Applications of HUGIN to Diagnosis and Control of Autonomous Vehicles

We present an application of HUGIN to solve problems related to diagnosis and control of autonomous vehicles. The application is based on a distributed architecture supporting diagnosis and control of autonomous units. The purpose of the architecture is to assist the operator or piloting system in managing fault detection, risk assessment, and recovery plans under uncertainty. To handle uncertainty, we focus on the use of probabilistic graphical models (PGMs) as implemented in the HUGIN tool. We describe the application of PGMs to three problems of diagnosis and control of autonomous vehicles. Based on the HUGIN tool, limited memory influence diagrams (LIMIDs) are used to represent and solve complex problems of diagnosis and control of autonomous ground and underwater vehicles. In particular, we describe how battery monitoring and control problems related to an underwater and a ground vehicle are solved and how to solve the problem of assessing the quality of a sonar image related to an underwater vehicle.