Designing the programs and algorithms that control the robo-beasts is something that remains a challenge for engineers. There are a lot of moving parts and the suite of sensors, contacts and cameras all relay a lot of information at once, and the robot has to sift through these rapidly just to walk or stay upright.

Now engineers at Robotics Systems Lab are making great strides in allowing these types of robots to learn and recover from changing environments. Their article, published in Science Robotics, details their latest achievements in 'smart' legged robotics control.

Here is the abstract, click the link above for the full article:

Legged robots pose one of the greatest challenges in robotics. Dynamic and agile maneuvers of animals cannot be imitated by existing methods that are crafted by humans. A compelling alternative is reinforcement learning, which requires minimal craftsmanship and promotes the natural evolution of a control policy. However, so far, reinforcement learning research for legged robots is mainly limited to simulation, and only few and comparably simple examples have been deployed on real systems. The primary reason is that training with real robots, particularly with dynamically balancing systems, is complicated and expensive.

In the present work, we introduce a method for training a neural network policy in simulation and transferring it to a state-of-the-art legged system, thereby leveraging fast, automated, and cost-effective data generation schemes. The approach is applied to the ANYmal robot, a sophisticated medium-dog–sized quadrupedal system. Using policies trained in simulation, the quadrupedal machine achieves locomotion skills that go beyond what had been achieved with prior methods: ANYmal is capable of precisely and energy-efficiently following high-level body velocity commands, running faster than before, and recovering from falling even in complex configurations.