LG Innotek (011070.KS) is partnering with Nvidia to develop sensing technology for the eyes and hands of humanoid robots. The move aims to expand the company's business into the robot components market, building on expertise accumulated in smartphone cameras and self-driving car sensors.
LG Innotek said on the 22nd that it unveiled camera modules and a multimodal vision system for robots at GTC Berlin, Nvidia's technology event held in the German capital on the 20th.
The products include a 3-megapixel camera module and a sensing system that combines cameras, lidar and radar. They help robots determine the position, distance and movement of nearby objects. The system applies Nvidia's Holoscan Sensor Bridge technology to rapidly transmit sensor data to artificial intelligence computers. By cutting transmission time, it allows robots to recognize and respond to their surroundings more quickly.
The two companies agreed to broaden their cooperation to sensing systems that serve as a humanoid robot's vision, as well as technology for robotic hands that grip and manipulate objects. They also plan to develop sensing products for Nvidia's next-generation AI platform and unveil them at CES and GTC next year.
LG Innotek has been expanding into robotics and autonomous driving in recent years. Last year, it agreed to jointly develop a vision sensing system for robots with U.S.-based Boston Dynamics. In July this year, it began mass production of camera modules at its plant in Paju, Gyeonggi Province, for supply to a North American robot maker. In autonomous driving, it is working with U.S. lidar maker Aeva and self-driving software firm Applied Intuition.
"This partnership is a good opportunity to expand LG Innotek's differentiated integrated sensing solutions into the global market," said Moon Hyuk-soo, chief executive of LG Innotek.
