lfd: Learning from Demonstrations for Robotic Manipulation

lfd is a software framework for generalizing robot trajectories from demonstrations into new situations, thus enabling robots to learn to perform tasks from demonstrations.

This software follows the line of work of these papers: [ISRR2013], [IROS2013], [IROS2014].

Source code is available on github.

Contents

• Installation
  • Dependencies
  • Instructions
  • Running the Test Suite
• Additional Installation to Use the GPU
  • Dependencies
  • Instructions
• Documenting
  • Dependencies
  • Instructions
• Miscellaneous
  • Settings files
  • Downloading test data
  • Cache files

Index

• Index
• Module Index
• Search Page

References

[ISRR2013]

John Schulman, Jonathan Ho, Cameron Lee, Pieter Abbeel, “Learning from Demonstrations through the Use of Non-Rigid Registration,” in Proceedings of the 16th International Symposium on Robotics Research (ISRR), 2013.

[IROS2013]

John Schulman, Ankush Gupta, Sibi Venkatesan, Mallory Tayson-Frederick, Pieter Abbeel, “A Case Study of Trajectory Transfer through Non-Rigid Registration for a Simplified Suturing Scenario,” in Proceedings of the 26th IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2013.

[IROS2014]

Alex X. Lee, Sandy H. Huang, Dylan Hadfield-Menell, Eric Tzeng, Pieter Abbeel, “Unifying Scene Registration and Trajectory Optimization for Learning from Demonstrations with Application to Manipulation of Deformable Objects,” in Proceedings of the 27th IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2014.