Avoiding data saturation in autonomous driving

November 27, 2017 // By Graham Prophet
Silicon design house EnSilica (Wokingham, UK) has developed a radar imaging co-processor, to address the specific problem of automotive data overload, furthering progress in the development of self-driving cars.

The eSi-ADAS radar Imaging Co-processor will accelerate the commercialisation of autonomous vehicles by solving the current problem of radar data overload and resolution by handling this in a dedicated co-processor to enable tracking of potentially hundreds of objects in real time. Its designers claim up to 20x lower power requirements than today’s lower resolution systems; and with lower system costs; as well as increasing speed and accuracy by up to 10x.

EnSilica has until recently primarily acted as a silicon design house and is now managing the entire process right through to supply for many of its customers. The eSi-ADAS radar Imaging Co-processor is one of the next generation products developed by the company to sell under its own branding. This can be either as IP to be integrated onto a customer’s chip, or as a stand-alone chip from EnSilica. This is a major shift in the company's business model.

“Until now, we have been under the radar, so to speak, with this product,” explained David Doyle, EnSilica’s Commercial Director. “Our focus has been quietly promoting our ADAS solution into the automotive market. This has been really successful and elements of our ADAS are shipping in production volume already. The timing is now right to leverage this adoption and take it to manufacturers around the world, capitalising on the success so far.”

Advanced Driver Assistance Systems (ADAS) are becoming increasingly complex as they advance from their simple origins of cruise control and parking assistance to fully autonomous driving. One of the challenges facing OEMs is that a combination of Video, radar and LIDAR is needed to cover all operational conditions, which is expensive. This is because current radar sensors lack the imaging capability, 3D and 360° coverage that is available from LIDAR but LIDAR does not work well at night or in fog so the two have to be used in conjunction with Video. This creates a problem of large amounts of data that needs to be processed in real time.