Bipedal robot Cassie completes 5k run

Dubbed Cassie, the running robot was developed under the direction of robotics professor Jonathan Hurst with a 16-month, $1m grant from the US Defence Advanced Research Projects Agency (DARPA).

Cassie, the first bipedal robot to use machine learning to control a running gait on outdoor terrain, completed the 5K on Oregon State’s campus untethered on a single battery charge.

Boston Dynamics’ Atlas robot performs parkour

“The Dynamic Robotics Laboratory students in the OSU College of Engineering combined expertise from biomechanics and existing robot control approaches with new machine learning tools,” said Hurst, who co-founded Agility in 2017. “This type of holistic approach will enable animal-like levels of performance. It’s incredibly exciting.”

Cassie, designed with knees that bend like an ostrich’s, taught itself to run using a deep reinforcement learning algorithm. Running requires dynamic balancing and Cassie is said to have learned to make infinite subtle adjustments to stay upright while moving.