Imagine driving a car – one without self-driving capability – to a mall, airport, or parking lot, and using an app to send the car to park.
Software company Seoul Robotics uses NVIDIA technology to make this possible by turning non-self-driving cars into self-driving vehicles.
Based in Korea, the company is initially focused on improving first and last mile logistics, such as parking. Its Level 5 control tower is a mesh network of sensors and computers placed on infrastructure around a facility, such as buildings or streetlights – rather than on individual cars – to capture an unobstructed view of the environment.
The system allows cars to move autonomously by directing their vehicle to any, or so-called V2X communication systems. These systems transmit information from a vehicle to the infrastructure, to other vehicles, to all surrounding entities – and vice versa. V2X technology, which equips many modern cars, is used to improve road safety, traffic efficiency and energy saving.
Seoul Robotics’ platform, dubbed LV5 CTRL TWR, collects 3D data from the environment using cameras and lidar. Computer vision and deep learning-based AI analyze the data, determining the most efficient and safest paths for vehicles in the covered area.
Then, thanks to V2X, the platform can manage a car’s existing features, such as adaptive cruise control, lane keeping and brake assist, to move it safely from one place to another. place to place.
LV5 CTRL TWR is built using NVIDIA CUDA libraries for creating GPU-accelerated applications, as well as the Jetson AGX Orin module for high-performance AI at the edge. NVIDIA GPUs are used in the cloud for global fleet path planning.
Seoul Robotics is a member of NVIDIA Metropolis – a partner program centered around an application framework and set of developer tools that power vision AI applications – and NVIDIA Inception, a free global program that supports startups in point.
Autonomy thanks to the infrastructure
Seoul Robotics is paving the way for a new path to Level 5 Autonomy, or full driving automation, with the so-called “autonomy via infrastructure.”
“Instead of equipping the vehicles themselves with sensors, we are equipping the surrounding infrastructure with sensors,” said Jerone Floor, vice president of products and solutions at Seoul Robotics.
Using V2X capabilities, LV5 CTRL TWR sends infrastructure commands to cars, turning vehicles right or left, moving from A to B, braking and more. It achieves a car positioning accuracy of plus or minus four centimeters.
“No matter how smart a vehicle is, if another car is coming from a corner, for example, it won’t be able to see it,” Floor said. “LV5 CTRL TWR provides vehicles with the latest information gathered through a holistic view of the environment, so they are never ‘blind’.”
These communication protocols already exist in most vehicles, he added. LV5 CTRL TWR acts as the AI-powered brain of instructive mechanics, requiring nothing more than a firmware update in cars.
“Early on, we knew we needed deep learning in the system in order to achieve the very high performance required to meet security goals – and for that, we needed GPU acceleration,” Floor said. “So we designed the system from scratch based on NVIDIA and CUDA GPUs.”
NVIDIA CUDA libraries help the Seoul Robotics team render massive amounts of data from 3D sensors in real time, as well as accelerate training and inference for their deep learning models.
As a member of Metropolis, Seoul Robotics received early access to software development kits and the NVIDIA Jetson AGX Orin for advanced artificial intelligence.
“The Jetson AGX Orin’s compute capabilities allow us to get the LV5 CTRL TWR to cover more area with a single module,” Floor added. “Furthermore, it handles a wide temperature range, allowing our system to operate in both indoor and outdoor units, rain or shine.”
Deployment around the world
The LV5 CTRL TWR is in early commercial deployment at a BMW manufacturing facility in Munich.
According to Floor, cars often have to change locations once they’re made, from electrical repair stations to parking lots for test drives and more.
Equipped with the LV5 CTRL TWR, the BMW factory automated this movement of cars, saving time and money. Automating car transfers also improves employee safety and frees them up to focus on other tasks, like headlight alignment and more, Floor said.
And from the time a vehicle is fully manufactured until it is delivered to the customer, it travels an average of up to seven parking lots. Moving cars manually costs manufacturers between $30 and $60 per car per batch, meaning the LV5 CTRL TWR can address a $30 billion market.
The technology behind LV5 CTRL TWR can be used across industries, Floor pointed out. Beyond car factories, Seoul Robotics plans to deploy its platform around the world – in retail stores, airports, traffic intersections and more.
NVIDIA Jetson AGX Orin 32GB production modules are now available.
Learn more about NVIDIA Metropolis and apply to join Creation of NVIDIA.
Featured image courtesy of BMW Group.