A new method to achieve smooth walking transitions in hexapod robots

Robots that can navigate different terrains both quickly and efficiently can be highly beneficial, allowing them to successfully complete complex missions in challenging environments. For example, these robots can help monitor complex natural environments, such as forests, or search for survivors after natural disasters.

One of the most common types of robots designed to navigate different terrains are legged robots, whose bodies are often inspired by the body structure of animals. To move quickly across different terrains, legged robots must be able to adapt their movements and walking style based on detected changes in their environmental conditions.

Researchers at the Higher Institute of Applied Science and Technology in Damascus, Syria, recently developed a new method to enable a smooth transition between the different gaits of a hexapod robot.

Their proposed gait control technique, introduced in a paper published in Heliyon, is based on so-called central pattern generators (CPGs), computational approaches that mimic biological CPGs. These are the neural networks underlying many rhythmic movements performed by humans and animals (i.e. walking, swimming, jogging, etc.).

“Our recent publication is a fundamental part of a larger project that aims to revolutionize the locomotion control of hexapod robots,” Kifah Helal, corresponding author of the paper, told Tech Xplore.

“Although machine learning techniques have not yet been integrated, the architecture we have designed lays the foundation for such advanced applications. Our methodology is designed with future machine learning integration in mind and, when implemented, ensures that it will compensate for failures will improve significantly.”

Helal and his colleagues first wanted to design and simulate a six-legged (hexapod) robot. This simulated robot platform was then used to test their proposed control architecture based on CPGs.

“Our control method uses the principles of CPGs where each leg of the hexapod robot is controlled by a clear rhythmic signal,” Helal explains. “The essence of different corridors lies in the phase differences between these signals. The core contribution of our paper is the new interaction design between the oscillators, which ensures seamless corridor transitions.”

Helal and his colleagues also developed a workspace trajectory generator, a computing tool that translates the output of oscillators integrated into a hexapod robot into trajectories for its feet, so that these trajectories remain effective during transitions. In initial testing, their proposed control architecture was found to enable stable, efficient, and rapid gait changes in both a simulated and real hexapod robot.

“The most striking results of our research are the harmonious mix of smoothness and speed of the transition,” Helal said. “Essentially, it’s the combination of fluidity and speed that sets our work apart from other previous efforts. We also validated a mapping function that ensures the robot’s foot trajectory remains effective during these transitions.”

The new architecture introduced by this team of researchers could soon be tested in further experiments and applied to robots with different legs, allowing them to quickly adapt to changes in the environment while maintaining their agility.

In their next research, Helal and his colleagues plan to further improve their method, address potential failures and further improve performance when robots encounter particularly challenging terrains.

“Looking ahead, we plan to dive deeper into machine learning to further refine our robot’s adaptability to the environment,” Helal added. “We are particularly excited about exploring interference compensation and integrating pain detection as feedback mechanisms.

“These developments will not only improve the interaction of the robot with its environment, but also pave the way for more autonomous and resilient robotic systems.”

© 2024 Science X Network