ΑΙhub.org
 

Bipedal robot developed at Oregon State learns to run


by
11 August 2021



share this:

Cassie the robotImage courtesy of Jonathan Hurst, Oregon State University.

By Steve Lundeberg

Cassie the robot, invented at Oregon State University and produced by OSU spinout company Agility Robotics, has made history by traversing 5 kilometres outdoors in just over 53 minutes. The robot was developed under the direction of robotics professor Jonathan Hurst with a 16-month, $1 million grant from the Advanced Research Projects Agency of the U.S. Department of Defense.

Since Cassie’s introduction in 2017, OSU students funded by the National Science Foundation have been exploring machine learning options for the robot.

“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, with knees that bend like an ostrich’s, taught itself to run with what’s known as a deep reinforcement learning algorithm. Running requires dynamic balancing – the ability to maintain balance while switching positions or otherwise being in motion – and Cassie has learned to make infinite subtle adjustments to stay upright while moving.

“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,” said Jeremy Dao, a Ph.D. student in the Dynamic Robotics Laboratory.

“Deep reinforcement learning is a powerful method in AI that opens up skills like running, skipping and walking up and down stairs,” added Yesh Godse, an undergraduate in the lab.

Hurst said walking robots will one day be a common sight – much like the automobile, and with a similar impact. The limiting factor has been the science and understanding of legged locomotion, but research at Oregon State has enabled multiple breakthroughs.

ATRIAS, developed in the Dynamic Robotics Laboratory, was the first robot to reproduce human walking gait dynamics. Following ATRIAS was Cassie, then came Agility’s humanoid robot Digit.

“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,” Hurst said.

During the 5K, Cassie’s total time of 53 minutes included about 6.5 minutes of resets following two falls: one because of an overheated computer, the other because the robot was asked to execute a turn at too high a speed.

In a related project, Cassie has become adept at walking up and down stairs.



tags:


Oregon State University

            AUAI is supported by:



Subscribe to AIhub newsletter on substack



Related posts :

The Machine Ethics podcast: Data Collective with E.M. Lewis-Jong

Ben chats to E.M. Lewis-Jong about the promise of AI and making human connection easier, speech recognition and supporting linguistic diversity, making useful technologies that have a purpose, and more.

AI for ethology: an interview with Isla Duporge

  08 Sep 2026
Deep learning is becoming a powerful tool for understanding animal behaviour and tracking populations.

AI in cardiology: The path to practical application carries risks

  07 Sep 2026
Can artificial intelligence help us better combat cardiovascular diseases? Legal researcher Hannah van Kolfschooten urges caution, as there are still many legal issues that need to be resolved.

AI-powered camera system offers low-cost way to monitor bumblebees

  04 Sep 2026
Researchers have developed a semi-automated method that uses remote cameras to survey bumblebees and potentially other insects.

Interview with Noah Golowich – theoretical foundations for learning in games and dynamic environments

  03 Sep 2026
Noah Golowich tells us about his research into the theory of decision making and learning in games, which have applications in Multi-Agent Reinforcement Learning.

Forthcoming machine learning and AI seminars: September 2026 edition

  02 Sep 2026
A list of free-to-attend AI-related seminars that are scheduled to take place in the next couple of months.
AI pioneers

Combining cultures, from code to canvas: an interview with Ken Goldberg

  01 Sep 2026
AI Pioneer Ken Goldberg on the clash of cultures within robotics, his career bridging art and science, and the meteoric rise of agentic robotics.

What happens when AI runs out of pictures?

AI is data-hungry and needs thousands of images to learn how to detect tumours or product defects, but often very few are available. A new method aims to change that.



AUAI is supported by:







Subscribe to AIhub newsletter on substack




 















©2026.05 - Association for the Understanding of Artificial Intelligence