The Omega Turn: A General Turning Template for Elongate Robots

Baxi Chong, Tianyu Wang, Kelimar Diaz, Christopher Pierce, Eva Erickson, Julian Whitman, Yuelin Deng, Esteban Flores, Ruijie Fu, Juntao He, Jianfeng Lin, Hang Lu, Guillaume Sartoretti, Howie Choset, Daniel I. GoldmanarXiv 2025 · DOI · arXiv

Abstract

Elongate, limbless robots are uniquely suited for navigating cluttered environments, yet executing sharp, efficient turns remains a significant challenge. This paper presents a robust turning template derived from the biological "omega turn" of the nematode C. elegans. By modeling the turn as a superposition of two traveling waves, we demonstrate a controller that allows both snake-like and multi-legged robots to achieve high-angle, compact turns in complex terrains, including granular media and outdoor rock piles.

Introduction

Robotic snakes and other elongate systems are ideal for search-and-rescue or industrial inspection in tight spaces. However, while forward locomotion is well-understood, effective turning is difficult to implement. Traditional strategies, such as adding a constant offset to a body wave, often result in wide, sweeping movements that lead to collisions in confined areas. The goal of this research is to develop a turning strategy that is both agile (large turn angle) and compact (small footprint) while remaining robust to unpredictable environmental interactions.

Biological Inspiration and Methodology

The researchers employed a comparative theory-biology approach, beginning with the microscopic worm C. elegans. By tracking the worm's movements and applying Principle Component Analysis (PCA), they decomposed the complex "omega turn"—where the organism's head nearly touches its tail—into fundamental mathematical components.

The characteristic loop of the C. elegans omega turn serves as the biological template for robotic agility.

The Wave Superposition Template

The core discovery was that the omega turn can be prescribed by superimposing two distinct traveling waves: a "forward wave" and a "turning wave" with different spatial frequencies. Using geometric mechanics, the team calculated "height functions" to optimize the timing and amplitude of these waves, maximizing the turn angle while preventing body self-collision.

To handle real-world obstacles, the researchers implemented a compliant control framework. This allows the robot to adjust wave amplitudes based on sensed joint torques, effectively using physical contact with the environment to facilitate the turn.

Experimental Results

The template was tested on a 16-joint snake robot and a 10-legged elongate robot across diverse environments.

  • Superior Agility: On flat ground, the omega turn achieved an average angular displacement of 105.7° per cycle, vastly outperforming traditional offset turns (61.6°) and geometric turns (9.0°).
  • Compactness: The omega turn demonstrated the smallest "swept area" per degree of turn, making it the most space-efficient strategy for cluttered environments.
  • Environmental Robustness: In field tests on random rock piles, the omega turn maintained a robust 100.1° per cycle. In granular media (simulating sand), it produced ~78.5° of turn, whereas other common strategies failed to produce significant reorientation.
A snake robot executing a robust omega turn on a pile of rocks, demonstrating the template's effectiveness in unstructured terrain.

Discussion and Future Work

The universality of this template is a key finding; the same wave superposition enabled a multi-legged robot to turn effectively, suggesting that body-driven turning is a fundamental strategy for any elongate system. Furthermore, mechanical intelligence—leveraging physical interactions via compliant control—improved turning performance by approximately 30%.

Future research will focus on developing autonomous transitions between forward crawling and the omega turn, as well as integrating the template into high-level path planners for fully autonomous missions. Extending these 2D motions into 3D "volumetric" turns could further enhance maneuverability in complex vertical structures.

Summary updated

BibTeX
@misc{chong2025omega,
  author = {Baxi Chong and Tianyu Wang and Kelimar Diaz and Christopher Pierce and Eva Erickson and Julian Whitman and Yuelin Deng and Esteban Flores and Ruijie Fu and Juntao He and Jianfeng Lin and Hang Lu and Guillaume Sartoretti and Howie Choset and Daniel I. Goldman},
  title = {{The Omega Turn: A General Turning Template for Elongate Robots}},
  year = {2025},
  eprint = {2510.12970},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2510.12970},
}
Figure