IEEE Robotics and Automation Letters (RA-L) 2022

RCIK: Real-Time Collision-Free Inverse Kinematics Using a Collision-Cost Prediction Network

RCIK: Real-Time Collision-Free Inverse Kinematics Using a Collision-Cost Prediction Network

by Mincheul Kang, Yoonki Cho, and Sung-Eui Yoon

Korea Advanced Institute of Science and Technology (KAIST)






Abstract

In this paper, we present real-time collision-free inverse kinematics (RCIK) that accurately performs consecutively provided six-degrees-of-freedom commands in environments containing static and dynamic obstacles. Our method is based on an optimization-based IK approach to generate IK candidates with high feasibility for the command. While checking various constraints (e.g., collision and joint velocity limits), we select the best configuration among generated IK candidates through our objective function, considering the continuity of joints and collision avoidance with obstacles. To avoid dynamic obstacles efficiently, we propose a novel, collision-cost prediction network (CCPN) that estimates collision costs using an occupancy grid updated from sensor data in real-time. We evaluate our method in three dynamic problems using a real robot, the Fetch manipulator, and four static problems using three different configurations of robots. We show that the proposed method successfully performs the consecutively given commands in realtime, mainly thanks to the collision-cost prediction network, while avoiding dynamic and static obstacles. The results of tested problems are available in the accompanying video.



Contents

Paper: PDF (5.37MB)
Source Code: GitHub, ZIP
Model: fetch_7dof.pt


Bibtex:
@ARTICLE{9616379,
  author={Kang, Mincheul and Cho, Yoonki and Yoon, Sung-Eui},
  journal={IEEE Robotics and Automation Letters}, 
  title={RCIK: Real-Time Collision-Free Inverse Kinematics Using a Collision-Cost Prediction Network}, 
  year={2022},
  volume={7},
  number={1},
  pages={610-617},
  doi={10.1109/LRA.2021.3128238}}