Accelerate Automotive AI Systems
From ADAS to Autonomous Driving
Faster, Power-Efficient Processing
Automotive AI-enabled applications vary greatly, but they all require stable, reliable, and often complex models. However, ASICs are a part of the automotive control systems that are often passively cooled, creating a challenge to meet functional requirements within power and cost constraints.
Expedera's Origin™ Deep Learning Accelerator (DLA) is a configurable AI solution that addresses the full range of performance and power efficiency requirements for automotive use cases, including ADAS and autonomous driving. Origin meets the low latency requirements of critical systems to deliver real-time performance and enable faster decision making. It can also support multiple models to pull in multiple sensor data streams, including lidar and radar, without compromising performance or accuracy. Origin excels at vision-based use cases and is capable of evaluating both downsampled and high resolution video streams in real time. Consider, for example, a backup camera that might downsample to identify a hazard 2 meters away. The performance requirements for downsampling are very different from a front-facing camera outputting a Full HD video stream which must identify a danger 200 meters away at highway speeds. Expedera’s Origin IP enables automotive system designers to meet real-time performance requirements within tight performance, power and area constraints.
to our News
Sign up today and receive helpful
resources delivered directly
to your inbox.