Quasi-Static Analysis on Transoral Surgical Tendon-Driven Articulated Robot Units

Wire actuation in tendon-driven continuum robots enables the transmission of force from a distance, but it is understood that tension control problems can arise when a pulley is used to actuate two cables in a push-pull mode. This paper analyzes the relationship between angle of rotation, pressure, as well as variables of a single continuum unit in a quasi-static equilibrium. The primary objective of the quasi-static analysis was to output pressure and the analysis, given the tensions applied. Static equilibrium condition was established, and the bisection method was carried out for the angle of rotation. The function for the bisection method considered pressure-induced forces, friction forces, and weight. {\theta} was 17.14{\deg}, and p was 405.6 Pa when Tl and Ts were given the values of 1 N and 2 N, respectively. The results seemed to be consistent with the preliminary design specification, calling for further simulations and experiments.

Deep reinforcement learning for guidewire navigation in coronary artery phantom

In percutaneous intervention for treatment of coronary plaques, guidewire navigation is a primary procedure for stent delivery. Steering a flexible guidewire within coronary arteries requires considerable training, and the non-linearity between the control operation and the movement of the guidewire makes precise manipulation difficult. Here, we introduce a deep reinforcement learning(RL) framework for autonomous guidewire navigation in a robot-assisted coronary intervention. Using Rainbow, a segment-wise learning approach is applied to determine how best to accelerate training using human demonstrations with deep Q-learning from demonstrations (DQfD), transfer learning, and weight initialization. `State' for RL is customized as a focus window near the guidewire tip, and subgoals are placed to mitigate a sparse reward problem. The RL agent improves performance, eventually enabling the guidewire to reach all valid targets in `stable' phase. Our framework opens anew direction in the automation of robot-assisted intervention, providing guidance on RL in physical spaces involving mechanical fatigue.