2019 · 18 citations · 30 references
EngineeringComputer ScienceRobotics EngineeringRobot ControlRobot LearningRoboticsHand-eye CoordinationLearning From DemonstrationManipulation ControlPolicy LearningReinforcement LearningVisual ServoingInverse Reinforcement LearningJacobian EstimationLearned Reward FunctionsReinforcement Learning For RoboticsReinforcement Learning Visual ServoingReward Learning From DemonstrationRobot Learning From DemonstrationTask Learning From Demonstration
We present a robot eye-hand coordination learning method that can directly learn visual task specification by watching human demonstrations. Task specification is represented as a task function, which is learned using inverse reinforcement learning(IRL [1]) by inferring a reward model from state transitions. The learned reward model is then used as continuous feedbacks in an uncalibrated visual servoing(UVS [2]) controller designed for the execution phase. Our proposed method can directly learn from raw videos, which removes the need for hand-engineered task specification. Benefiting from the use of a traditional UVS controller, the training on real robot only happens at initial Jacobian estimation which takes an average of 4-7 seconds for a new task. Besides, the learned policy is independent from a particular robot, thus has the potential of fast adapting to other robot platforms. Various experiments were designed to show that, for a task with certain DOFs, our method can adapt to task/environment changes in target positions, backgrounds, illuminations, and occlusions.
30
A survey of robot learning from demonstration
Brenna Argall, Sonia Chernova, Manuela Veloso et al. · Robotics and Autonomous Systems · 2008 · 3.3K citations
Apprenticeship learning via inverse reinforcement learning
Pieter Abbeel, Andrew Y. Ng · 2004 · 2.9K citations
Eye–Hand Coordination in Object Manipulation
Roland S. Johansson, Göran Westling, Anders Bäckström et al. · Journal of Neuroscience · 2001 · 879 citations · Full text