Should you need inspiration to lace up today, take it from Cassie: a bipdal robot that just made history by running its first outdoor 5k.
Built by Oregon State University and Agility Robotics, the robot tackled the distance totally untethered – and on a single battery charge – completing the race in 53 minutes, which is about the same speed as a gentle stroll for humans.
Though admittedly unlikely to beat any human records anytime soon, Cassie’s historic race is all the more remarkable given the robot taught itself to run using a deep reinforcement learning algorithm.
‘Deep reinforcement learning is a powerful method in AI that opens up skills like running, skipping and walking up and down stairs,’ said Yesh Godse, an undergraduate in the Dynamic Robotics Laboratory at Oregon State University.
Running requires you to maintain your balance while switching positions, so Cassie has learned to make infinite subtle adjustments to stay upright on the move. Designed with knees that bend like an ostrich, it’s the first robot to successfully maintain a running gait on outdoor terrain.
‘Cassie is a very efficient robot because of how it has been designed and built, and we were really able to reach the limits of the hardware and show what it can do,’ added Jeremy Dao, a Ph.D. student in same lab.
Unfortunately, the run wasn’t all smooth sailing. Cassie’s race time included around 6.5 minutes of troubleshooting. The engineers dealt with an overheated computer that caused the robot to collapse – relatable in this heatwave – and a high-speed turn that knocked it off its legs.
‘In the not very distant future, everyone will see and interact with robots in many places in their everyday lives, robots that work alongside us and improve our quality of life,’ said robotics professor Jonathan Hurst. So, expect to see Cassie at a parkrun near you in the not-too-distant future.
This content is created and maintained by a third party, and imported onto this page to help users provide their email addresses. You may be able to find more information about this and similar content at piano.io
