The problem of fault detection and isolation (FDI) in cooperative manipulators is addressed in this paper. Four FDI procedures are developed to deal with free-swinging joint faults, locked joint faults, incorrectly measured joint position, and incorrectly measured joint velocity. Free-swinging and locked joint faults are isolated via neural networks. For each arm, a Multilayer Perceptron (MLP) is used to reproduce the dynamics of the fault-free robot. The outputs of each MLP are compared to the actual joint velocities in order to generate a residual vector which is then classified by an RBF network. The remaining faults are isolated based on the kinematic constraints imposed on the cooperative system. Results obtained via simulations and via an actual cooperative manipulator robot are presented.
Fault Detection; Fault Isolation; Robotic Manipulators; Co-operation; Neural Networks
