It’s called ANYmal, and as the name suggests, it works as if it were. It is a four-legged robot, the size of a dog and with great autonomy. Because locomotion – the act of moving from one place to another – is one of the most difficult skills to develop in machines. Normally, they learn to walk in familiar controlled environments to avoid accidents or falls, but researchers at ETH Zurich Polytechnic in Switzerland have enabled ANYmal to face real-world challenges on its own. The latest version of this robot can climb stairs, walk on very narrow beams and navigate natural environments with vegetation, rocks, snow and all kinds of slippery and complex surfaces that it is not familiar with. The creators published their progress in the magazine Scientific robotics.
To achieve its objectives, ANYmal overcame courses strewn with obstacles on 76,000 square meters of land designed by the researchers. The circuit required a lot of precision in walking or a large dose of strength to cross it without falling. “We reconstructed harsh environments likely found at natural disaster sites, where debris could break down further when walked on, and at construction sites, where there are oil stains that create slippery surfaces,” explains Fabian Jenelten, professor of robotics at the University of Ottawa. ETH Zurich and lead author of the study.
This robot is capable of operating autonomously, that is, it can make decisions and adapt to its environment without constant human intervention. To do this, it is equipped with various cameras, proximity sensors and gyroscopes which allow it to perceive its environment. Their lidars (laser devices) accurately measure distances and create maps in real time. In addition, this machine can detect gas leaks and transport up to 10 kilos.
Quadruped robots learn to walk in two ways: with deep learning methods that allow them to handle different terrains or with traditional methods that achieve more precise movements. So they end up being either robust or accurate, or one of the two, and when faced with the real world, they tend to fail. Jenelten’s team decided to address this weakness in robotics by combining the best of both worlds: the precision of traditional methods and the robustness of deep learning. Thus, ANYmal achieved proprioception, that is to say the awareness of the position of its body – on which humans and animals rely – to move. “In our work, we try to combine both aspects, enabling quadruped robots to conquer new environments that seem relevant for future applications,” explains Jenelten.
ANYmal is not unique. Your distant relative place, from the company Boston Dynamics, is a quadruped robot with very similar characteristics. Ford has acquired some that are already circulating in its factories to track and prevent possible breakdowns on vehicle assembly lines. Then, the defense company Sword Defense Systems installed in these robots a precision rifle that hits targets more than a kilometer away and has thermal vision for shooting in the dark, while others Tasks They moved to rural New Zealand to become sheep farmers.
The inventors of ANYmal believe that in a few years they will be able to actively assist humans in rescue missions, for example after an earthquake or explosion, where, using lidars, they will create maps anticipating landslides or instability. routes. Looking to the future, says Jenlets, “these robots have enormous potential to completely replace humans in dangerous and life-threatening environments.”
Robots: more than a replacement, an assistance
Rodolfo Haber, director of the CSIC Automation and Robotics Center, suggests considering that these robots “more than a replacement for humans, they will be of help and support”. In larger cities, he explains, they can perform support functions for people with hearing, visual and other disabilities. And machines will not necessarily replace workers: “A robot can help a person who works, for example, in construction to reach 60 years stronger,” explains Haber.
Locomotion with legs “was very fashionable” in certain years, according to Haber. In Spain, for example, CSIC researchers created six-legged machines that performed mine detection tasks. Many quadruped robot projects have been abandoned, but the researcher believes that advances in artificial intelligence can improve the models, so many companies and industries will take them up and there will be a large market at the industrial and social level .
The creators and robotics expert at CSIC believe that advances in ALL locomotion are valuable for the future of legged robots. Given these advances and the constant evolution of technology and artificial intelligence, it is likely that in a few years it will be common to walk the streets among four-legged robots, which assist people in all types of tasks.
Subscribe to continue reading
Read without limits