Catenary-based visual servoing for tether shape control between underwater vehicles

OCEAN ENGINEERING

卷:200

Article

Laranjeira, Matheus¹  Dune, Claire¹  Hugel, Vincent¹ 

[1] Univ Toulon & Var, Lab Cosmer, Toulon, France

关键词: Underwater robotics;    Visual servoing;    Control;    Sensory based robot control;   

DOI  :  10.1016/j.oceaneng.2020.107018

来源: Elsevier

【 摘 要 】

This paper deals with the shape control of a passive tether that links underwater vehicles. The pair of robots is composed of a leader that explores the environment, and a follower that is in charge of regulating the tether shape by moving appropriately. Both robots are equipped with standard low-cost sensors: cameras, depth sensor and inertial measurement unit (IMU). This work introduces an innovative visual servoing control scheme based on a parameterized catenary model of the tether, assuming that it is negatively buoyant The tether is detected in the image through color segmentation and its 3-D shape parameters are estimated through a nonlinear optimization procedure thanks to an adapted Gauss-Newton algorithm. Simulations and experimental results show that the relative positioning of the robots can be achieved through simultaneous tether shape visual servoing in both leader and follower's cameras, even in situations of tether partial visibility and strong disturbances, which proves the approach to be reliable.

【 授权许可】

Free