Adaptive Neural Network Fault-Tolerant Control of Continuum Robots Using Cosserat Rod Model

This research proposes an adaptive neural network fault-tolerant control strategy utilizing the Cosserat rod model for continuum robots. The existing control schemes assume that the precise spatial derivative of the continuum robot’s internal force can be obtained directly. However, this assumption is difficult to satisfy in real applications due to the complex environment and the limitations of sensors. Hence, we tackle this challenge by approximating the internal force’s spatial derivative with a radial basis function neural network. The safety and reliability of the control scheme are further improved by adaptive compensation. In the simulation, the proposed control scheme is compared with sliding mode control and adaptive neural network sliding mode control schemes. The outcomes of the simulation show that the proposed scheme successfully achieves trajectory tracking without satisfying the assumption above and has certain advantages under the actuator fault.