Muhammad T. Khan, Muhammad U. Qadir, Anam Abid, Fazal Nasir, and Clarence W. de Silva
Fault detection, negative selection algorithm, artificial immunesystem
Fault detection is a challenging problem in complex autonomous systems like robots. For reliable operation, faults have to be detected quickly and accurately. This paper presents an immune-inspired fault-detection method based on negative selection theory of self-/non-self-discrimination. In the methodology proposed here, a generic fault-detection method based on an artificial immune system is presented. It is shown that the developed scheme can be employed for both detection and identification of multiple faults. The method is applied to anomaly detection in sensors of robots. The developed methodolgy is validated by implementing it on a mobile robot in a simulated environment. The results are shown to support the developed methodology.
Important Links:
Go Back
