Mobile Robot Self-diagnosis with a Bank of Adaptive Particle Filters
The diagnosis of mobile robot faults is one of the most serious problems which have to be overcome if one considers applications of mobile robotics in real life, outside laboratories. It would be desirable to perform the diagnosing routine in parallel with the standard activity of the robot, e.g., navigation, but without generating additional computational overhead. Recently the particle filter has become a very popular tool for state estimation of mobile robots. This is because it makes it easier to solve, e.g., the simultaneous localization and mapping problem. One of the biggest drawbacks of the method is its high computational burden closely related to the number of particles used. Therefore, it is often necessary to work out a compromise between the computational time and the quality of results. This work proposes a fault diagnosis system for a mobile robot which is based on a bank of adaptive particle filters. The idea behind is to reduce the total number of particles used in state estimation through activating and deactivating individual filters when needed, as well as by adapting the number of particles in each filter.
KeywordsMobile Robot Fault Detection Fault Diagnosis Nonlinear Stochastic System Fault Isolation
Unable to display preview. Download preview PDF.
- 1.Andrieu, C., Doucet, A., Singh, S.S., Tadić, V.: Particle methods for change detection, system identification, and control. In: IEEE Proc. (2004)Google Scholar
- 3.de Freitas, N., Dearden, R., Hutter, F., Morales-Menendez, R., Mutch, J., Poole, D.: Diagnosis by a waiter and a mars explorer. In: IEEE Proc. (2004)Google Scholar
- 4.Goel, P., Dedeoglu, G., Roumeliotis, S., Sukhatme, G.: Fault detection and identification in a mobile robot using multiple model estimation and neural network. In: Proc. IEEE Int. Conf. Robotics and Automation (2000)Google Scholar
- 5.Gordon, N., Salmond, D., Smith, A.: Novel approach to nonlinear/non-gaussian bayesian state estimation. In: IEEE Proc.–F, vol. 35, pp. 107–113 (1993)Google Scholar
- 6.Hutter, F., Dearden, R.: The gaussian particle filter for diagnosis of non-linear systems. In: Proc. 14th Int. Conf. Principles of Diagnosis, Washington D.C., USA (2003)Google Scholar
- 9.Kozłowski, K., Pazderski, D.: Modeling and control of a 4-wheel skid-steering mobile robot. Int. J. Appl. Math. Comput. Sci. 14(4) (2004)Google Scholar
- 15.Plagemann, C., Stachniss, C., Burgard, W.: Efficient failure dtection for mobile robots using mixed abstraction particle filters, vol. 22, pp. 93–107 (2006)Google Scholar
- 16.Straka, O., Šimandl, M.: Particle filter adaptation based on efficient sample size. In: Proc. 14th IFAC Symp. System Identification, pp. 991–996 (2006)Google Scholar
- 17.Straka, O., Šimandl, M.: A survey of sample size adaptation techniques for particle filters. In: Proc. 15th IFAC Symp. System Identification, vol. 15 (2009)Google Scholar
- 20.Zając, M., Uciñski, D.: Adaptive particle filter in fault detection and isolation of mobile robots. Pr. Nauk. Polit. Warsz., Elektronika 175, 635–644 (2010)Google Scholar