STOMP: Stochastic trajectory optimization for motion planning
- Mrinal Kalakrishnan, Sachin Chitta, E. Theodorou, P. Pastor, S. Schaal
- Computer ScienceIEEE International Conference on Robotics and…
- 9 May 2011
It is experimentally show that the stochastic nature of STOMP allows it to overcome local minima that gradient-based methods like CHOMP can get stuck in.
A Generalized Path Integral Control Approach to Reinforcement Learning
- E. Theodorou, J. Buchli, S. Schaal
- Computer ScienceJournal of machine learning research
- 1 March 2010
The framework of stochastic optimal control with path integrals is used to derive a novel approach to RL with parameterized policies to demonstrate interesting similarities with previous RL research in the framework of probability matching and provides intuition why the slightly heuristically motivated probability matching approach can actually perform well.
Information theoretic MPC for model-based reinforcement learning
- Grady Williams, Nolan Wagener, E. Theodorou
- Computer ScienceIEEE International Conference on Robotics and…
- 1 May 2017
An information theoretic model predictive control algorithm capable of handling complex cost criteria and general nonlinear dynamics and using multi-layer neural networks as dynamics models to solve model-based reinforcement learning tasks is introduced.
Aggressive driving with model predictive path integral control
- Grady Williams, P. Drews, Brian Goldfain, James M. Rehg, E. Theodorou
- Computer ScienceIEEE International Conference on Robotics and…
- 16 May 2016
A model predictive control algorithm designed for optimizing non-linear systems subject to complex cost criteria using a stochastic optimal control framework using a fundamental relationship between the information theoretic notions of free energy and relative entropy is presented.
Information-Theoretic Model Predictive Control: Theory and Applications to Autonomous Driving
- Grady Williams, P. Drews, Brian Goldfain, James M. Rehg, E. Theodorou
- EngineeringIEEE Transactions on robotics
- 7 July 2017
An information-theoretic approach to stochastic optimal control problems that can be used to derive general sampling-based optimization schemes is presented and applied to the task of aggressive autonomous driving around a dirt test track.
Reinforcement learning of motor skills in high dimensions: A path integral approach
- E. Theodorou, J. Buchli, S. Schaal
- Computer ScienceIEEE International Conference on Robotics and…
- 3 May 2010
This paper derives a novel approach to RL for parameterized control policies based on the framework of stochastic optimal control with path integrals, and believes that this new algorithm, Policy Improvement with Path Integrals (PI2), offers currently one of the most efficient, numerically robust, and easy to implement algorithms for RL in robotics.
Learning variable impedance control
- J. Buchli, F. Stulp, E. Theodorou, S. Schaal
- Computer ScienceInt. J. Robotics Res.
- 1 June 2011
The results show that the power of variable impedance control is made available to a wide variety of robotic systems and practical applications and can be used not only for planning but also to derive variable gain feedback controllers in realistic scenarios.
Model Predictive Path Integral Control: From Theory to Parallel Computation
- Grady Williams, Andrew Aldrich, E. Theodorou
- Engineering
- 31 January 2017
The current simulations illustrate the efficiency and robustness of the proposed approach and demonstrate the advantages of computational frameworks that incorporate concepts from statistical physics, control theory, and parallelization against more traditional approaches of optimal control theory.
Model Predictive Path Integral Control using Covariance Variable Importance Sampling
- Grady Williams, Andrew Aldrich, E. Theodorou
- EngineeringArXiv
- 3 September 2015
In this paper we develop a Model Predictive Path Integral (MPPI) control algorithm based on a generalized importance sampling scheme and perform parallel optimization via sampling using a Graphics…
Skill learning and task outcome prediction for manipulation
- P. Pastor, Mrinal Kalakrishnan, Sachin Chitta, E. Theodorou, S. Schaal
- Computer Science, PsychologyIEEE International Conference on Robotics and…
- 9 May 2011
This work presents a Reinforcement Learning based approach to acquiring new motor skills from demonstration that allows the robot to learn fine manipulation skills and significantly improve its success rate and skill level starting from a possibly coarse demonstration.
...
...