embedded news: Driver not responsible for self-driving features - Embedded.com

embedded news: Driver not responsible for self-driving features

A roundup of this week’s embedded news including a report from the legal industry on who’s responsible for self-driving features in autonomous vehicles, plus regular features, news, products and funding news which includes a startup applying AI to autonomous vehicles and mobile robotics with its $200 million funding round.

It would seem that many of the electronics trade media and analysts provide a lot of coverage and comment on the advances in self-driving features and autonomous vehicle technology. While there is now a growing realization that full level 5 autonomy is not that easy, I came across a significant report this week that might just make the industry think again about how far it wants to go: the Law Commission of England and Wales and the Scottish Law Commission published a 315-page joint report, which recommends introducing a new Automated Vehicles Act, to regulate vehicles that can drive themselves. It recommends drawing a clear distinction between features which just assist drivers, such as adaptive cruise control, and those that are self-driving.

Under the proposals, when a car is authorized by a regulatory agency as having “self-driving features” and those features are in-use, the person in the driving seat would no longer be responsible for how the car drives. Instead, the company or body that obtained the authorization (an ‘Authorised Self-Driving Entity’) would face regulatory sanctions if anything goes wrong.

Well, there you go, if this recommendation is followed, then car manufacturers and potentially developers of systems and software could become liable for accidents caused by self-driving features.

On that note, we take you back to this week’s news, products and articles. On the web site, you’ll see features on how vapor phase soldering can aid prototyping, and secure edge device provisioning. In news, we have PICMG’s new COM-HPC carrier board design guide, and quantum cryptography IP developer PQShield’s latest funding. Products include Synaptics’ new reference design for wireless docking over Wi-Fi 6/6E, and Alfamation’s test support for APIX3 automotive infotainment technology.

In news below, we cover the latest semiconductor industry forecast from Future Horizons, plus news from CEVA, TT Electronics, iSYSTEM, Samsung Electronics, Ceremorphic, Atmosic, Appletec, DSP Concepts, Owl Autonomous Imaging, Wayve, and EMnify.

News & Products

The semiconductor industry is heading for its next downturn, Malcolm Penn, CEO of Future Horizons, warned in his annual forecast for the sector, delivered online during IFS2022. “It is simply a question of when: the fourth quarter of this year or the first quarter of 2023,” predicted Penn. There are no soft landings,” he added. Penn posited that the global semiconductor market would grow roughly by 10% this year. That would represent better performance than the 8.8% recently forecast by World Semiconductor Trade Statistics (WSTS).

CEVA Fortrix SecureD2D IP
(Image: CEVA)

CEVA, Inc., through its wholly owned subsidiary, Intrinsix, has introduced the Fortrix SecureD2D IP solution for secure data exchange between different chiplets within a heterogenous system-on-chip (HSoC), offering secure authentication and firmware boot/code load between chiplets. The IP was selected and deployed as part of the Department of Defense State-of-the-Art Heterogeneous Integrated Packaging (SHIP) program and has already been adopted by Lockheed Martin and a ‘world-leading’ semiconductor company. The IP consists of a controller communicating over a secure fabric to hardware-based crypto accelerators which perform rapid encryption and decryption to enable cryptographic functions such as ECDSA, SHA2, AES, and others. A low-level firmware API and a customizable high-level application are also part of the IP package to allow rapid integration into secure chiplets.

TT Electronics_FlexSense_Transmissive Encoder Illustration (1)
(Image: TT Electronics)

TT Electronics has launched FlexSense series optical encoder sensors featuring on-the-fly programmability, comprising FS210 transmissive incremental encoder and FS310 reflective incremental encoder sensors. These sensors integrate several features including auto-alignment capability, a closed-loop LED driver, on-chip diagnostics, and a state-of-the-art 8x interpolator. In addition, the fully programmable array system-on-chip (SOC) enables adaptability to a multiple range of code-disk diameters and pulses per revolution, as well as several LEDs on the market, including TT’s OP207CL encoder LED. If the encoder code-disk is later altered, FlexSense can be reprogrammed to meet changing requirements.

(Image: iSYSTEM)

iSYSTEM has added the NXP S32K344 emulation adapter to its portfolio to help accelerate software tests. The adapter is populated with an NXP S32K344 “umbrella” device offering extra pins to enable trace for all S32K3 derivatives. The adapter can also be used for standalone operations as an evaluation board. With this emulation adapter and with the aid of iSYSTEM’s BlueBox and winIDEA debugging tool, developers can now get the benefits of tracing: they can non-intrusively analyze the timing behavior of an application, perform code coverage metrics and measure the CPU load or localize flaws and errors in the code for further debugging. These immediate insights into the software help improve code quality and shorten development times.

Samsung Fingerprint_IC
(Image: Samsung Electronics)

Samsung Electronics has launched what it said is the industry’s first all-in-one security chip solution that reads biometric information through a fingerprint sensor, stores and authenticates encrypted data with a tamper-proof SE, and analyzes and processes data with a secure processor. With the three key functions integrated in a single chip, its new S3B512C can help card manufacturers reduce the number of chips required and optimize card design processes for biometric payment cards. The solution’s encrypted fingerprint data is stored in a SE that has received globally accredited certifications such as EMVCo and CC EAL 6+.

Ceremorphic, an AI chip startup emerging from stealth mode this week, is readying a heterogeneous AI processor aimed at model training in data centers, automotive, high-performance computing, robotics and other emerging applications. Venkat Mattela, founding CEO of Redpine Signals, launched the AI processor startup in April 2020 after selling Redpine Signals’ wireless business to Silicon Labs for $308 million. Mattela and an 18-member team along with key patents were spunout to become Ceremorphic. Its IP portfolio includes a multi-thread processor macro-architecture called ThreadArch dating back to 2003 (licensed to Silicon Labs, but not sold).

Partner content: IoT device manufacturer survey (deadline 1 February, 2022)

Timesys is conducting an industry-wide, global survey to better understand where IoT device manufacturers are on their cybersecurity journey. The survey, which should take less than 5 minutes to complete, includes driving factors for security, security processes being implemented, specific technologies being used, and security challenges being faced. Timesys emphasizes individual answers will remain completely anonymous and strictly confidential. It is open to all individuals, regardless of location. All participants will get early access to the survey results, and 5 participants will randomly be selected to receive a USD$50 Visa gift card. Survey closes 1 February, 2022. Take the survey.

Atmosic has introduced the ATM33 series Bluetooth 5.3 family of SoCs to its portfolio of advanced energy harvesting and ultra-low power radio technology products, featuring a 0.7mA receiver and a transmitter that can operate efficiently with output power from 0dBm to 10dBm. The ATM33 integrates a power-optimized Arm Cortex M33F operating up to 64MHz and supporting Arm TrustZone. The new series has improved energy harvesting efficiency, a wider harvesting input range, a lower cold start voltage, lithium-ion battery support, and low-power voice processing. In addition, the Atmosic sensor hub supports CPU-less operations, allowing the SoCs to perform functions such as access memory and transmit beacons even while in deep sleep.

Israeli company Appletec is producing a range of compact camera modules (CCMs) exclusively for Arrow Electronics, intended for developers who want to add embedded vision capability to their products. The Arrow CCMs use CMOS image sensors from onsemi and can utilize the image signal processor (ISP) on the customer’s main board where, to minimize cost, they do not include their own ISP. Two fixed focus CCMs are available initially, offering 0.3 or 4MP sensors and fields of view ranging from 49 to 84 degrees. Frame rates range from 36.7fps (at 720p) and 75fps (at VGA) up to 360fps. A CCM with auto-focus and 13MP sensor will be added later offering 30fps and 64 degree FOV. A fourth CCM, based on a 1.26MP, 30fps sensor with integrated image flow processor will enable vision functionality even for MCU platforms without additional hardware components.


In addition to launching its energy harvesting chips (see above) Atmosic has raised $72 million in new funding let by early investor Sutter Hill Ventures, taking the total raised since being founded in 2016 to over $100 million. The company said business has grown significantly during the past year with a 10x increase in shipments, and a 500% expansion of its customer base, including the recent partnership announcement with Universal Electronics.

Yuchun Lee DSP Concepts
Yuchun Lee

DSP Concepts has closed a $28 million series C funding round, taking its total funding to date to $55 million. The investment was led by entrepreneur investor, Mr. Yuchun Lee, who joins the company as executive chairman, with participation from new and previous investors including Sony Innovation Fund, Subaru, and Porsche Digital. The company said it is enabling the “audio of things” and is the creator of the Audio Weaver platform that powers sound and voice features in many of the world’s leading consumer and automotive brands.

Owl Autonomous Imaging, a Silicon Catalyst portfolio company and developer of patented monocular 3D thermal imaging and ranging solutions for automotive active safety systems, has announced $15 million series A funding led by State Farm Ventures. Current ADAS sensor suites typically comprise mutually dependent visible-light cameras and radar, but when one of these sensors becomes ineffective, so too does the entire sensor suite. Owl has created a new modality known as monocular 3D thermal ranging to dramatically improve pedestrian safety, based on specialty HD thermal imaging and computer vision algorithms which exploit angular, temporal and intensity data to produce ultra-dense point clouds and highly refined object classification.

Wayve, a London, U.K.-based startup applying AI to autonomous vehicles and mobile robotics, announced it has been backed by prominent investors such as Microsoft in a $200 million series B funding round. Founded in 2017 from research in computer vision and deep learning at the University of Cambridge, Wayve claims it has developed a deep learning technology that permits autonomous driving in dense urban environments. Instead of relying on the traditional AV stack, HD maps and hand-coded rules, the startup said it has developed a data-driven learned driver that can scale, adapt and generalize its driving intelligence to new places.

Berlin, Germany-based EMnify, a supplier of IoT roaming services on both LTE-M and NB-IoT platforms, has raised a further €50 million to increase its footprint in international markets, notably in the U.S., where it already has 20% of its customer base and which generates 30% of revenues. The company describes itself as a provider of ‘cloud building blocks’ for cellular in the IoT stack such that its SIM solutions and API can connect and secure small and large IoT deployments to its application back end.

Related Content:

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.