Remote Manipulation Of Untethered Bots In Soft Media

Scientific and engineering literature extensively discusses untethered miniature devices, also called bots, as those parts of medical robotic systems that can be manipulated remotely within the human body. There is clear evidence that this topic has gained significant traction based on the spike in publication over the past seven years (Koleoso et al., 2020). Robotic systems that employ micro- and millimeter size bots are thought to have numerous potential medical applications including drug delivery (Jang et al., 2019), localized heating or cooling, cessation of bleeding, clearing of occlusions, establishing a diagnosis based on sensing or biopsy, and minimally invasive surgeries. Most studies focuses on the actuation and control of such bots in fluid environment, typically at low Reynolds numbers. Yet, many medical applications would involve soft tissues, rather than fluids. While controlling tethered devices, such as needles, based on appropriate models of interaction with tissue has been the subject of considerable interest. However, no similar work has been published for untethered miniature devices. The goal of the present thesis is to develop trajectory planning and control methods for untethered solid millimeter size bots in soft tissue under the action of magnetic force. This study employs a recently developed model describing the interaction of untethered bots with soft materials during bot motion by implementing a numerical simulation of the model. Given the highly non-linear and past trajectory (history) dependent nature of the interaction with the medium, in addition to significant model parameter variability throughout soft media, the present thesis hypothesizes that a trajectory control based on the so-called Sliding Mode Control (SMC) method will make it possible for a bot to follow a large class of practical trajectories accurately. Numerical simulations of the proposed SMC method were conducted and experimentally validated for a class of trajectories that could be well approximated by circular and straight segments. The robustness of the proposed method was analyzed numerically by simulating trajectories in media with random variations of various model parameters as well as in the presence of significant delays in the feedback signal used to estimate the bot position. All the work carried out in this thesis applies to bots that are spherical making it possible to simplify the model, trajectory control, and their experimental validation. Furthermore, experimental validations are limited in this work to bots that are on the order of millimeters in diameter. While smaller bots may be of interest in medicine, they would be significantly harder to actuate and image during experimental validation experiments. At the same time, millimeter size bots are commensurate with the typical dimensions of needles widely employed in medical applications. Keywords: Medical Robotics, Micro Robots, Motion control, Robophysics, Soft Media, Untethered Device