2015/09/06 by Yunpeng Pan, Pan, Yunpeng, Evangelos A. Theodorou +3 · 15 citations
Computer Science · Engineering · Mathematics · Physics and Astronomy · #Algorithm #Artificial intelligence #Computer science #Control (management) #Control Systems and Identification #FOS: Electrical engineering #Gaussian Processes and Bayesian Inference #Generalizability theory #Generalization #Mathematical optimization #Mathematics #Model Reduction and Neural Networks #Optimal control #Path (computing) #Probabilistic logic #Representation (politics) #Sample (material) #Sampling (signal processing) #Statistics #Systems and Control (eess.SY) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1509.01846
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2015/09/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06
We present a data-driven optimal control framework that can be viewed as a generalization of the path integral (PI) control approach. We find iterative feedback control laws without parameterization based on probabilistic representation of learned dynamics model. The proposed algorithm operates in a forward-backward manner which differentiate from other PI-related methods that perform forward sampling to find optimal controls. Our method uses significantly less samples to find optimal controls compared to other approaches within the PI control family that relies on extensive sampling from given dynamics models or trials on physical systems in model-free fashions. In addition, the learned controllers can be generalized to new tasks without re-sampling based on the compositionality theory for the linearly-solvable optimal control framework. We provide experimental results on three different systems and comparisons with state-of-the-art model-based methods to demonstrate the efficiency and generalizability of the proposed framework.