Multi-sensor, multi-level data fusion and behavioral analysis based fault detection and isolation in mobile robots

This paper presents a multi-sensor fusion based fault detection and isolation (FDI) method for robotic system. The proposed approach is independent of analytical model which suits dynamic, uncertain and unstructured nature of the robot operating environment. In contrast to single-sensor data-driven FDI methods, incorporation of multi-sensor fusion architecture makes the proposed FDI scheme more robust and reliable. The proposed scheme incorporates stages including pre-processing, local-data fusion, change detection, credibility computation, decision-level information fusion and fault isolation. The proposed methodology is applied on a simulated multiple robots navigating the environment. The performance of developed FDI method is examined under different types of IR and camera faults. With this novel fusion based FDI method; superior fault detection, isolation and tolerance capabilities are achieved in mobile robot system.