TY - GEN
T1 - Information theoretic MPC for model-based reinforcement learning
AU - Williams, Grady
AU - Wagener, Nolan
AU - Goldfain, Brian
AU - Drews, Paul
AU - Rehg, James M.
AU - Boots, Byron
AU - Theodorou, Evangelos A.
N1 - This work was made possible by the ARO through MURI award W911NF-11-1-0046, DURIP award W911NF-12-1-0377, NSF award NRI-1426945, and supported by the NSF Graduate Research Fellowship under Grant No. 2015207631.
PY - 2017/7/21
Y1 - 2017/7/21
N2 - We introduce an information theoretic model predictive control (MPC) algorithm capable of handling complex cost criteria and general nonlinear dynamics. The generality of the approach makes it possible to use multi-layer neural networks as dynamics models, which we incorporate into our MPC algorithm in order to solve model-based reinforcement learning tasks. We test the algorithm in simulation on a cart-pole swing up and quadrotor navigation task, as well as on actual hardware in an aggressive driving task. Empirical results demonstrate that the algorithm is capable of achieving a high level of performance and does so only utilizing data collected from the system.
AB - We introduce an information theoretic model predictive control (MPC) algorithm capable of handling complex cost criteria and general nonlinear dynamics. The generality of the approach makes it possible to use multi-layer neural networks as dynamics models, which we incorporate into our MPC algorithm in order to solve model-based reinforcement learning tasks. We test the algorithm in simulation on a cart-pole swing up and quadrotor navigation task, as well as on actual hardware in an aggressive driving task. Empirical results demonstrate that the algorithm is capable of achieving a high level of performance and does so only utilizing data collected from the system.
UR - https://www.scopus.com/pages/publications/85028017652
UR - https://www.scopus.com/pages/publications/85028017652#tab=citedBy
U2 - 10.1109/ICRA.2017.7989202
DO - 10.1109/ICRA.2017.7989202
M3 - Conference contribution
AN - SCOPUS:85028017652
T3 - Proceedings - IEEE International Conference on Robotics and Automation
SP - 1714
EP - 1721
BT - ICRA 2017 - IEEE International Conference on Robotics and Automation
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2017 IEEE International Conference on Robotics and Automation, ICRA 2017
Y2 - 29 May 2017 through 3 June 2017
ER -