Motion Planning and Robust Control for the Endovascular Navigation of a Microrobot

Microrobots show great targeted-delivery potential in precision medicine. This article presents the use of a model-free approach to the navigation control of a microrobot in the cardiovascular environment. With the proposed approach, the microrobot can adapt to the non-Newton behavior of blood and environmental disturbances when it moves in blood vessels without knowledge of blood-velocity distribution. The referred trajectory of the navigated microrobot is generated by using a breadth-first search and genetic algorithm, aiming to minimize the energy consumption. The proposed navigation controller combines sliding mode control, backstepping control, and disturbance compensation. A high-gain extended state observer is designed to estimate and reject the uncertainties of model parameters and environmental disturbances. Simulations are performed to demonstrate the effectiveness of the proposed approach in a small artery compared with other control methods.