While most emerging humanoid robot technologies rely on the latest in high-tech sensors to keep themselves oriented to the space around them, Fourier Intelligence is utilizing a simpler approach, integrating their GR-1 robot with six RGB cameras to “see” its environment. The data received via the cameras is then fed to a neural network, called the “Fourier Perception Network,” which “then translates data into a three-dimensional occupancy grid, helping GR-1 navigate passable and impassable areas.”
WATCH: Astribot’s S1 robot completes tasks with remarkable speed and precision
“With our pure vision solution, GR-1 is poised to play a pivotal role in diverse applications such as medical rehabilitation, family services, reception and guidance, security inspection, emergency rescue, and industrial manufacturing,” said Fourier’s Roger Cai.
Fourier says that this approach accomplishes “enhancing GR-1’s environmental perception, achieving safer and more efficient operations with human-like precision.” The company claims that tests have been a success, with the robot showing “high efficiency and accuracy” in navigating the world around it, including the detection and avoidance of vehicles and other objects.
