2021/05/10 by Michael Lutter, Shie Mannor, Lutter, Michael +7 · 3 citations
Computer Science · Engineering · #Advanced Control Systems Optimization #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Reinforcement Learning in Robotics #Robotics (cs.RO) #Systems and Control (eess.SY) #cs.LG #cs.RO #cs.SY #eess.SY #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2105.04682
Accepted at International Conference on Machine Learning (ICML) 2021
arxiv created 2021/05/10 · openalex publication_date 2021/05/10 · arxiv updated 2021/05/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Classical value iteration approaches are not applicable to environments with continuous states and actions. For such environments, the states and actions are usually discretized, which leads to an exponential increase in computational complexity. In this paper, we propose continuous fitted value iteration (cFVI). This algorithm enables dynamic programming for continuous states and actions with a known dynamics model. Leveraging the continuous-time formulation, the optimal policy can be derived for non-linear control-affine dynamics. This closed-form solution enables the efficient extension of value iteration to continuous environments. We show in non-linear control experiments that the dynamic programming solution obtains the same quantitative performance as deep reinforcement learning methods in simulation but excels when transferred to the physical system. The policy obtained by cFVI is more robust to changes in the dynamics despite using only a deterministic model and without explicitly incorporating robustness in the optimization. Videos of the physical system are available at \urlhttps://sites.google.com/view/value-iteration.