1 / 24

Value Function Approximation on Non-linear Manifolds for Robot Motor Control

Value Function Approximation on Non-linear Manifolds for Robot Motor Control. Masashi Sugiyama 1) 2) Hirotaka Hachiya 1 ) 2) Christopher Towell 2) Sethu Vijayakumar 2) 1) Computer Science, Tokyo Institute of Technology 2) School of Informatics, University of Edinburgh.

svein
Download Presentation

Value Function Approximation on Non-linear Manifolds for Robot Motor Control

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Value Function ApproximationonNon-linear Manifolds for Robot Motor Control Masashi Sugiyama1)2)Hirotaka Hachiya1)2) Christopher Towell2) Sethu Vijayakumar2) 1) Computer Science, Tokyo Institute of Technology2) School of Informatics, University of Edinburgh

  2. Maze Problem: Guide Robot to Goal Possibleactions Up Left Right Position (x,y) reward Down Goal • Robot knows its position but doesn’t know which direction to go. • We don’t teach the best action to take at each position but give a reward at the goal. • Task: make the robot select the optimal action.

  3. Markov Decision Process (MDP) • An MDP consists of • : set of states, • : set of actions, • : transition probability, • : reward, • An action the robot takes at state is specified by policy . • Goal: make the robot learn optimal policy

  4. Definition of Optimal Policy • Action-value function: discounted sum of future rewards when taking in and following thereafter • Optimal value: • Optimal policy: • is computed if is given. • Question: How to compute ?

  5. Policy Iteration (Sutton & Barto, 1998) • Starting from some initial policy iterate Steps 1 and 2 until convergence. Step 1.Compute for current Step 2.Update by • Policy iteration always converges to if in step 1 can be computed. • Question: How to compute ?

  6. Bellman Equation • can be recursively expressed by • can be computed by solving Bellman equation • Drawback: dimensionality of Bellman equation becomes huge in large state and action spaces high computational cost

  7. Least-Squares Policy Iteration (Lagoudakis and Parr, 2003) • Linear architecture: • is learned so as to optimally approximate Bellman equation in the least-squares sense • # of parameters is only : • LSPI works well if we choose appropriate • Question: How to choose ? : fixed basis functions : parameters : # of basis functions

  8. Popular Choice: Gaussian Kernel (GK) • Smooth • Gaussian tail goes over partitions Partitions : Euclidean distance : Centre state

  9. Approximated Value Function by GK • Values around the partitions are not approximated well. Optimal value function Approximated by GK Log scale 20 randomly located Gaussians

  10. Policy Obtained by GK • GK provides an undesired policy around the partition. Optimal policy GK-based policy

  11. Aim of This Research • Gaussian tails go over the partition. • Not suited for approximating discontinuous value functions. We propose new Gaussian kernel to overcome this problem.

  12. State Space as a Graph • Ordinary Gaussian uses Euclidean distance. • Euclidean distance does not incorporate state space structure, so tail problems occur. • We represent state space structure by a graph, and use it for defining Gaussian kernels. (Mahadevan, ICML 2005)

  13. Geodesic Gaussian Kernels • Natural distance on graph is shortest path. • We use shortest path in Gaussian function. • We call this kernel geodesic Gaussian. • SP can be efficiently computed by Dijkstra. Shortest path Euclidean distance

  14. Example of Kernels • Tails do not go across the partition. • Values smoothly decrease along the maze. GeodesicGaussian Ordinary Gaussian

  15. Comparison of Value Functions Optimal Ordinary Gaussian Geodesic Gaussian • Values near the partition are well approximated. • Discontinuity across the partition is preserved.

  16. Comparison of Policies Ordinary Gaussian GeodesicGaussian • GGKs provide good policies near the partition.

  17. Experimental Result Average over 100 runs • Ordinary Gaussian: tail problem • Geodesic Gaussian: no tail problem Geodesic Fraction of optimal states Ordinary Number of kernels

  18. Robot Arm Reaching 2-DOF robot arm • Task: move the end effector to reach the object State space 180 Object End effector Obstacle Joint 2 (degree) 0 Joint 2 Joint 1 Reward: +1 reach the object 0 otherwise -180 -100 0 100 Joint 1 (degree)

  19. Robot Arm Reaching Successfully avoids the obstacle and can reach the object. Ordinary Gaussian Geodesic Gaussian Moves directly towards the objectwithout avoiding the obstacle.

  20. Khepera Robot Navigation • Khepera has 8 IR sensors measuring the distance to obstacles. • Task: explore unknown maze without collision Reward: +1 (forward) -2 (collision) 0 (others) Sensor value: 0 - 1030

  21. State Space and Graph Discretize 8D state space by self-organizing map. 2D visualization Partitions

  22. Khepera Robot Navigation When facing obstacle, goes backward (and goes forward again). Ordinary Gaussian Geodesic Gaussian When facing obstacle, makes a turn (and go forward).

  23. Experimental Results Average over 30 runs Geodesic Ordinary • Geodesic outperforms ordinary Gaussian.

  24. Conclusion • Value function approximation: good basis function needed • Ordinary Gaussian kernel: tail goes over discontinuities • Geodesic Gaussian kernel: smooth along the state space • Through the experiments, we showed geodesic Gaussian is promising in high-dimensional continuous problems!

More Related