Osaka, Japan – Researchers have developed an artificial intelligence system that improves the movement efficiency of snake-like robots, potentially extending their operational missions in disaster zones and exploration environments.

The work was led by Dr. Akio Yamano at the Graduate School of Engineering, Osaka Metropolitan University, focusing on optimizing how snakebots move using deep reinforcement learning.

Snake-like robots are designed to navigate tight spaces, rough terrain, and water surfaces, making them useful for rescue operations in hazardous environments such as earthquake-affected regions.

However, traditional slithering movement requires many motors working in coordination, which drains battery power quickly and limits mission duration.

The research introduces an AI approach supported by an “observation buffer” that processes sensor data, including angular velocity, acceleration, and body state, to stabilize movement and improve straight-line navigation.

The system enables two main motion types. Undulating movement is used for uneven terrain, while a rolling motion is more efficient on flat ground. In rolling mode, the robot reshapes into a circular form and moves by shifting its center of gravity, using gravity to reduce motor demand.

Researchers found that on level surfaces, rolling motion achieved about twice the travel speed per unit of energy compared to undulating movement.

Dr. Yamano explained that combining both motion types can extend operational time in disaster response and planetary exploration missions.

“Our group is developing various interesting capabilities,” he said. “We aim to create robots that autonomously assess the situation and use precise navigation technologies to carry out useful tasks.”

Podcast: Play in new window | Download