Robotic dog learned to walk on its own in just an hour

Scientists conducted a study to find out how animals learn to walk and not stumble. They built

a four-legged, dog-sized robot that helped them figure out the details.


By learning to walk in just an hour, the robot was able tosuccessfully use complex leg mechanics. The Bayesian optimization algorithm in it directs learning: foot sensor data is compared with target data from a simulated virtual spinal cord. The robot learns to walk by constantly comparing sent and expected information and adapting its motor control.

It is noted that the robot was able to optimize its movements faster than the animal. His computer consumes only five watts of power while walking.

“Our robot is, as it were, “born”, knowing nothing aboutthe anatomy of their legs and how they work,” the scientists explained. “We cannot easily examine the spinal cord of a living animal. But we can simulate it in a robot... This is fundamental research at the intersection of robotics and biology. The robotic model gives us answers to questions that biology cannot answer.”