Time-bounded mission planning in time-varying domains with semi-MDPS and Gaussian processes

Conference item

Uncertain, time-varying dynamic environments are ubiquitous in real world robotics. We propose an online planning framework to address time-bounded missions under time-varying dynamics, where those dynamics affect the duration and outcome of actions. We pose such problems as semi-Markov decision processes, where actions have a duration distributed according to an a priori unknown time-varying function. Our approach maintains a belief over this function, and time is propagated through a discrete search tree that efficiently maintains a subset of reachable states. We show improved mission performance on a marine vehicle simulator acting under real-world spatio-temporal ocean currents, and demonstrate the ability to solve co-safe linear temporal logic problems, which are more complex than the reachability problems tackled in previous approaches.

Authors: Duckworth, P; Lacerda, B; Hawes, N

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Journal of Machine Learning Research
• Host title: Proceedings of the 2020 Conference on Robot Learning
• Pages: 1654-1668
• Series: Proceedings of Machine Learning Research
• Series number: 155
• Publication date: 2021-10-04
• Acceptance date: 2020-10-14
• Event title: 4th Annual Conference on Robot Learning (CoRL2020)
• Event location: Virtual Conference
• Event website: https://sites.google.com/robot-learning.org/corl2020/
• Event start date: 2020-11-16
• Event end date: 2020-11-18
• ISSN: 2640-3498
• Language: English
• Keywords: MCTS; semi-MDPs; FFR; representing uncertainty; Gaussian processes