Speeding across the ground at up to 9 miles per hour, this four-legged cheetah robot is sure to make you jump.
It was developed by engineers at the Massachusetts Institute of Technology and is even able to quickly overcome ice and gravel.
Billed as “virtually indestructible,” they said he learned to race across different terrain through trial and error, much the same way humans learn.
Thanks to state-of-the-art simulation technology, their mini cheetah robot was able to accumulate 100 days of experience at different sites in just three hours of real time, the researchers say.
Speeding across the ground at up to 9 miles per hour, this four-legged cheetah robot is sure to make you jump.
It was developed by MIT engineers and is even capable of moving quickly over ice and gravel.
Billed as “virtually indestructible”, they said they learned how to navigate various terrains through trial and error.
Almost 25 years have passed since the first animal robots were created, and since then, scientists around the world have identified, among other things, examples of machines that walk, dance and open doors.
One of the most famous is Spot, a mini robot dog from Boston Dynamics, which was filmed last year doing a choreographed dance to the music of South Korean K-pop sensation BTS.
But when it comes to a running robot, MIT researchers explain, it can be quite difficult to program it for difficult terrain.
“We have developed an approach whereby robot behavior is improved based on simulated experience, and our approach also allows these learned behaviors to be successfully applied in the real world,” said lead researcher Gabriel Margolis.
“Of all the environments he sees in this simulator, some will teach robots skills that are useful in the real world. When working in the real world, our controller determines and applies the appropriate skills in real time.”
When asked why it was much more difficult to make the robot run instead of walk, he added: “Achieving fast running requires pushing the hardware to its limits.
“The robot must respond quickly to changes in the environment, such as when it hits ice while running on grass.
When it comes to a running robot, as MIT researchers explain, it can be quite difficult to program it for difficult terrain.
Thanks to state-of-the-art simulation technology, their mini cheetah robot was able to accumulate 100 days of experience at different sites in just three hours of real time, the researchers say.
The researchers said their particular training system was so successful that the robot broke the record for the fastest run recorded for its particular platform.
“People run fast on grass and slow down on ice – we adapt.
“Giving robots a similar ability to adapt requires rapid identification of terrain changes and rapid adaptation to prevent the robot from falling over.”
Co-author Ge Yang added: “Programming how a robot should act in every possible situation is just very difficult.
“The process is tedious because if the robot fails in a certain area, the human engineer will need to determine the cause of the failure and manually set up the robot controller, a process that can require significant human time.
“Trial and error learning eliminates the need for humans to specify exactly how the robot should behave in each situation.”
The researchers said their particular training system was so successful that the robot broke the record for the fastest run recorded for its particular platform.
The previous high speed record was set by MIT’s Biomimetic Robotics Laboratory in 2019, when scientists reached a top running speed of 8.2 miles per hour (3.7 m/s). Latest mini cheetah robot hits over 8.7 mph (3.9 m/s)
A robot of MIT engineers was filmed scurrying around at speeds of up to 9 miles per hour.
When asked why it was much more difficult to make the robot run rather than walk, the researchers replied: “Achieving fast running requires pushing the equipment to its limits.”
The previous high speed record was set by MIT’s Biomimetic Robotics Laboratory in 2019, when scientists reached a top running speed of 8.2 miles per hour (3.7 m/s).
In comparison, the latest mini cheetah robot has reached speeds of over 8.7 miles per hour (3.9 m/s).
“Developing controllers for maneuver maneuvers such as sprinting and high-speed turns with robotic legs is challenging,” the authors write in their paper.
“These tasks require high torques to apply ground forces and control the body at high speed, pushing both the sensing and execution stack in the robot hardware to their limits.
“We are achieving record-breaking agility for the Mini Cheetah, reaching speeds of up to 3.9 m/s.”
The researchers are currently building more low-cost quadrupedal machines, aiming to create a set of 10 pieces, each of which they hope to donate to other labs.
WHAT IS BOSTON DYNAMICS MINI ROBO DOG?
Boston Dynamics first showed off Spot, the most advanced robot dog ever made, in a video released in November 2017.
The firm best known for its 1.7-meter-tall Atlas humanoid robot has unveiled a new “lighter” version of its Spot robot.
A robot dog was shown running around the yard with the promise that more information would be “coming soon” from the infamous secret firm.
“Spot is a small, four-legged robot that fits comfortably in an office or home,” the company’s website says.
It weighs 25 kg (55 lbs) or 30 kg (66 lbs) including arm.
Spot is all-electric and can run for about 90 minutes on a single charge, depending on what it does, the firm says, boasting, “Spot is the quietest robot we’ve built.”
The Spot was first introduced in 2016, and a previous version of the mini version of the Spot, with the odd retractable neck, was showcased helping around the house.
A previous video from the company shows the robot exiting the firm’s headquarters and entering what appears to be a home.
There, he helps load the dishwasher and carries the trash can.
Also at one point, he collides with a fallen banana peel and plummets, but uses his retractable neck to get back up.
“Spot is one of the quietest robots we have ever built,” the company says, “due to its electric motors.
“It has a variety of sensors, including depth cameras, a solid-state gyroscope (IMU), and proprioception sensors in the limbs.
“These sensors help with navigation and mobile manipulation.
“Spot performs some tasks autonomously, but often uses a human for high-level guidance.”