AI-Enabled Quadruped Robots Parkour Test Footage

DEEP Robotics has released a video of its quadrupedal robot doing parkour and shifting its center of gravity midair, showcasing how AI enhances locomotion & intelligence for robotics By Abi Wylie / 22 May 2024
AI-Enabled Quadruped Robots Parkour Test Footage
Follow UST

Leading quadrupedal robot developer DEEP Robotics has released a video of its quadruped robot involved in parkour, showcasing the role of AI in enhancing the levels of locomotion and intelligence for robotic systems.

In the video, the Lite3, one of the company’s signature robot dogs, is seen scurrying across a stack of large boxes, leaping across gaps, navigating around stairs, and climbing easily onto an elevated platform about half a meter tall.

A more stunning feat is that the Lite3 can shift its center of gravity when tossed in the air, spin its body and land on its four feet, or quickly roll over after falling on its back.

Robot dogs pre-programmed to vault over objects can struggle in the face of obstacles, and are certainly unable to spin in mid-air. Through a series of algorithmic innovations and reinforcement learning iterations, engineers from DEEP Robotics have given a huge boost to the dexterity, autonomy and self-balancing abilities of Lite 3.

Zunwang Ma, a robot control engineer at DEEP Robotics, explained that they successfully applied the Proximal Policy Optimization (PPO) method combined with depth camera sensor inputs and trained an end-to-end policy. This approach effectively broadens the limitations of traditional model-based control methods, enabling the quadruped robot to traverse high platforms and gaps without requiring precise perception and robot modeling.

In a simulation mode, an avatar of the dog is always kept initially at a slant. By devising policies and setting up rewards to guide the avatar, AI engineers enable the quadruped to shift in the air and maintain its dynamical balance in the virtual world.

Other improved metrics of Lite3 include faster walking speed, topping up at 4 meters per second. When pushed to its limits, the robot can scramble up platforms higher than 50 centimeters, as well as jump over a 80-centimeter-wide gap.

Meanwhile, the X30, a larger, industrial-grade robot dog also produced by DEEP Robotics, amazed visitors with its autonomous adaptation to an array of unstructured, uneven surfaces such as grass, gravel, and rubble.

The X30 stands out with the ability to traverse rugged terrains and navigate random objects with ease. It is designed to withstand external interference like pushing or pulling, maintaining its balance, which proves to be a big advantage in outdoor operations.

Built to IP67 standards, the robot is able to function under the rain, on plateaus or Gobi deserts, or in any challenging environments with huge variances in temperature, altitude or degrees of humidity.

Algorithmic Improvements

Through a series of algorithmic innovations and reinforcement learning iterations, engineers from DEEP Robotics have given a huge boost to the dexterity, autonomy and self-balancing abilities of Lite3 and X30.

The company is looking to incorporate advanced AI technologies into its robotics systems to enable bigger breakthroughs in embodied AI. Application-wise, legged robots from DEEP Robotics have proven practical in an array of use cases including autonomous inspection, emergency detection and rescue.

Overall, thanks to a large amount of data acquired from real-world application of its products, DEEP Robotics has met with relatively fewer difficulties in the sim-to-real transfer, meaning the deployment of robotics algorithms to a physical robot prototype.

Going forward, the company aims to integrate reinforcement learning further into its development to broaden the application scenarios of its robots and explore the next frontiers of AI-powered robotics.

“With reinforcement learning, we can use fewer policies to achieve greater terrain adaptation and coverage, making the robot more robust and intelligent,” Zunwang Ma explained.

Posted by Abi Wylie Connect & Contact
Latest Articles