Risk-Averse Decision Support for Optimal Use of Electric Vehicles
conference paper
Optimizing the use of an electric vehicle such that energy consumption, battery aging and operational costs are minimal while constraints are kept is a wicked problem with many uncertain, non-observable or unknown factors. AI technologies help here with decision support in logistics and energy estimations within the vehicle. However, separating these concerns disregards that operational decisions do not only set a vehicle's current behavior, but also impact its ability to perform in the future. We address this interweaving of current behavior and the future development of the situation as well as of the vehicles capabilities within a novel dual awareness loop that combines situation awareness with inner system reflection. Providing a probabilistic reasoning system, we demonstrate this for an electric vehicle use case, where we concurrently diagnose and predict the vehicle's state and thus its capabilities given past, current, and expected operations that in turn depend on the predicted situation. This allows for better predictions of range and risks given expected situations, environmental effects, and consequences of decisions, providing for decision support within an operational strategy.
Topics
TNO Identifier
1012027
ISSN
2153-0017
ISBN
979-8-3315-0592-9
Publisher
IEEE
Source title
2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), Edmonton, AB, Canada
Files
To receive the publication files, please send an e-mail request to TNO Repository.