Parent-child underwater robot-based manipulation system using AUV and agent robot = 무인자율수중로봇 기반 에이전트 로봇을 활용한 무관절형 수중 매니퓰레이션 시스템

다국어 초록 (Multilingual Abstract)

Underwater vehicles possessing manipulative abilities prove instrumental in many missions, enablingdirect observation and interaction with targeted objects alongside operation in hazardous or shallowenvironments instead of human divers. While such manipulators are predominantly integrated with remotely operated vehicles (ROVs) as robotic arms, few studies focus on maneuvering. Even with the robustness and immediate responsiveness of ROVs to varying conditions through manual operator control, they presentlimitations in the execution of tasks over extensive territories and complex, narrow terrains due to their size.

Simultaneously, there is a growing need for developing autonomous underwater vehicles (AUVs) that can effectively navigate broad areas. However, entering complex terrains andvehicle control during hovering operations introduces new challenges. Specifically, the vehicle'spose is impacted as the center shifts when the robot arm moves, necessitating the development of arefined control model. Even slight errors could disrupt delicate operations that maintain the vehicle's pose. Moreover, existing operational vehicle limitations render the simultaneous execution of multiple manipulation tasks, such as the relocation of extensive pipes, virtually impossible.

Thisstudyproposes a parent-child robot-based underwater manipulation system comprising two underwater vehicles with split roles. The larger autonomous underwater vehicle (AUV) acts as the base to transportandcontrol a tethered child robot that is deployed to perform actual manipulation tasks. Conventional AUVs face several challenges when performing manipulation missions because of their large andmassivebody and the complexity of dynamic control with a rigid arm manipulator in the precise hoveringstate. To address this problem, we proposed an underwater manipulation system that uses a largeparent AUV with high-performance sensors to navigate to the desired location and a child robot withhigh mobility to perform manipulation tasks. In this study, a parent-child robot-based manipulationsystemwas developed and mathematically modeled considering hydrodynamics. Based on the system model, a nonlinear time delay (TD) controller was applied to the child robot to ensure robustness of themovement, andcomplementary sensing methods were proposed according to the observation distance. Subsequently, wedetailed the mission process such that the manipulation tasks could be performed autonomously. To implement and verify the proposed method, we conducted three-dimensional (3D) simulations, water tankexperiments, and sea trials. In the 3D simulation and water tank experiments, we evaluated the position control using a TD controller during a specific disturbance and performed an operational evaluationof the parent-child system. In the water tank, we performed the operation of the child robot with the tether length adjustment of the winch system and the manipulation task. In the sea trial, the predefined mission was successfully completed, which demonstrated the feasibility of properlyoperating both the robots. The results of the experiments demonstrated the high potential and capabilities of the proposed manipulation system.

참고문헌 (Reference)

1. Numerical simulation of undersea cable dynamics, S Schechter, CM Ablow, 10(6):443–457, , 1983

2. Inspection-class remotely operated vehicles—a review, Gerard Dooly, Thomas Newe, Joseph Coleman, Edin Omerdi´c, Romano Capocci, Daniel Toal, 5(1):13, , 2017

3. Moveit!: autonomous underwater freefloating manipulation, Pere Ridao, Dina Youakim, David Ribas, Francesco Spadafora, Narcis Palomeras, Maurizio Muzzupappa, 24(3):41–51, , 2017

4. Invariant extended kalman filtering for underwater navigation, Easton R. Potokar, Joshua G. Mangelson, Kalin Norman, 6(3):5792–5799, , 2021

5. Development and validation of a lumpedmass dynamics model of a deep-sea rov system, FR Driscoll, RG Lueck andMNahon, 22(3):169–182, , 2000

6. Development of a flap-type mooring-less wave energy harvesting system for sensor buoy, Hangil Joe, Hyunwoo Roh, Hyeonwoo Cho, Son-Cheol Yu, 133:851–863, , 2017

7. Automatic generation and detection of highly reliable fiducial markers under occlusion, Francisco Jos´e, Manuel Jes´us Mar´ın-Jim´enez, Rafael Mu˜noz-Salinas, Sergio Garrido-Jurado, Madrid-Cuevas, 47(6):2280– 2292, , 2014

8. Underwater walking mechanism of underwater amphibious robot using hinged multi-modal paddle, Seokyong Song, Young-woon Song, Son-Cheol Yu, Taesik Kim, 19(4):1691– 1702, , 2021

9. Magnetic survey and autonomous target reacquisition with a scalar magnetometer on a small auv, Jeffrey Gee, Andrew Nager, Robert Hess, Andrew Pietruszka, Eric Terrill, Eric Gallimore, 37(7):1246–1266, , 2020

10. Dexterous underwater manipulation from onshore locations: Streamlining efficiencies for remotely operated underwater vehicles, Arturo Gomez Chavez, Tomasz Łuczynski, Gianluca Antonelli, Andreas Birk, Christian Mueller, Tobias Doernbach, Daniel Koehntopp, Andras Kupcsik, Ajay K Tanwani, Sylvain Calinon, 25(4):24–33, , 2018