Integrated Trajectory Planning and Fault Detection for a Skid-Steered Mobile Robot deployed in Post-Disaster Scenarios

The development of intelligent response systems for first aid and rescue operations after natural disasters is a rapidly growing field of research. In this paper, we present the ongoing research in the project 'CHeMSys: Cooperative Heterogeneous Multi-drone System for disaster prevention and first response', which addresses the problem of detecting faults that affect the capability of a skid-steered tracked mobile robot in following an assigned trajectory. To achieve this, the dynamics of the trajectory tracking error are reformulated in terms of an uncertain model with norm-bounded uncertainty that accounts for delays in the control loop and possible skidding and slipping phenomena. First, the optimal feasible trajectory for the mobile robot is planned. Then, using some set-based mathematical properties of the feasible trajectory, a multivariate logic with low computational complexity is used to detect the faults and a reconfiguration strategy is applied in case of a sensor fault. The algorithm has been shown to be effective in distinguishing actuator faults from faults in the robot's measurement system and is based on set-theoretic arguments. To demonstrate the effectiveness of the proposed solution, we present the results of several numerical simulations performed with a simulator implemented in the MATLAB/Simulink environment.