference system which generates the desired trajectory to be tracked, using an invariance principle for hybrid systems. The existence of a dwell-time between any two transmissions is discussed. We have implemented the proposed strategy on a benchmark where the controller is located on a remote computer which communicates with the mobile robot via a IEEE 802.11g wireless network. The proposed event-triggering strategy is shown to significantly reduce the need for communication compared to a classical time-triggered setup while ensuring similar, if not better, tracking performances.