A million-dollar robot named Cassie set a record when it ran a 5k (3.1 miles) in just under an hour.
It fell down twice.
It looks a lot like an ostrich, or at least the lower half of an ostrich. It was built at Oregon State University and it used machine learning to figure out how to stay upright as it navigated uneven terrain. Did we mention that it fell down twice? It was accompanied by a pack of students who helped it get up and get started again.
Two legs good
One of the professors is co-founder of the new company which will produce the Cassie robot. “This technology will simply explode at some point, when we create vehicles so automated and robots so efficient that deliveries and shipments are almost free,” he said.
The key is the use of legs rather than wheels. Legs can carry a robot into a lot of places where wheels won’t go. Cassie is lighter than earlier prototypes and has a sealed system that allows it to work in inclement weather.
The makers foresee autonomous vehicles carrying Cassie to homes and workplaces, where Cassie will carry deliveries into the buildings.
The new machine was presented at Amazon’s exclusive MARS (Machine-Learning Automation, Robotics and Space Exploration) conference. The makers showed off its ability to steer and to fall down without breaking — an important skill for robots.
Indramat motion control
Indramat technology was game-changing when it was presented in the 20th century, and the components are sealed for use in harsh environments. That’s about all it has in common with Cassie. If you need support or service for Indrmat components, we can help you get up and running again fast…even if you never run a 5k.