Why This Startup Created A Deep Learning Chip For Autonomous Vehicles Posted on : Jun 23 - 2018

Israeli artificial intelligence (AI) startup, Hailo Technologies, has closed a $12.5 million Series A from Maniv Mobility, OurCrowd, and NextGear to develop a chip for deep learning on edge devices and processing of high-resolution sensory data in real time.

According to a report from Markets and Markets, edge computing will be worth $6.72 billion by 2020, and IC Insights reported that integrated circuits in cars are expected to generate global sales of $42.9 billion in 2021. Combine those statistics with the automotive industry's need for advanced driver assistance systems (ADAS), and there's a demand for computational power for deep learning algorithms for ADAS.

In 2017, McKinsey reported in the study Self Driving Car Technology: when will robots hit the road? that ADAS systems grew to 140 million in 2016 from 90 million units in 2014.

"Because of the low latency required for autonomous driving and advanced driving assistance, deep learning with convolutional neural networks, running on in-vehicle hardware, is necessary," offers Tom Coughlin, IEEE Fellow and President at Coughlin Associates. "Only a fast, well-trained deep learning system could hope to make the sort of split-second decisions that human drivers make to avoid accidents and thus act as an effective autonomous driving system."

The new deep learning chip from Hailo Technologies is designed to run AI applications embedded on edge devices installed in drones, smart home appliances, AR/VR platforms, and wearables. But, the company said in a press release they will target the mobility market and autonomous vehicles first. Samples of the processors are expected in the market in the first half of 2019.

"We believe that current computing solutions don't stack up for running neural networks (i.e., deep learning) at scale in resource-constrained environments," said Orr Danon, CEO, Hailo Technologies. "Our observation is that the key deficiency is in the architecture of the computer, which was designed for running classical rule-based software. With our technology, it will be possible to bring state-of-the-art deep learning into devices outside the data center at reasonable power and cost. We believe this will enable many interesting use cases, automotive being a leading one." View More