Comparison of methods for using risk information in path planning for autonomous surface vessels

Abstract

Path planning is an essential part of autonomous surface vessel (ASV) operations. Risk models can be used to give estimates of risk related to the operation of technical systems, such as ASVs. Combining risk models directly with control systems can be a way to make autonomous systems more capable of assessing and evaluating risk, and in this way allow them to make better decisions. Focusing on path planning, this study aims at investigating how an online risk model used to inform an ASV of risk compares to a static safety domain. The results show that the paths selected when using the risk model can consider different environmental conditions, and the choice of path can be adapted based on how conservative the ASV should be with respect to risk. In this way the risk model provides a more detailed basis for making decisions than the static safety domain.