Learning Dynamical System for Grasping Motion

08/15/2021
by   Xiao Gao, et al.
0

Dynamical System has been widely used for encoding trajectories from human demonstration, which has the inherent adaptability to dynamically changing environments and robustness to perturbations. In this paper we propose a framework to learn a dynamical system that couples position and orientation based on a diffeomorphism. Different from other methods, it can realise the synchronization between positon and orientation during the whole trajectory. Online grasping experiments are carried out to prove its effectiveness and online adaptability.

READ FULL TEXT

page 1

page 2

research
06/28/2022

Dext-Gen: Dexterous Grasping in Sparse Reward Environments with Full Orientation Control

Reinforcement learning is a promising method for robotic grasping as it ...
research
03/25/2020

Merging Position and Orientation Motion Primitives

In this paper, we focus on generating complex robotic trajectories by me...
research
03/23/2021

Learning 6DoF Grasping Using Reward-Consistent Demonstration

As the number of the robot's degrees of freedom increases, the implement...
research
07/22/2018

FastOrient: Lightweight Computer Vision for Wrist Control in Assistive Robotic Grasping

Wearable and Assistive robotics for human grasp support are broadly eith...
research
06/20/2019

Boosting for Dynamical Systems

We propose a framework of boosting for learning and control in environme...
research
03/17/2021

Learning Descriptor of Constrained Task from Demonstration

Constrained objects, such as doors and drawers are often complex and sha...
research
10/20/2021

Periodic DMP formulation for Quaternion Trajectories

Imitation learning techniques have been used as a way to transfer skills...

Please sign up or login with your details

Forgot password? Click here to reset