Study on underwater navigation of crawler type mining robot

In South Korea, a self-propelled mining robot, named MineRo, was developed in 2007. In 2009 and 2010, shallow water tests in around 100m depth have been executed for the purpose of evaluations of collecting, lifting and driving performance. Particularly, in the 2010's test, a path tracking control of the robot was performed near harbor Hupo at East Sea of South Korea. For the path tracking control, most of all, the underwater navigation informed robot's position should be ensured. This paper proposes two navigation algorithms, dead-reckoning and extended kalman filter(EKF) based on the kinematics of robot. Firstly, the kinematic parameters, such as slip and slip-angle are calculated from velocity of DVL and yaw-rate in real time. In case of dead reckoning, the position is directly estimated from the kinematics substituting the parameters. In EKF, by applying the parameters to the Jacobian matrix, the accuracy is improved. The performances of two algorithms are verified through simulation using the data of various driving test of the robot.