An integrated collision avoidance system for autonomous underwater vehicles

Abstract

This paper presents an integrated approach to the design of an obstacle avoidance system for autonomous underwater vehicles. A type of incremental stochastic algorithm known as rapidly-exploring random tree (RRT) is merged with a manoeuvre automaton (MA) representation to form the motion planner. The improved planner is not only capable of taking into account the dynamics of the system, but it also allows complex non-linear manoeuvres to be performed. In addition, it is highly computationally efficient. This paper also extends it to take into account dynamic targets. The output trajectory from the planner is then tracked using a kinematic based state-dependent Riccati equation (SDRE) controller. In other words, by omitting the requirement of a dynamic model of the system, one can extend the application to a more encompassing types of vehicle whilst reducing the number of tuning parameters. Detailed simulation studies are conducted using a non-linear AUTOSUB model to investigate the feasibility of the proposed methods.

Acknowledgement

The author would like to thank the IMarEST for the Stanley Gray Fellowship award and also Dr. Frazzoli of UCLA for his most enlightening explanation regarding the MA concept.