How This Agile, Soccer-Playing Dog Robot Could Lead to Better Future Robotics

You might soon be able to play soccer with robots. 

MIT researchers have developed a legged robotic system that can dribble a ball under the same conditions as humans, according to a newly published paper. It's part of the fast-developing field of nimble robots that could lead to innovations in their use. 

"With Robots becoming more and more agile in design and function, this is potentially having a major impact on our world of work," Tim Warrington, the CEO of robot company Bots.co.uk, told Lifewire in an email interview. "The benefit of agile robots is that they are able to move quickly and precisely in challenging environments, making them ideal for tasks such as search and rescue, manufacturing, and healthcare."

Robots That Move and Play Like Humans

MIT's robot uses a mixture of onboard sensing and computing to traverse different natural terrains such as sand, gravel, mud, and snow and adapt to their varied impact on the ball's motion. Its sensors let it perceive the environment, allowing it to feel where it is, "understand" its position, and "see" some of its surroundings. There are also actuators that let it apply forces and move itself and objects. In between the sensors and actuators sits the computer, or "brain," tasked with converting sensor data into actions, which it will apply through the motors. 

The robot starts without knowing how to dribble the ball. It only receives positive reinforcement when it does right. When the robot runs on snow, it cannot see the snow but can feel it through its motor sensors. But soccer is trickier than walking, so the team used cameras on the robot's head and body for a new sensory modality of vision, in addition to the new motor skill. 

"Our robot can go in the wild because it carries all its sensors, cameras, and computers on board. That required some innovations in terms of getting the whole controller to fit onto this onboard computer," MIT researcher Gabe Margolis said in a news release. "That's one area where learning helps because we can run a lightweight neural network and train it to process noisy sensor data observed by the moving robot. This is in stark contrast with most robots today: Typically a robot arm is mounted on a fixed base and sits on a workbench with a giant computer plugged right into it. Neither the computer nor the sensors are in the robotic arm! So, the whole thing is weighty, hard to move around."

The best-known agile robot is probably the Boston Dynamics "Atlas" humanoid robot, which may be the world's most advanced robot, Brendan Englot, the Director of the Stevens Institute for Artificial Intelligence at Stevens Institute of Technology, pointed out in an email. There are also agile quadruped robots, such as the MIT Mini Cheetah. 

"Aerial robots can be just as agile as our most advanced aircraft technology—the Yamaha Fazer unmanned helicopter is a great example," Englot said. "Underwater robots have some catching up to do, but the BIOSwimmer from Boston Engineering, which combines a fish-inspired design with a propeller thruster at the end of its tail, is a great example of how underwater robots can be agile as well."

Agile robots are necessary because of how agile humans are, Englot said. "If we want intelligent, reliable, and safe robots that can work productively alongside humans (or help keep humans out of harm's way), these robots should be just as agile as their human co-workers are," he added. 

The Future: Robots and AI

In the near future, nimble robots like the one invented by MIT will get faster while still being safe to work around humans and will be easier to instruct or program, Bob Rogers, the CEO of Oii.ai, a data science company, said via email. 

"Imagine hooking up a natural language capability like ChatGPT to an agile robot: rather than encoding instructions in QR codes on objects, these robots will be able to take verbal instructions and ask questions when needed," he added.

Robots have been primarily used for relatively simple tasks, Warrington said. However, with increased sophistication, they are being used for more complex tasks requiring more agility and coordination, such as performing surgery, cleaning up hazardous waste, and even delivering packages.

"And with the latest and much reported advances in artificial intelligence and machine learning, together with the increased agility, [we] will see real-time decision making," he added. "At the same time, robots will navigate through even more challenging environments, potentially opening up opportunities in medicine, exploration, and, of course, the one we don't like to think about, the military."