Edge AI Computing Solution Enables Self-Driving Capabilities for Autonomous Vehicles


Edge AI Computing Solution Enables Self-Driving Capabilities for Autonomous Vehicles

The Autonomous Vehicle Landscape

Autonomous vehicles have become an indicator of innovation, and a forward-thinking approach to transportation and commute efficiency wherever they are deployed in the world. Incorporating autonomous vehicles into private and public transport services help to enhance the level of service provided and in return, the overall experience and satisfaction of its passengers. Thanks to the recent development of self-driving solutions and supporting technologies developed by major industry players, such as advanced sensor technology, vehicle-to-everything communication systems (V2X) and advanced driver assistance systems (ADAS), autonomous vehicles are gradually becoming a widespread reality for the daily consumer.

Solution Requirements

A leading self-driving solutions provider, with operations in over 22 countries, was in the market for a powerful industrial computer that could comprehensively support the workloads and functionality that its autonomous shuttles aim to offer for its clients globally. The system will serve as the core hardware platform to provide an intelligent, safe and high-performing mobility solution that could transport up to 15 people on private sites and public roads without a driver on board.

In order to support the self-driving system’s data processing capabilities of both images and large sets of data from multiple sources, the solution provider needed an IPC that had both high CPU and GPU computing power to process these heavy workloads with the lowest level of latency. In addition, a dual 4G and 5G module integration, and wide operating temperature functionality were essential for the system to perform seamlessly no matter the location or environment that the shuttles operate in.


Each autonomous shuttle was equipped with four units of the SINTRONES ABOX-5200 series fanless computer – two to support the shuttle’s driving system, one to serve as the router for network connectivity and one to provide on-board infotainment. The ABOX-5200 is a powerful computer from the SINTRONES Edge AI computing product line and leverages high end Intel© Gen8 CoreTM i7 CPU and NVIDIA® GeForce GTX 1060 GPU to accelerate computational workloads, which was essential for an application such as autonomous vehicles. The wide operating temperature range of -40°C to 60°C ensured that the computer system can operate with ease in both low and high temperature environments. In addition, a wide power input range of 9-48V safeguards the system from sudden changes in the vehicle’s power supply.

Driving System

The SINTRONES Edge AI GPU computing solution- the ABOX series offered a diverse set of I/Os that allowed the self-driving system to integrate with and process data from multiple sources such as LiDAR sensors, IP Cameras, GNSS antenna, and rain and light sensors. This enabled the shuttle to perform precise obstacle detection, video recording and remote monitoring, exact positioning, and lights and wipers automation. Such interconnected modules played a key role in allowing the autonomous shuttle to perceive, decide and act on its own according to its environment.

Router for Network Connectivity

A dual 4G and 5G module integrated technology provided versality for the shuttle to utilize the network available at the deployed region. As some regions across the world have yet to achieve 5G coverage, it was a must for the ABOX solution to be able to operate in both 4G and 5G enabled regions.

On-board Infotainment

The ABOX supported four HDMI® ports and with 4G/5G network connectivity, passenger infotainment programs are seamlessly delivered to each of the displays on the shuttles.


SINTRONES’ team of engineers with over 15 years of industry experience evaluated the client’s complex design requirements and delivered a clear proposal in a short span of time. The team was also able to offer a cost-effective version of the standard ABOX-5200 model to fulfill the unique hardware requirements of the client. This not only provided a customized solution for the specific application but was also considerate of the client’s budget. The strong technical support and ensured quality of SINTRONES computers also earned the trust and confidence of the client.

With SINTRONES’ powerful Edge AI computing solution, the self-driving solution provider is able to offer mobility solutions that are transforming the autonomous vehicle landscape. 50 units of the ABOX-5200 has now been deployed as of July 2021.

To learn more about SINTRONES Edge AI computing solutions and product offerings, please visit https://sintrones.com/ai-gpu-computing.html or email sales@sintrones.com.