Search papers, labs, and topics across Lattice.
This paper addresses the challenges of modeling and controlling Tendon-Driven Continuum Robots (TDCRs) by introducing a differentiable learning framework that combines high-fidelity dynamics modeling with robust neural control. The authors develop a GRU-based dynamics model that utilizes bidirectional multi-channel connectivity and residual prediction to mitigate compounding errors during long-horizon predictions. Experimental results show that their approach not only achieves accurate tracking but also enhances robustness against unseen payloads, significantly outperforming traditional Jacobian-based methods by eliminating self-excited oscillations.
A novel GRU-based dynamics model for tendon-driven robots eliminates self-excited oscillations, achieving superior robustness and accuracy in control.
Tendon-Driven Continuum Robots (TDCRs) pose significant modeling and control challenges due to complex nonlinearities, such as frictional hysteresis and transmission compliance. This paper proposes a differentiable learning framework that integrates high-fidelity dynamics modeling with robust neural control. We develop a GRU-based dynamics model featuring bidirectional multi-channel connectivity and residual prediction to effectively suppress compounding errors during long-horizon auto-regressive prediction. By treating this model as a gradient bridge, an end-to-end neural control policy is optimized through backpropagation, allowing it to implicitly internalize compensation for intricate nonlinearities. Experimental validation on a physical three-section TDCR demonstrates that our framework achieves accurate tracking and superior robustness against unseen payloads, outperforming Jacobian-based methods by eliminating self-excited oscillations.
