Swinging Search and Crawling Control

Please be aware that the video accompanying this article may take some time to load, depending on the speed of your internet connection to GitHub.

A snake-inspired path planning algorithm based on reinforcement learning and self-motion for hyper-redundant manipulators

SSCC = Swinging Search (self-motion for path planning) and Crawling Control (crawling into the pipe like a snake)

Self-motion is a special kind of motion for redundant manipulators. Given a fixed base and an end target, there can be more than one possible solution configuration due to its redundancy. Searching configurations (for a collision-free path) within these constraints (fixing both ends of a manipulator) can be likened to a rope swinging in the air.

Keywords:

• Redundant Manipulator
  • Path Planning
  • Control
• Reinforcement Learning

Videos

Crawling Control: a 16 DoF manipulator crawling into a simple pipe.

Resources

1. PDF
2. Videos
3. Codes
4. Self-Motion demo