Axiomtek’s Hailo chips integrated for Smart City Edge AI infrastructure


Based in Taiwan Axiomtek announced it will offer hardware with Hailo’s AI processor chips, the latest in a series of design wins for the edge AI chip startup. Axiomtek plans to target smart city applications with the new system to enable more efficient processing of critical tasks like face recognition and object recognition closer to the place of data acquisition.

The new RSC100 ‘Plato’ is “well suited for use in smart city applications including smart surveillance, smart factory, smart agriculture and smart transportation,” says Ken Pan, product manager at Axiomtek.

The move is the latest in a series of steps leading Hailo processors to systems that support facial biometrics at the edge.

According to market research company IDC, the global market for edge computing hardware is growing reach $ 80.7 billion in 2024. One of the major workloads these systems will run is AI / ML applications. According to research company Omdia, global sales of AI software for smart city applications are expected to increase to $ 4.9 billion by 2025, compared to $ 673.8 million in 2019.

The RSC100 is equipped with flexible and high-performance storage, including 16 GB eMMC onboard as well as expansion slots and interfaces for SSD and NVMe storage. According to Axiomtek, the system has two GbE ports that allow IP cameras and LiDAR connectivity with greater bandwidth for video analytics. Other expansion modules allow customers to add connectivity for their environments such as Wi-Fi / Bluetooth, 5G / 4G / LTE.

Axiomtek points out that the rugged Plato is designed for harsh environments and operates in a temperature range of -20 ° C to + 70 ° C.

Axiomtek Plato

Hailo recently behaved $ 136 million in Series C funding, bringing total venture capital funding to $ 224 million. Its Hailo-8 chip is said to be based on neural networks of the brain, and its “hyper-efficient” processor is better suited for edge workloads than chips that are more commonly found in data centers. It consumes only 2.5W of power while it can perform up to 26 tera operations per second.

Article Topics

AI chips | Axiomtek | Biometrics | Biometrics on the edge | Edge Computing | Face recognition | Hailo | Object recognition | Smart cities


Comments are closed.