The four-legged robot ANYmal is now able to independently climb mountains. In fact, it recently scaled the 1,098-meter summit of Etzel on the southern side of Lake Zurich in just 31 minutes – without falling once or making any other errors.
This was all made possible thanks to a new control system developed by researchers at the Swiss Federal Institute of Technology in Zurich (ETH). Machine learning techniques have made it possible for the robot to combine its visual perception of the environment with its sense of touch for the first time. In this way, ANYmal does not have to rely on its cameras, but can rather use sensor data at the same time.
“The robot has learned to combine visual perception of its environment with proprioception – its sense of touch – based on direct leg contact. This allows it to tackle rough terrain faster, more efficiently and, above all, more robustly”, explains Marco Hutter, Professor of Robotics at ETH Zurich, in a press release.
Before attempting to scale the mountain, ANYmal prepared for its mission in a virtual training camp. In this way, the neural network of the robot’s control system learned how the robot could optimally overcome obstacles and when it could rely on environmental data.
However, in future ANYmal will not merely be restricted to mountaineering. “With this training, the robot is able to master the most difficult natural terrain without having seen it before”, comments Hutter. This would allow the robot to assist emergency services in the aftermath of an earthquake or forest fire, for example.
ANYmal itself was also developed at ETH Zurich. The robot is being commercialized by the ETH spin-off ANYbotics, which is headquartered in Zurich.
Watch the video:
Related news
Meet with an expansion expert
Our services are free of charge and include:
- Introduction to key contacts in industry, academia, and government
- Advice on regulatory framework, taxes, labor, market, and setting up a company
- Custom-made fact-finding visits, including office and co-working space