Spread the love

This robot is a parkour pro: would Boston Dynamics have anything to worry about ?

© ANYbotics

When we talk about robotics and agility, we necessarily think of Boston Dynamics and its famous robots. The biped Atlas in particular, astonishing by his physical capacities surpassing those of most human beings. Little Spot and his four legs is also one of the stars of the company. However, Boston Dynamics is not alone in the arena. ANYbotics, a Swiss company founded in 2016 by doctoral students and researchers from ETH Zurich also has reason to be proud of its robot.

Called ANYmal, it has been in development for several years and today demonstrates quite exceptional skills. A few days ago, ETH Zurich shared a video showing the progress of its little quadruped.

A new Swiss parkour champion

Designed by a team supervised by Marco Hutter, ANYmal was already known for its ability to move on the rocky terrain of Swiss hiking trails. More recently, he has acquired additional agility skills.

Nikita Rudin , a doctoral student with a passion for parkour, is part of the ANYmal design team. The latter explains: “ Before the start of the project, several of my fellow researchers believed that legged robots had already reached the limits of their development potential .” He continues: “But I had a different opinion. In reality, I was convinced that it was possible to do much more with the mechanics of legged robots “.

< p>As we can see in the video below, ANYmal can now overcome a series of obstacles with astonishing ease. It can maintain its balance even on very complex terrain.

Subscribe to Lemon Squeezer

The power of machine learning and model-based

If ANYmal is so effective, it's because the approach adopted by its development team combines both machine learning and a design method called model-based. Thanks to machine learning, ANYmal learns from its interactions with its environment, which allows it to refine its navigation techniques.

At the same time, control model-based gives him a framework based on pre-established physical laws, so he can understand how his actions influence his movement in the environment. He therefore uses mathematical models to plan his movements more precisely, anticipating how each action he takes will affect him. This synergy between the model-based and machine learning allows it to overcome obstacles more intelligently and demonstrate greater strength. ;#8217;adaptation.

Development team member Fabian Jenelten helped apply these advanced techniques to ANYmal . He explains: “ Combining the two approaches allows us to get the most out of ANYmal ”. Obviously, if ANYmal has shown particular care in its development, it is not simply for antics.

What is ANYmal for ?

With such good skills, ANYmal will be the ideal candidate to intervene in hostile environments or difficult to access for humans. Construction sites, collapsed buildings following a disaster, disaster areas, complex industrial facilities, etc. The range of scenarios in which it can intervene is theoretically quite wide.

According to the information provided by the official ANYbotics website, this brave little robot is not not just a parkour ace. It is also equipped with advanced devicesallowing it to inspect its environment with precision: 4K camera, thermal camera, ultrasonic microphone, powerful LED projector and Lidar (light detection and ranging), a laser-based remote sensing technology.

Its total operating autonomy is 90 minutes and it is capable of returning completely autonomously to its base to recharge. It takes less than two hours to regain 70% of its battery and three hours for a full charge. It can also carry a load of up to 10 kg.

ANYmal proves that Boston Dynamics is not the only company qualified to design advanced robots. Despite its young age, the Swiss company already has solid arguments to carve out a place in the sun. It also manufactures another model, ANYmal X, specialized in environmental inspection even more hostile.

  • ANYmal is a quadruped robot manufactured by the Swiss company ANYbotics .
  • Thanks to machine learning and model-based design, he is able to move very easily in rugged environments.
  • It can be used in the inspection of areas that are difficult to access or dangerous to be accessed human.

📍 To not miss any news from Presse-citron, follow us on Google News and WhatsApp.

[ ]

Teilor Stone

By Teilor Stone

Teilor Stone has been a reporter on the news desk since 2013. Before that she wrote about young adolescence and family dynamics for Styles and was the legal affairs correspondent for the Metro desk. Before joining Thesaxon , Teilor Stone worked as a staff writer at the Village Voice and a freelancer for Newsday, The Wall Street Journal, GQ and Mirabella. To get in touch, contact me through my teilor@nizhtimes.com 1-800-268-7116