Robot Era puts humanoid biped to world’s ‘first outdoor snow test’ on AI capacity

The going gets tough when the contraption is put in an unfamiliar, natural environment, where the robot has to figure out the next step on its own.

Robot Era (星动纪元), a Chinese general-purpose robot startup, recently claimed to have become the world’s first robotic developer to realize end-to-end reinforcement learning of its humanoid robot in an outdoor snowy environment.

In a video clip released by the Beijing-based firm, its humanoid robot, named Xiao Xing, is seen walking bipedally on a snow-covered strip of land somewhere in the Chinese capital.

At certain points of the footage, Xiao Xing ascends and descends a gentle slope without slipping or tripping over snow.

In a separate scene, the silver-clad robot climbs up and down a set of makeshift, rickety stairs.

Given what is shown in the video footage, Xiao Xing doesn’t seem to have much difficulty keeping its balance even when it moves down a somewhat steep rise from the top of the steps.

Real-world adaptability

Walking upright with dexterity and stability has been a basic test of motion control for all builders of humanoid robot.

The going gets tough when the contraption is put in an unfamiliar, natural environment, where the robot has to figure out the next step on its own.

This type of “blind” walking test is designed to examine the machine learning abilities of the robot, specifically, how its built-in environmental perception and locomotion algorithms can perform independently of human intervention.

End-to-end reinforcement learning

“Walking on snowy land has always been a tremendous challenge for robot,” said Xi Yue, a senior executive at Robot Era, which was founded in August 2023. “This is because snow on the ground is soft and slippery, and walking stably on it is even hard for humans.”

Snow test thus will put high requirements on the self-balancing abilities and kinematics of humanoid robots, as well as their fast adaptability to the environment.

Xi told media that Xiao Xing did not rely on pre-programmed algorithms but learned to navigate the snow test completely through AI learning.

He explained that end-to-end reinforcement learning will enable Xiao Xing to interact with the surroundings in real time and autonomously adapt to different terrains, including snowy land.

“This not just enhances the robot’s adaptability in an inclement environmental condition, but also attests to the sophistication of Robot Era’s algorithms in the field of humanoid robot, in particular the exceptional and innovative traits of its artificial general intelligence.”

An ‘8-year-old’ and its ‘grown-up’ variant

Robot Era is one of a dozen Chinese tech champions and up-and-comers that either doubled down on or ventured into humanoid robotics last year, lured by the prospect of their wider adoption in a number of areas.

As the company’s first product, Xiao Xing met the public for the first time in July 2023 at the World Artificial Intelligence Conference in Shanghai.

The robot, with the stature of an seven-year-old, stands 1,210mm tall and weighs in at 28kg.

Its upgraded “grown-up” version Xiao Xing Max, is almost two heads taller, coming in at a height of 1,662mm and a weight of 46kg.

The other key specs, such as degrees of freedom, torque and payload, remain unknown.

Robot Era was incubated by Tsinghua University’s Institute for Interdisciplinary Information Sciences and Shanghai Qi Zhi Institute. It so far has closed an angel round for an undisclosed amount of money.

Avatar photo
Ni Tao

Ni Tao is the founder and editor-in-chief of cnrobopedia. Prior to cnrobopedia, he had a full decade of experience with a major state-run English-language newspaper as a tech reporter and opinion writer. He is also a communications specialist, having provided consultancy services to established firms like Siemens, Philips, ABinBev, Diageo, Trip.com Group (Nasdaq: TCOM, HK: 9961), Jianpu Technology (NYSE: JT) and a handful of domestic startups. A graduate of Fudan University, he writes widely about China's business and tech scenes and other topics for global publications including South China Morning Post, SupChina, The Diplomat, CGTN, Banking Technology, among others, and tries to impart his experience to students at Fudan University Journalism School, where he is a part-time lecturer. When he's not writing about robotics, you can expect him to be on his beloved Yanagisawa saxophones, trying to play some jazz riffs, often in vain and occasionally against the protests of an angry neighbor. Get in touch with him by dropping a line at nitao0927@gmail.com.

Articles: 789