HomeTechnologyArtificial intelligenceWatch this robot dog scramble over tricky terrain just by using its...

Watch this robot dog scramble over tricky terrain just by using its camera

Unlike existing robots on the market, such as Boston Dynamics’ Place, which moves using internal maps, this robot just uses cameras to guide its movements in the wild, said Ashish Kumar, a graduate student at UC Berkeley who is one of the authors of a paper detailing the work; it will be presented on the Robot Learning Conference next month. Other attempts to use signals from cameras to guide robot movements were limited to flat terrain, but they succeeded in getting their robot to climb stairs, climb rocks and jump over gaps.


The four-legged robot is first trained to move through different environments in a simulator so that it has a general idea of ​​what it’s like to walk in a park or climb up and down stairs. When deployed in the real world, images from a single camera on the front of the robot guide its movement. The robot learns to adjust its gait to navigate things like stairs and uneven ground using Reinforcement Learning, an AI technique that allows systems to improve through trial and error.

Removing the need for an internal map makes the robot more robust, as it is no longer constrained by possible errors in a map, said Deepak Pathak, an associate professor at Carnegie Mellon, who was part of the team.

It’s extremely difficult for a robot to convert raw pixels from a camera into the kind of precise and balanced motion it needs to navigate its environment, said Jie Tan, a research scientist at Google who was not involved in the study. He says the work is the first time he’s seen a small and inexpensive robot demonstrate such impressive mobility.

Must Read
Summit explores role of ethics in development of artificial intelligence

The team has made a “breakthrough in robot learning and autonomy,” said Guanya Shi, a University of Washington researcher who studies machine learning and robot control, who was also not involved in the study.

Akshara Rai, a research scientist at Facebook AI Research who works on machine learning and robotics, and was not involved in this work, agrees.

“This work is a promising step towards building such astute robots with legs and deploying them in the wild,” Rai said.

While the team’s work is helpful in improving the way the robot walks, it doesn’t help the robot predetermine where to go, Rai says. “Navigation is important for deploying robots in the real world,” she says.

More work is needed before the robot dog can prance through parks or get things into the house. While the robot can understand depth through the front camera, it can’t handle situations like slippery ground or tall grass, says Tan; it can step in puddles or get stuck in mud.



Please enter your comment!
Please enter your name here

Most Popular

Recent Comments