Implementation of gaits for achieving omnidirectional walking in a quadruped robot

Komal K. Teru, T. Komal Kumar, Rajesh Kumar, S. Mogily, Divyanshu Goel, Nahas Pareekutty, Suril Vijaykumar Shah, K. Madhava KrishnaAir 2015 · DOI

TL;DR

This paper tackles the challenge of enabling quadruped robots to walk omnidirectionally with improved stability. By introducing a modified crawl gait with a positive stability margin, the authors demonstrate enhanced robustness against terrain variations and weight distribution shifts, validated through simulations and real-world experiments.

What Problem Does This Paper Address?

Quadruped robots offer significant advantages over wheeled systems, particularly in navigating uneven and unpredictable terrain. However, achieving stable and versatile locomotion, especially omnidirectional movement (moving forward, backward, and sideways, as well as rotating), remains a complex engineering challenge. Existing static walking gaits, while simpler to implement, often suffer from a zero stability margin, making them prone to toppling when encountering slight disturbances or uneven surfaces. This paper aims to bridge this gap by developing a more robust omnidirectional walking strategy that guarantees a positive stability margin, thereby increasing the robot's reliability in real-world applications.

Methods

The core contribution of this paper lies in the development of a "modified crawl gait." This new gait builds upon conventional crawl and rotation gaits but introduces an additional phase where all four legs are in contact with the ground. This "all-legs-on-ground" phase is strategically placed to shift the robot's center of gravity (CoG) forward, thereby increasing the stability margin. The authors mathematically formulate the relationships between key gait parameters: stability margin, stride length, and duty factor. They then compare their modified crawl gait to the conventional one, demonstrating its superior stability margin both geometrically and mathematically. The paper also details the planning methodology for omnidirectional static walking, which involves planning foot placement, support and swing sequences, leg trajectories, and importantly, seamless transitions between different gaits (e.g., from forward crawl to rotation).

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • The "All-Legs-On-Ground" Phase is Key: Introducing a phase where all four legs are simultaneously on the ground is a simple yet effective way to increase the stability margin of a quadruped robot. This phase allows for controlled shifting of the robot's center of gravity.
  • Mathematical Formulation of Stability: The paper provides a clear mathematical framework for understanding the interplay between stride length, duty factor, and stability margin, which is crucial for designing robust gaits.
  • Gait Transitions are Crucial for Omnidirectionality: Achieving true omnidirectional movement requires not just individual stable gaits but also the ability to transition smoothly and stably between them. The paper highlights the importance of common foot positions for efficient transitions.
  • Trade-off Between Stability and Speed: The modified gait, while significantly improving stability, comes at the cost of reduced speed. This highlights a fundamental trade-off that needs to be considered in robot locomotion design.

Key Results

The paper demonstrates through mathematical analysis, computer simulations, and experimental validation that the modified crawl gait consistently maintains a positive stability margin throughout its operation. This is a significant improvement over conventional gaits that often have a zero stability margin. Experimental results on a physical quadruped robot showed that the robot could walk on rough surfaces without falling or wobbling, validating the enhanced stability. While specific quantitative metrics for stability margin improvement are presented in graphs (e.g., Figure 10), the core finding is the consistent positive margin across various gaits and transitions.

Limitations and Future Work

The authors acknowledge that the improved stability of their modified gait comes at the expense of reduced walking speed. This trade-off is a primary area for future investigation. The paper suggests that future work should focus on optimizing the balance between stability and speed, potentially by exploring more advanced control strategies or hybrid gait designs. Further research into more complex terrains and dynamic environments is also implied as a natural progression.

Summary updated

BibTeX
@inproceedings{teru2015implementation,
  author = {Komal K. Teru and T. Komal Kumar and Rajesh Kumar and S. Mogily and Divyanshu Goel and Nahas Pareekutty and Suril Vijaykumar Shah and K. Madhava Krishna},
  title = {{Implementation of gaits for achieving omnidirectional walking in a quadruped robot}},
  booktitle = {Air},
  year = {2015},
  doi = {10.1145/2783449.2783482},
}
Figure