|
Study on Direct Teaching of End Effector for Surgical Assistant Robot Based on Current Feedback |
|
Youqiang Zhang, Minhyo Kim, Cheolsu Jeong, Dongchan Kim, Gyoungjun Lee, Yeri Sim, Sangrok Jin(Pusan National University, Korea) |
| This paper presents a force-free control algorithm for the direct teaching of end-effectors. In order to estimate and control torque by motor current without a force/torque sensor, the gravity and friction model of the device are derived through repeated experiments. The LuGre model is applied to the friction model, and static and dynamic parameters are obtained using a curve fitting function and genetic algorithm. Direct teaching control is designed using a force-free control algorithm that compensates the estimated torque from the motor current for gravity and friction and then converts it into a position control input. Operation sensitivity is verified through hand guiding experiments. |
|
|
Autonomous Guidewire Navigation with Deep Reinforcement Learning in Simulation for Cardiovascular Intervention |
|
JuEun Choi(Asan Medical Center, Korea), Hyunsuk Yoo(REMO inc, Korea), Sangeun Park(University of Ulsan College of Medicine, Korea), Youngjin Moon, Jaesoon Choi(Asan Medical Center & University of Ulsan, Korea) |
| The guidewire navigation requires complex control of long flexible instruments. The experiences of highly skilled physicians are crucial to the success of the procedure. Autonomously controlled robots are one avenue for overcoming these limitations in cardiovascular intervention.
We intend to develop the method of automatic control of the guidewire with machine learning-based control methods such as reinforcement learning by using our developed robotic system for cardiovascular intervention. This study aims to control a guidewire that moves to target positions in the simulated vascular vessels with Dueling Deep Q-Network. |
|
|
Animal Feasibility Study of Robotic System for Transurethral Resection of Bladder Tumors |
|
Jaeho Hyun, Dongho Lee, Bomi Yang(Asan Medical Center, Korea), Bumsik Hong, Bumjin Lim, Jaesoon Choi, Youngjin Moon(Asan Medical Center & University of Ulsan, Korea) |
| This paper confirms the feasibility of a robotic system for transurethral resection of bladder tumor. The system mainly consists of master and slave compartments: the master part includes a master device and user console, and the slave one has a steerable endoscope, manipulator, and control base. The manipulator moves the endoscope assembly and adjusts motion of the endoscope, so its movement is four degrees of freedom of translation, rolling, elevation, pivoting. To evaluate performance on steering and resection using the developed robotic endoscopic device, two animal trials are performed. |
|
|
In vivo and Clinical Evaluation of a Novel Vascular Intervention Assist Robot |
|
Youngjin Moon(Asan Medical Center & University of Ulsan, Korea), JuEun Choi(Asan Medical Center, Korea), Jaesoon Choi(Asan Medical Center & University of Ulsan, Korea) |
| A novel robotic system using bi-motional roller-cartridge assemblies and haptic manipulator for modularized vascular intervention instruments' teleoperation has been developed and evaluated in a pilot clinical trial. The roller-cartridge assembly can accommodate instruments of various sizes with an active clamping mechanism and manipulates them individually or simultaneously inserting and rotating. The primary performance associated with positioning accuracy and precision in translational and rotational motions of endovascular tools using the robotic system was tested, and the results showed sufficiently high accuracy and precision for application in the intervention procedure. |
|
|
Design of Acceleration Feedback of UGV Using a Stewart Platform |
|
Keunwoo Park, Eunhye Youn, Sunbum Kim, Yeonsu Kim, Geehyuk Lee(KAIST, Korea) |
| A fast-speed Unmanned Ground Vehicle (UGV) is utilized for logistics or carrying casualties in dangerous environments, substituting humans. Controlling the UGV’s acceleration is essential for a human operator concerning stability. However, it is not possible to recognize the accurate acceleration of UGV only with visual information. We designed haptic feedback using a Stewart platform for a pilot to control a UGV’s acceleration and verified its effect through a user study. The result showed that a user with haptic feedback could drive a UGV with less acceleration on rough terrain than without haptic feedback. |
|
|
Ensuring Reliable OpenFlow Channel for SDN Based Cyber-Physical System |
|
Hyeontae Joo, Hwangnam Kim(Korea University, Korea) |
| OpenFlow channel connecting the two representative components of SDN, the SDN controller and the switch, is configured over a wireless network, it often falls into a bad state, which can lead to uncertainty in transmitting the necessary network state data. In this paper, we applied Multipath TCP (MPTCP) rather than Transport Control Protocol (TCP) to the unstable wireless OpenFlow channel to improve the uncertainty about the transmission failure rate. We confirmed that the OpenFlow channel has high reliability by applying various schedulers
for MPTCP. |
|