Closed Loop Control of an MR-Conditional Robot with Wireless Tracking Coil Feedback

Modeling and Control of an MR-Safe Pneumatic Radial Inflow Motor and Encoder (PRIME)

Magnetic resonance (MR) conditional actuators and encoders are the key components for MR-guided robotic systems. In this article, we present the modeling and control of our MR-safe pneumatic radial inflow motor and encoder. A comprehensive model is developed that considers the primary dynamic elements of the system, including: 1) motor dynamics, 2) pneumatic transmission line dynamics, and 3) valve dynamics. After model validation, we present a simplified third order model that facilitates design of a first order sliding mode controller (TO-SMC). Finally, the motor hardware is tested in a 7T MRI. No image distortion or artifacts were observed. We posit the MR-safe motor and dynamic model will lower the entry barriers for researchers interested in MR-guided robots and promote wider adoption of MR-guided robotic systems.

Curvature-based force estimation for an elastic tube

Contact force is one of the most significant feedback for robots to achieve accurate control and safe interaction with environment. For continuum robots, it is possible to estimate the contact force based on the feedback of robot shapes, which can address the difficulty of mounting dedicated force sensors on the continuum robot body with strict dimension constraints. In this paper, we use local curvatures to estimate the magnitude and location of single or multiple contact forces based on Cosserat rod theory. We validate the proposed method in a thin elastic tube and calculate the curvatures via Fiber Bragg Grating (FBG) sensors or image feedback. For the curvature feedback obtained from multicore FBG sensors, the overall force magnitude estimation error is $0.062 \pm 0.068$ N and the overall location estimation error is $3.51 \pm 2.60$ mm. For the curvature feedback obtained from image, the overall force magnitude estimation error is $0.049 \pm 0.048$ N and the overall location estimation error is $2.75 \pm 1.71$ mm. The results demonstrate that the curvature-based force estimation method is able to accurately estimate the contact force.

A Brief Insight on Magnetic Resonance Conditional Neurosurgery Robots

The brain is a delicate organ in the human body that requires extreme care. Brain-related diseases are unavoidable. Perse, neurosurgery is a complicated procedure that demands high precision and accuracy. Developing a surgical robot is a complex task. To date, there are only a handful of neurosurgery robots in the market that distinctly undergo clinical procedures. These robots have exorbitant cost that hinders the utmost care progress in the area as they are unaffordable. This paper looked at the historical perspective and presented insight literature of the magnetic resonance conditional stereotactic neurosurgery robots that find their ways in clinics, abandoning research projects and promising research yet to undergo clinical use. In addition, the study also gives a thorough insight into the advantage of magnetic resonance imaging modalities and magnetic resonance conditional robots and the future challenges in automation use. Image compatibility test data and accuracy results are also examined because they guarantee that these systems work correctly in particular imaging settings. The primary differences between these systems include actuation and control technologies, construction materials, and the degree of freedom. Thus, one system has an advantage over the other.