Seoul Robotics, a software company developing autonomous solutions, is using Nvidia’s graphics processing units (GPUs) to develop an autonomous valet service that could be used at malls, airports, parking garages and more.
Seoul Robotics’ platform, called Level 5 Control Tower, collects 3D data from the environment using cameras and lidar. Computer vision and deep learning analyze the data and determine the most efficient and safest paths for vehicles within covered areas.
The system allows cars to move autonomously by directing their vehicle-to-everything (V2X) technology that passes information from a vehicle to infrastructure, other vehicles and surrounding entities. The Level 5 Control Tower is a mesh network of sensors and computers placed on infrastructure around a facility such as buildings or light poles to capture unobstructed views of the environment.
The platform is built using Nvidia’s Cuda libraries for creating GPU-accelerated applications as well as the Jetson AGX Orin module for artificial intelligence (AI) at the edge. The GPUs are used in the cloud for global fleet path planning.
“Instead of outfitting the vehicles themselves with sensors, we’re outfitting the surrounding infrastructure with sensors,” said Jerone Floor, vice president of product and solutions at Seoul Robotics.
The idea is interesting and different than other systems that use the car as the main focal point of the autonomous features. Now the vehicles would still need to have some sort of control mechanism and sensors to connect with the infrastructure, but it might be a cheaper way of introducing initial autonomous functions to public areas.
Using V2X, the platform sends commands from infrastructure to cars, making the vehicles turn right or left, move from one point to the next, brake and more. Accuracy in positioning is about 4 cm, Seoul Robotics said.
“No matter how smart a vehicle is, if another car is coming from around a corner, for example, it won’t be able to see it,” Floor said. “LV5 CTRL TWR provides vehicles with the last bits of information gathered from having a holistic view of the environment, so they’re never ‘blind.’”
