Snakes are smooth-movers. Traversing rough terrains comes easily to the reptiles, as evidenced by their pursuit of iguanas up rocks in the iconic Planet Earth II scene. By studying the movements of snakes when scaling obstacles, engineers from Johns Hopkins University have built a mechanical version that can stably climb large steps, proving its potential to navigate real-life obstacles.
“We look to these creepy creatures for movement inspiration because they’re already so adept at stably scaling obstacles in their day-to-day lives,” senior author Chen Li, an assistant professor of mechanical engineering at Johns Hopkins University, said in a statement. “Hopefully our robot can learn how to bob and weave across surfaces just like snakes.”
Bringing together the fields of robotics, biology, and physics, Li’s lab first published a paper on the topic in 2019, which provided the biological inspiration for their current work. Other studies have focused on the movement of snakes on flat planes. However, the team hope their new paper, published in Royal Society Open Science, will help advance the creation of robots to navigate the tricky 3D terrain that often has to be overcome in search and rescue missions.
The researchers observed how the kingsnake, commonly found in both deserts and pine-oak forests, contorted its body when climbing different steps in the lab. Three distinct parts of the snake’s body were involved. The front and rear sections oscillated laterally, like a wave, on the top and bottom of the step providing stability for the moving snake, whereas the middle section remained rigid and hovered parallel to the step it was trying to climb. As the snake traveled, the researchers found that the front body section got longer, the rear part got shorter, whilst the middle section remained roughly the same length.
Varying the friction and height of the steps also impacted the snakes' approach. A taller and more slippery surface caused the snakes to decrease their speed and wiggle their front and rear sections less.
Mimicking this movement in robotic form was not always a smooth ride. The initial design saw the robotic snake climb a small ways before it became unstable and flipped over. To overcome this issue, Qiyuan Fu, a graduate student at Johns Hopkins University and co-author of the paper, added a suspension system, like that in a car, to each body segment of the robot. This enabled the segments to compress onto the step’s surface when needed, which improved its stability.
In the end, the researchers reported that their robot snake climbed a height 38 percent of its body length, at nearly a 100 percent success rate. The other success of this model, compared to previous studies, was getting the robot to maintain its speed. Almost matching the real snake’s speed, a race between the two would be a close call.
“The animal is still far more superior, but these results are promising for the field of robots that can travel across large obstacles,” remarked Li.
Further work to navigate more complex terrain, such as forest floors and earthquake rubble, is the team’s next aim – which will probably involve watching more snakes on a (3D) plane.