Hybrid Visual Servoing of Tendon-Driven Continuum Robots

Abstract This article introduces a novel hybrid visual servoing (HVS) framework and evaluates it on a tendon-driven continuum robot as a challenging nonlinear robotic platform. The HVS combines image-based visual servoing (IBVS) with deep learning-based visual servoing (DLBVS) to overcome the limitations of each method and improve overall performance. IBVS offers higher accuracy and faster convergence in feature-rich environments, while DLBVS enhances robustness against disturbances and offers a larger workspace. By enabling transitions between IBVS and DLBVS, the proposed HVS provides effective control under the tested experimental conditions and disturbances. The effectiveness of this approach is validated through simulations and real-world experiments, demonstrating that HVS achieves reduced iteration time, faster convergence, lower final error, and smoother performance compared to DLBVS alone, while maintaining DLBVS’s robustness in challenging conditions such as occlusions, lighting changes, actuator noise, and physical impacts.