Guided policy search
Direct policy search can effectively scale to high-dimensional systems, but complex policies
with hundreds of parameters often present a challenge for such methods, requiring …
with hundreds of parameters often present a challenge for such methods, requiring …
LQR-trees: Feedback motion planning via sums-of-squares verification
R Tedrake, IR Manchester… - … Journal of Robotics …, 2010 - journals.sagepub.com
Advances in the direct computation of Lyapunov functions using convex optimization make it
possible to efficiently evaluate regions of attraction for smooth non-linear systems. Here we …
possible to efficiently evaluate regions of attraction for smooth non-linear systems. Here we …
Learning control in robotics
S Schaal, CG Atkeson - IEEE Robotics & Automation Magazine, 2010 - ieeexplore.ieee.org
Recent trends in robot learning are to use trajectory-based optimal control techniques and
reinforcement learning to scale complex robotic systems. On the one hand, increased …
reinforcement learning to scale complex robotic systems. On the one hand, increased …
LQR-trees: Feedback motion planning on sparse randomized trees.
R Tedrake - Robotics: Science and Systems, 2009 - direct.mit.edu
Recent advances in the direct computation of Lyapunov functions using convex optimization
make it possible to efficiently evaluate regions of stability for smooth nonlinear systems …
make it possible to efficiently evaluate regions of stability for smooth nonlinear systems …
Robust task-based control policies for physics-based characters
We present a method for precomputing robust task-based control policies for physically
simulated characters. This allows for characters that can demonstrate skill and purpose in …
simulated characters. This allows for characters that can demonstrate skill and purpose in …
[图书][B] Neural approximations for optimal control and decision
Many scientific and technological areas of major interest require one to solve infinite-
dimensional optimization problems, also called functional optimization problems. In such a …
dimensional optimization problems, also called functional optimization problems. In such a …
Standing balance control using a trajectory library
C Liu, CG Atkeson - 2009 IEEE/RSJ International Conference …, 2009 - ieeexplore.ieee.org
This paper presents a standing balance controller that explicitly handles pushes. We employ
a library of optimal trajectories and the neighboring optimal control method to generate local …
a library of optimal trajectories and the neighboring optimal control method to generate local …
Multiple balance strategies from one optimization criterion
CG Atkeson, B Stephens - 2007 7th IEEE-RAS International …, 2007 - ieeexplore.ieee.org
Multiple strategies for standing balance have been observed in humans, including using the
ankles to apply torque to the ground, using the hips and/or arms to generate horizontal …
ankles to apply torque to the ground, using the hips and/or arms to generate horizontal …
Cross-entropy optimization of control policies with adaptive basis functions
This paper introduces an algorithm for direct search of control policies in continuous-state
discrete-action Markov decision processes. The algorithm looks for the best closed-loop …
discrete-action Markov decision processes. The algorithm looks for the best closed-loop …
Min-max differential dynamic programming: Continuous and discrete time formulations
In this work, the first min-max Game-Theoretic Differential Dynamic Programming (GT-DDP)
algorithm in continuous time is derived. A set of backward differential equations for the value …
algorithm in continuous time is derived. A set of backward differential equations for the value …