Online Learning for Energy Efficient Navigation in Stochastic Transport Networks
The first part of the thesis introduces an online machine learning framework for navigation of electric vehicles, with the objective of adaptively and efficiently navigating the vehicle in a stochastic traffic environment. We assume that the road-specific probability distributions of vehicle energy consumption are unknown, and thus, we need to learn their parameters through observations. Furthermore, we take a Bayesian approach and assign prior beliefs to the parameters based on longitudinal vehicle dynamics. We view the task as a combinatorial multi-armed bandit problem, and utilize Bayesian bandit algorithms, such as Thompson Sampling, to address it. We establish theoretical performance guarantees for Thompson Sampling, in the form of upper bounds on the Bayesian regret, on single-agent, multi-agent and batched feedback variants of the problem. To demonstrate the effectiveness of the framework, we perform simulation experiments on various real-life road networks.
In the second half of the thesis, we extend the online learning framework to find paths which minimize or avoid bottlenecks. Solutions to the online minimax path problem represent risk-averse behaviors, by avoiding road segments with high variance in costs. We derive upper bounds on the Bayesian regret of Thompson Sampling adapted to this problem, by carefully handling the non-linear path cost function. We identify computational tractability issues with the original problem formulation, and propose an alternative approximate objective with an associated algorithm based on Thompson Sampling. Finally, we conduct several experimental studies to evaluate the performance of the approximate algorithm.
Online Minimax Path Problem
Online Shortest Path Problem
Energy Efficient Navigation
Data Science och AI
Åkerblom, N., Chen, Y., Chehreghani, M. H. Online Learning of Energy Consumption for Navigation of Electric Vehicles
Åkerblom, N., Hoseini, F. S., Chehreghani, M. H. Online Learning of Network Bottlenecks via Minimax Paths
EENE: Energieffektiv Navigering för Elfordon
FFI - Fordonsstrategisk forskning och innovation (2018-01937), 2019-01-01 -- 2022-12-31.
Annan data- och informationsvetenskap
Datorseende och robotik (autonoma system)
Informations- och kommunikationsteknik
Room 8103, EDIT Building, Rännvägen 6. Zoom (password request: email@example.com)
Opponent: Prof. Joakim Jaldén, Department of Intelligent Systems, KTH Royal Institute of Technology