Cloud Horse Intelligent Transportation: Building Autonomous Vehicle Infrastructure with NVIDIA Jetson

Cloud Horse Intelligent Transportation is a technology innovation company focused on L4 level autonomous driving technology in urban scenarios. While pursuing technological innovation, it is also facing a series of challenges:

• Control algorithms for autonomous vehicles, increasing computational power requirements

• Flexible deployment to meet different scenarios and needs

• Maintaining reliability in various challenging work environments

• Network and functional safety requirements

• Shorter R&D cycles and marketization

The domain controller products from Cloud Horse Intelligent Transportation provide a powerful computing platform for applications such as L4 autonomous driving, autonomous industrial transport vehicles, and dedicated work vehicles. These domain controller products have flexible specifications and can integrate up to four NVIDIA Jetson Orin™ system modules, with computational power exceeding 1000 TOPS, and feature high redundancy, modularity, and scalability design.

Different application scenarios, including smart connected taxis, often require different sensor solutions, so the domain controller needs to be adjusted accordingly. With the support of NVIDIA Jetson, Cloud Horse Intelligent Transportation provides rich system interfaces in its domain controller products to meet this demand.

With the advantages of a high computing power platform, NVIDIA Jetson can create smarter autonomous driving systems for dedicated vehicles. Its outstanding stability and reliability allow Cloud Horse Intelligent Transportation’s domain controllers to simplify existing HPC/IPC-based solutions and achieve faster deployment.

Additionally, Jetson supports domain controllers to achieve excellent redundancy design and integrate network security and functional safety design into the overall design.

Cloud Horse Intelligent Transportation’s domain controller products have been deployed in smart connected taxis and unmanned cleaning vehicles. These vehicles utilize computer vision and autonomous planning algorithms to enhance efficiency and handle extremely complex road environments in first-tier cities like Shanghai. They also integrate multiple AI models, including pure vision multi-task Transformer large models, point cloud object detection models, and front-fusion based BEV models. With these models, Cloud Horse Intelligent Transportation has also generated an easy-to-use control platform for users. Smart connected taxis have even begun road testing without human drivers.

With Jetson’s rich interface support and flexible scalability, the domain controller products can adapt to mainstream sensor solutions and adjust configurations to meet the demands of quantitative product design. The rich sensor adaptability allows users to deploy various CIS, LiDAR, millimeter-wave radar, and other sensors in their dedicated vehicles.

NVIDIA Jetson also implements network security and functional safety designs throughout the product lifecycle (including design, development, testing, deployment, and operation). Based on Jetson, Cloud Horse Intelligent Transportation has applied redundancy design in the main controller system, communication system, and power system, supporting fail-safe operations. This has allowed Cloud Horse Intelligent Transportation to pass one of the first driverless road tests in China.

Smart shuttle bus based on Cloud Horse Intelligent Transportation’s domain controller

Contact:[email protected]

GTC 2024 will be held from March 18 to 21, 2024, at the San Jose Convention Center in California, USA, and the online conference will also be open simultaneously.Scan the QR code below to register for the GTC conference.