Sources
GPMP: Code of “Gaussian process movement primitive”, Automatica, 2023.
NEUM-DS: Code of “Learning a flexible neural energy function with a unique minimum for globally stable and accurate demonstration learning”, IEEE Transactions on Robotics, 2023.
Neural energy learning with gurantee of a unique minimum
Online planning + state-dependent compliant control: Learning Stable State-Dependent Variable Impedance Control for Compliant Manipulation,” IEEE/ASME Transactions on Mechatronics, 2024
Compliant control for assemby
Compliant control for polish
Constrained motion planning
Bimanual assembly with different end-direction constraints
Test-tube insertion with different start and end-direction constraints
Obstacle avoidance: with in 10ms (Optimization-free)
Diffusion Movement Primitives: Multimodal Trajectory Generators for Generalizable Few-Shot Demonstration Learning
LASA data set
One-shot learning: case 1
One-shot learning: case 2
One-shot learning: case 3
Few-shot multimodal learning
One-shot learning with via-point constraints
Book insertion
Multimodal demonstrations
Demonstration 1
Demonstration 2
Generalizations
Generalization 1
Generalization 2: Multimodal sampling 1
Generalization 3: Multimodal sampling 2