Tracking trajectory control of dual-eyes visual-based underwater vehicle

Nowadays, a variety of robots have been studied and developed for undersea exploration. The purpose of undersea exploration is to retrieve resources such as methane hydrate rare metal from bottom of sea. The ability for underwater vehicle to recharge batteries automatically in the sea is vital function so that the vehicle achieves the required tasks with long duration time, raising next technical problems of how the guide the vehicle to the recharging station and how to avert surrounding hazardous environment. One approach to solve the problem is to control the vehicle's trajectory by using visual information. This paper proposes a method to guide a vehicle to desired trajectory with dual-eye visual feedback, which utilizes inverse Jacobian matrix to generate desired trajectory in the underwater space. The experimental results show that the proposed control scheme is effective to have the vehicle follow the desired trajectories.