Application of Learning Trajectory Preference Algorithm Based on On-Line Iterative Feedback in Robots
暂无分享,去创建一个
[1] David Silver,et al. Learning to search: Functional gradient techniques for imitation learning , 2009, Auton. Robots.
[2] P. Abbeel,et al. LQG-MP: Optimized path planning for robots with motion uncertainty and imperfect state information , 2011 .
[3] Maya Cakmak,et al. Keyframe-based Learning from Demonstration , 2012, Int. J. Soc. Robotics.
[4] Brett Browning,et al. A survey of robot learning from demonstration , 2009, Robotics Auton. Syst..
[5] Vijay Kumar,et al. Identification and Representation of Homotopy Classes of Trajectories for Search-based Path Planning in 3D , 2011, Robotics: Science and Systems.
[6] Aude Billard,et al. On Learning, Representing, and Generalizing a Task in a Humanoid Robot , 2007, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).
[7] Pieter Abbeel,et al. Autonomous Helicopter Aerobatics through Apprenticeship Learning , 2010, Int. J. Robotics Res..
[8] Maxim Likhachev,et al. Search-based planning for manipulation with motion primitives , 2010, 2010 IEEE International Conference on Robotics and Automation.
[9] Matei T. Ciocarlie,et al. Interactive Markers: 3-D User Interfaces for ROS Applications [ROS Topics] , 2011, IEEE Robotics Autom. Mag..
[10] Thierry Siméon,et al. The Stochastic Motion Roadmap: A Sampling Framework for Planning with Markov Motion Uncertainty , 2007, Robotics: Science and Systems.