Name
Title | ||
|---|---|---|
Motion capture and reinforcement learning of dynamically stable humanoid movement primitives |
Abstract | ||
|---|---|---|
Direct transfer of human motion trajectories to humanoid robots does not result in dynamically stable robot movements due to the differences in human and humanoid robot kinematics and dynamics. We developed a system that converts human movements captured by a low-cost RGB-D camera into dynamically stable humanoid movements. The transfer of human movements occurs in real-time. As need arises, the developed system can smoothly transition between unconstrained movement imitation and imitation with balance control, where movement reproduction occurs in the null space of the balance controller. The developed balance controller is based on an approximate model of the robot dynamics, which is sufficient to stabilize the robot during on-line imitation. However, the resulting movements cannot be guaranteed to be optimal because the model of the robot dynamics is not exact. The initially acquired movement is therefore subsequently improved by model-free reinforcement learning, both with respect to the accuracy of reproduction and balance control. We present experimental results in simulation and on a real humanoid robot. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/ICRA.2013.6631333 | Robotics and Automation |
Keywords | Field | DocType |
cameras,humanoid robots,learning (artificial intelligence),mechanical stability,mechanical variables control,motion control,robot dynamics,robot kinematics,balance control,dynamically stable humanoid movement primitives,human motion trajectory transfer,humanoid robot dynamics,humanoid robot kinematics,low-cost RGB-D camera,model-free reinforcement learning,motion capture,movement reproduction,red-green-blue-depth camera,robot stability | Motion capture,Motion control,Kinematics,Control theory,Control engineering,Artificial intelligence,Reinforcement learning,Humanoid robot,Robot control,Computer vision,Robot kinematics,Engineering,Robot | Conference |
Volume | Issue | ISSN |
2013 | 1 | 1050-4729 |
ISBN | Citations | PageRank |
978-1-4673-5641-1 | 5 | 0.47 |
References | Authors | |
15 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Rok Vuga | 1 | 19 | 2.23 |
| Ogrinc, M. | 2 | 5 | 0.47 |
| Andrej Gams | 3 | 385 | 29.54 |
| Tadej Petric | 4 | 178 | 20.60 |