January 15, 2021

Grade

SAN FRANCISCO, Nov. 23, 2020 /PRNewswire/ — Automotive Grade Linux (AGL), an open source project developing a shared software platform for in-vehicle technology, today announced the latest code release of the AGL platform, UCB 10, also known under the codename “Jumping Jellyfish.”

Developed through a joint effort by dozens of member companies, the AGL Unified Code Base (UCB) is an open source software platform that can serve as the de facto industry standard for infotainment, telematics and instrument cluster applications.

“The AGL platform is Yocto-based, and for Jumping Jellyfish, we updated to Yocto’s first Long Term Support (LTS) release named Dunfell,” said Dan Cauchy, Executive Director of Automotive Grade Linux at the Linux Foundation. “This is significant as it means the Yocto Project will provide patches, fixes, and updates for an extended period of time, something that is essential for automotive systems, which have a longer life cycle than many other embedded Linux applications.”

UCB 10/Jumping Jellyfish includes an operating system, board support package, middleware, application framework and application APIs. Notable updates to the AGL platform include:

  • Updated to Yocto Project’s first Long Term Support (LTS) release, 3.1 “Dunfell”
  • New WindowManager ‘agl-compositor’ replacing agl-service-windowmanager
  • Reworked autobuild scripts (update in services/apps required)
  • Board Support Package updates including i.MX8 support

The full list of additions and enhancements to UCB 10 can be found here.

Automotive Linux Summit
Automotive Linux Summit (ALS)  (ALS) will take place virtually on December 2-4, 2020, co-located with Open Source Summit Japan. Join thousands of open source developers, technologists and automotive industry leaders for 75+ sessions discussing new open source developments, best practices and insights across automotive security, virtualization, infotainment, instrument cluster, connectivity, Vehicle-to-Infrastructure (V2X), and more. All sessions are taking place in the Japan Standard Time Zone, with captioning & translation available in 16 different languages including Japanese, Chinese, Korean and more for every session. View the full schedule here: http://bit.ly/ALS_schedule and register for US$50 at http://bit.ly/ALS_register.

About Automotive Grade Linux (AGL)
Automotive Grade Linux is a collaborative open source project that is bringing together automakers, suppliers and technology companies to accelerate the development and adoption of a fully open software stack for the connected car. With Linux at its core, AGL is developing an open platform from the ground up that can serve as the de facto industry standard to enable rapid development of new features and technologies. Although initially focused on In-Vehicle-Infotainment (IVI), AGL is the only organization planning to address all software in the vehicle, including instrument cluster, heads up display, telematics, advanced driver assistance systems (ADAS) and autonomous driving. The AGL platform is available to all, and anyone can participate in its development. Automotive Grade Linux is hosted at the Linux Foundation. Learn more at automotivelinux.org.

About the Linux Foundation
Founded in 2000, the Linux Foundation is supported by more than 1,000 members and is the world’s leading home for collaboration on open source software, open standards, open data, and open hardware. Linux Foundation’s projects are critical

…...

Israeli startup Arbe, which has raised $55 million to date to develop a 4D imaging radar chipset, has today announced exclusively through EE Times that it has now launched its imaging radar processor chip as part of the chipset.

The company said this is the first automotive grade (AEC-Q100) dedicated imaging radar processing chip. The patented chip is capable of processing the raw data generated by 48 receiving channels and 48 transmitting channels, generating 30 frames per second, meeting automotive power constraints. This, it said, is higher than has ever been achieved on an automotive radar processing chip, while doing so in an “efficient and cost-effective manner”.

Additionally, the processor can scale from high resolution to ultra-high resolution and support over 100,000 detections per frame. According to Arbe, this ability to process such a high channel count provides unparalleled performance and safety to the automotive market. The radar processing chip enables the integration of smart detection algorithms, clustering, post processing and SLAM (simultaneous localization and mapping) into the chip. The processor is designed in accordance with the international standard for functional safety (ISO 26262), and the processor enables ASIL B (Automotive Safety Integrity Level) qualification for the radar chip.

Arbe’s chipset produces detailed 4D images, separates, identifies, and tracks objects in high resolution in both azimuth and elevation on top of range and Doppler resolutions, in a long range and a wide field of view, and complemented by AI-based post-processing and SLAM.  The company has also developed its own proprietary millimeter wave automotive grade radar RFIC chipset that includes a transmitter chip with 24 output channels and a receiver chip with 12 input channels. Using a 22FDX FDSOI CMOS process, Arbe’s RF chipset is designed to support TD-MIMO with strong performance characteristics for channel isolation, noise and transmit power.

The CEO of Arbe, Kobi Marenko, said, “The amount of processing capabilities that we incorporated on our radar chipset solution is one that has never been achieved before in automotive radar. Our technology will bring the safety of vehicles to a new level with low power and low cost. We are excited to ship the processer chip to Tier 1 customers as part of a chipset solution that supports their next gen radar system developments.”

The company said its processor provides more processing power, low latency, and low power while cutting the cost to implement a safe radar solution. We delved into these performance claims a little more to qualify them.

First, on processing power, Arbe said the processor is capable of processing 30 Gbps of data, representing a virtual array of over 2300 virtual channels. Today most radars are processing less than  10% of the bandwidth and usually up to 12 virtual channels. With regard to the latency claim, Arbe said 30 fps provides real-time frames every 33 ms, which enables a maximum latency from the end of a point cloud frame until it is received at the main ECU of 34ms.

What about power consumption? Arbe said the chip

…...