Model Estimation Within Planning and Learning
Author(s)
Geramifard, Alborz; Redding, Joshua D.; Joseph, Joshua Mason
DownloadRoy_Model estimation within.pdf (1.310Mb)
OPEN_ACCESS_POLICY
Open Access Policy
Creative Commons Attribution-Noncommercial-Share Alike
Terms of use
Metadata
Show full item recordAbstract
Risk and reward are fundamental concepts in the cooperative control of unmanned systems. In this research, we focus on developing a constructive relationship between cooperative planning and learning algorithms to mitigate the learning risk, while boosting system (planner & learner) asymptotic performance and guaranteeing the safety of agent behavior. Our framework is an instance of the intelligent cooperative control architecture (iCCA) where the learner incrementally improves on the output of a baseline planner through interaction and constrained exploration. We extend previous work by extracting the embedded parameterized transition model from within the cooperative planner and making it adaptable and accessible to all iCCA modules. We empirically demonstrate the advantage of using an adaptive model over a static model and pure learning approaches in an example GridWorld problem and a UAV mission planning scenario with 200 million possibilities. Finally we discuss two extensions to our approach to handle cases where the true model can not be captured exactly through the presumed functional form.
Date issued
2012-06Department
Massachusetts Institute of Technology. Aerospace Controls Laboratory; Massachusetts Institute of Technology. Department of Aeronautics and AstronauticsJournal
American Control Conference (ACC), 2012
Citation
Geramifard, A.; Redding, J.D.; Joseph, J.; Roy, N.; How, J.P., "Model estimation within planning and learning," American Control Conference (ACC), 2012 , vol., no., pp.793,799, 27-29 June 2012.
Version: Author's final manuscript
ISBN
978-1-4673-2102-0
978-1-4577-1095-7
ISSN
0743-1619