A Fault Tolerance Framework for Cooperative Robotic Manipulators
In this chapter we present a fault tolerance framework for cooperative manipulators. The fault detection and isolation (FDI) system uses the kinematic constraints of the cooperative system and neural networks to detect and treat four categories of faults: free-swinging joint faults (FSJF), where one or more joints lose actuation and become free-swinging; locked joint faults (LJF), where one or more joints become locked; incorrectly-measured joint position faults (JPF), where the measurement of one or more joint positions is incorrect; and incorrectly-measured joint velocity faults (JVF), where the measurement of one or more joint velocities is incorrect. After the faults are detected by the FDI system, the control system is reconfigured according to the nature of the isolated fault and the manipulation task is resumed.
KeywordsRadial Basis Function Network Joint Position Residual Vector Cooperative System Joint Velocity
- 11.McIntyre ML, Dixon WE, Dawson DM, Walker ID (2005) Fault identification for robot manipulators. IEEE Trans Robot Autom 21(5):1028–1034Google Scholar
- 13.Monteverde V, Tosunoglu S (1999) Development and application of a fault tolerance measure for serial and parallel robotic structures. Int J Modeling Simulat 19(1):45–51Google Scholar
- 16.Tinós R, Terra MH, Bergerman M (2001) Fault detection and isolation in cooperative manipulators via artificial neural networks. In: Proceedings of IEEE Conference on Control Applications, Mexico City, Mexico, pp 988–1003Google Scholar