Legged robotic systems are attractive alternatives to tracked/wheeled robots for applications with rough terrain and complex cluttered environments. The freedom to choose contact points with the environment enables them to overcome obstacles comparable to their leg length. With such capabilities, legged robots may one day rescue people in forests and mountains, climb stairs to carry payloads in construction sites, inspect unstructured underground tunnels, and explore other planets. Legged systems have the potential to perform any physical activity humans and animals are capable of.
Using a neuralnetwork computer system designed by ETH, the ANYmal robot “learned” to run faster and recover from falls in various positions, a skill not previously observed in other four-legged robots of its kind.