Adapting Neural-Based Models for Position Error Compensation in Robotic Catheter Systems

Robotic catheter systems with master–slave designs are employed for teleoperated navigation of flexible endovascular tools treating calcified lesions. Despite improved tool manipulation techniques, patient safety and lowering operative risks remain top priorities. Therefore, minimizing undesirable drifts imprecise during intravascular catheterization is essential. In the current designs, finite displacement lag between position command actual action at slave device affects smooth catheterization. this study, we designed developed a compact 2-DOF robotic system characterized influence step values, velocity, motion gap on error device. For uniform varying commands from master platform, results indicate that overall increases distance traveled respectively. Hence, proposed using recurrent neural networks—long short-term memory gated unit controllers to predict robot’s appropriate compensation value per translation step. An analysis in-silico studies CoppeliaSim showed neural-based can ensure mapping devices. Furthermore, implemented models within RCS length 120 mm. The result demonstrates suitably aid stepwise displacement. Thus, help match translational precise by could contribute alleviating patients’ concerns interventions.