ADAS vision processor offers both viewing and sensing functions
Indie Semiconductor has expanded its automotive camera video processor portfolio with the commercial release of iND87540 vision processor, a highly integrated system-on-chip (SoC) that enables viewing and sensing capability at the vehicle’s edge.
As government regulators, new car safety assessors and consumers demand higher performance driver and road user safety features, automakers are increasingly looking to camera-based Advanced Driver Assistance Systems (ADAS) that enable volume scalability across their vehicle classes. This demands a ‘distributed intelligence’ architectural approach to vision sensing, and high levels of integration coupled with low power consumption to meet the needs of mass market deployments. The iND87540 vision processor was developed to address these challenging design requirements. iND87540 is also a preferred option by end customers as a pre-processor for powerful central compute and provides real-time image processing for optimal detection performance.
Integrating real-time, on-chip image signal processing (ISP), digital signal processing (DSP) and customized hardware, the iND87540 enables viewing and sensing capabilities within the stringent power, latency and compact form factor needed for scalable vision architectures. The SoC provides computer vision processing that can run a range of algorithms, enabling ADAS functions including pedestrian detection, object detection, blind spot detection, cross traffic alerts as well as driver and occupant monitoring (DMS/OMS). This class-leading SoC hardware is complemented with value-add proprietary high-performance embedded algorithms such as auto calibration (AutoCAL®) and dirty lens detection.
“As OEMs strive to deploy vision-based viewing and sensing across their model ranges, distributed intelligence is emerging as a critical enabler for the proliferation of vision-based ADAS applications. Our launch of the iND87540 is capitalizing on this industry dynamic: delivering high-performance vision processing, without sacrificing the power, cost and size demands of the volume market,” said Abhay Rai, EVP and GM of indie’s Vision Business Unit. “By incorporating real-time video processing with object detection into a single SoC, indie is paving the way for multiple vision-enabled safety and convenience use cases across OEMs’ vehicle classes.”
According to S&P Global, shipments of automotive electronic control units (ECUs) incorporating vision-based processing are expected to grow from 232 million units in 2022, to nearly 400 million units by 2027.
The iND87540 has been developed to ISO 26262 ASIL-B requirements, is AEC-Q100 Grade 2 qualified and hosts the AutoSAR software stack. The SoC is available now for vehicle integration and is currently in advanced development with select Tier 1 customers.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
