Cassie establishes Guinness world record for bipedal robot 100-metre sprint

Cassie, a robot designed at the Oregon State University (OSU) and manufactured Agility Robotics established the Guinness World Record for the fastest 100 metres sprint a bipedal robot. Cassie did this crossing the dance in 24.73 seconds. Interestingly, the robot has no cameras or external sensors.
Cassie has knees that bend like an ostrich’s and was developed students and researchers from a wide range of backgrounds including mechanical engineering, robotics and computer science. It uses machine learning to control its gait on outdoor terrain. In 2021, it had covered a 5-kilometre dance in just 52 minutes.

“We have been building the understanding to achieve this world record over the past several years, running a 5K and also going up and down stairs. Machine learning approaches have long been used for pattern recognition, such as image recognition, but generating control behaviours for robots is new and different,” said graduate student Devin Crowley, in an OSU press statement. Crowley led the effort for the record.

“Cassie has been a platform for pioneering research in robot learning for locomotion. Completing a 5K was about reliability and endurance, which left open the question of, how fast can Cassie run? That led the research team to shift its focus to speed,” added Crowley.
The robot was trained for the equivalent of an entire year in a simulation environment. The researchers were able to compress this into a period of a week using parallelisation, where multiple processes and calculations happened at the same time. Essentially, Cassie went through a wide range of training experiences at the same time. For the record attempt, the researcher worked on optimising Cassie’s gait for speed.
“Starting and stopping in a standing position are more difficult than the running part, similar to how taking off and landing are harder than actually flying a plane. This 100-meter result was achieved a deep collaboration between mechanical hardware design and advanced artificial intelligence for the control of that hardware,” said Alan Fern, artificial intelligence professor at OSU, in a press statement. Fern was part of the research team.