Why aren’t there more real-life robots like BB-8? It seems roboticists are more concerned with building bots that mimic what we see in the natural world.
But sometimes, the solution found in nature is not the optimal one. In a paper published this week in the journal Nature Communications, a group of roboticists from École polytechnique fédérale de Lausanne and Université de Lausanne have bucked this biomimicry trend.
The Swiss scientists have designed a six-legged robot that walks differently from insects. What’s more, it moves faster than it would if its gait mimicked that of real insects.
The researchers monitored flies and found that they walk with what’s called a tripod gait, in which the insect’s front and rear leg on one side move forward at the same time as the center leg on the other side. This helps the fly maintain contact when climbing vertically or upside down. But it doesn’t allow the fly to walk very quickly on horizontal surfaces when it’s right-side-up. So the scientists designed their robot to move with a bipod gait instead, moving its legs in three pairs, loping along like a dog with an extra pair of legs. It turns out that a six-legged robot moves faster when it uses a bipod gait than when it used a tripod gait.
Interestingly, when the scientists put a polymer on the flies’ feet to block their stickiness, the flies started to tend towards a bipod gait. Ted Pavlic, who researches engineering and animal behavior at Arizona State University — who was not involved in this research — says the bipod gait is probably more efficient.
“That’s how both the insects and the simulation manage to find the same solution,” Pavlic tells Inverse.
In other words, he thinks this evidence shows that the insects can adapt toward the most efficient solution, a solution like the one the scientists programmed into the robot, even if the flies haven’t evolved that way.
So whether it’s the Boston Dynamics cheetah, the walking legs of Agility Robotics’ “Cassie”, or Sellafield’s giant insects, modern robots that bear a resemblance to the natural world might not always be the most efficient solution.