Simulation for Path Planning of OUC-II Glider with Intelligence Algorithm

Addressing the need for exploration of benthic zones utilising autonomous underwater gliders, this paper presents a simulation for an optimised path planning from the source node to the destination node of the autonomous underwater glider OUC-II Glider in near-bottom ocean environment. Near-bottom ocean current data from the College of Oceanic and Atmospheric Science, Ocean University of China have been used for this simulation. A cost function is formulated to describe the dynamics of the autonomous underwater glider in near-bottom ocean currents. This cost function is then optimised using various biologically-inspired algorithms such as genetic algorithm, Ant Colony optimisation algorithm and particle swarm optimisation algorithm. The simulation of path planning is also performed using Q-learning technique and the results are compared with the biologically-inspired algorithms. The results clearly show that the Q-learning algorithm is better in computational complexity than the biologically-inspired algorithms. The ease of simulating the environment is also more in the case of Q-learning techniques. Hence this paper presents an effective path planning technique, which has been tested for the OUC-II glider and it may be extended for use in any standard autonomous underwater glider.