Development of A Soft Robotic Approach for An Intra-abdominal Wireless Laparoscopic Camera

In Single-Incision Laparoscopic Surgery (SILS), the Magnetic Anchoring and Guidance System (MAGS) arises as a promising technique to provide larger workspaces and field of vision for the laparoscopes, relief space for other instruments, and require fewer incisions. Inspired by MAGS, many concept designs related to fully insertable magnetically driven laparoscopes are developed and tested on the transabdominal operation. However, ignoring the tissue interaction and insertion procedure, most of the designs adopt rigid structures, which not only damage the patients' tissue with excess stress concentration and sliding motion but also require complicated operation for the insertion. Meanwhile, lacking state tracking of the insertable camera including pose and contact force, the camera systems operate in open-loop control. This provides mediocre locomotion precision and limited robustness to uncertainties in the environment. This dissertation proposes, develops, and validates a soft robotic approach for an intra-abdominal wireless laparoscopic camera. Contributions presented in this work include (1) feasibility of a soft intra-abdominal laparoscopic camera with friendly tissue interaction and convenient insertion, (2) six degrees of freedom (DOF) real-time localization, (3) Closed-loop control for a robotic-assisted laparoscopic system and (4) untethering solution for wireless communication and high-quality video transmission. Embedding magnet pairs into the camera and external actuator, the camera can be steered and anchored along the abdominal wall through transabdominal magnetic coupling. To avoid the tissue rapture by the sliding motion and dry friction, a wheel structure is applied to achieve rolling motion. Borrowing the ideas from soft robotic research, the main body of the camera implements silicone material, which grants it the bendability to passively attach along the curved abdominal wall and the deformability for easier insertion. The six-DOF pose is estimated in real-time with internal multi-sensor fusion and Newton-Raphson iteration. Combining the pose tracking and force-torque sensor measurement, an interaction model between the deformable camera and tissue is established to evaluate the interaction force over the tissue surface. Moreover, the proposed laparoscopic system is integrated with a multi-DOF manipulator into a robotic-assisted surgical system, where a closed-loop control is realized based on a feedback controller and online optimization. Finally, the wireless control and video streaming are accomplished with Bluetooth Low Energy (BLE) and Analog Video (AV) transmission. Experimental assessments have been implemented to evaluate the performance of the laparoscopic system.