- ROS 2 Hardware Acceleration Working Group - meeting #12
- Why are some companies still short semiconductor chips when others are swimming in them?
- TSMC reportedly building 1nm chip fab in northern Taiwan
- AMD has launched two Xilinx FPGAs optimised for cost in embedded applications
- AMD Genoa detailed – Architecture makes Xeon look like a dinosaur
- New design center in Spain for next Gen semiconductors
- Pushing acceleration to the edge
- China conflicts prompt chip manufacturing battles in Europe
- Chipmakers, once in high demand, confront sudden challenges
- NASA’s EMIT mission taps FPGA processing to measure surface minerals
- US chip industry worried it may lose out to rivals over China ban
- Like the iPhone did, robots will open a new market
- How microservices enhance agility in embedded systems development
- Improving the autonomous navigation of mobile robots in crowded spaces using people as sensors
ROS 2 Hardware Acceleration Working Group
Source: ROS Discourse, November 3, 2022 
We discussed updates from ROSCon and IROS on hardware acceleration, the Robotic Processing Unit, updates and block diagram, standardization efforts in hardware acceleration updates, REP-2008 and REP-2014, 2022 Hardware Acceleration Report in Robotics 9, RobotPerf, a consortium of robotics leaders from academia, research labs, and industry whose mission is to build fair and useful robotics benchmarks that provide unbiased evaluations of robotics computing performance for hardware, software, and services—all conducted under prescribed conditions.
Source: Forbes, November 3, 2022 
Taiwan Semiconductor Manufacturing Company (TSMC) reportedly asked its staff to take vacations. Demand from smartphone makers has dropped sharply and is forecasted to get worse going into 2023. While some chip manufacturers appear to be swimming in inventories, some chips continue to be in short supply.
Source: Taiwan News, October 31, 2022 
Taiwan Semiconductor Manufacturing Company (TSMC) is reportedly planning a 1 nanometer fab in Taoyuan. The world's largest contract chipmaker did not deny or confirm the report. It added that it will continue to invest in advanced chip manufacturing in Taiwan.
Source: EENewsEurope, November 7, 2022 
AMD has unveiled two new family of FPGA chips for power-sensitive applications such as medical imaging, machine vision and automotive radar/lidar. New devices are built on the same 16nm process as previous devices but use a different mix of resources to reduce power consumption.
Source: SemiAnalysis, November 10, 2022 
This report will focus mostly on technology, architecture, performance, and SKUs. For those more concerned about volumes, ASPs, and revenue, check out our deep dive report here. Long story short, the server industry has been slow to switch to new suppliers due to long validation cycle times.
Source: Cisco Newsroom, November 10, 2022 
Cisco announced plans to launch a center for the design of next generation semiconductor devices in Spain. Announcement was made by Chuck Robbins, CEO of Cisco and H.E. Pedro Sánchez, Prime Minister of Spain. The center aims to contribute Cisco's knowledge and experience to help grow the European chips ecosystem.
Source: Semiwiki, November 4, 2022 
As more AI applications turn to edge computing, the need for more computational performance at the edge continues to increase. Custom hardware that is optimized for performance and high-level synthesis can radically reduce design cost. The total IC market revenue for AI will grow from $529 billion today to $1144 billion by 2030.
Source: The Register, November 8, 2022 
German chip manufacturer Elmos Semiconductor says it was informed by the government that the sale of its wafer fab to Chinese-owned Silex Microsystems "will most likely be prohibited" Germany has been making moves to protect Western semiconductor technology and supply chains from Chinese interests.
Source: The Economic Times, November 3, 2022 
Makers of computer chips seemed on top of the world a few months ago. Chips act as brains of computers and are needed in just about every device with an on-off switch. Democrats and Republicans agreed in July on a $52 billion subsidy package to boost U.S. chip production.
Source: allaboutcircuits, November 9, 2022 
Representing a joint effort to accurately characterize the minerals present on the Earth’s surface, NASA’s JPL leverages optical and FPGA-based advances to help in modeling the Earth’s climate.
Source: The Register, November 4, 2022 
US companies told to tough it out as they face a ban on selling to customers in China. Washington is trying to persuade its allies to join it in implementing export controls on any advanced chip technologies to China, as well as any equipment that could be used to manufacture such products.
Source: The Investor, November 7, 2022 
The Korea Institute for Robot Industry Advancement (KIRIA) President Shon Woong-hee says the next revolutionary technology will be led by robots. Applying robots to a variety of fields will open up a new market for the robot industry, he said.
Source: embedded, November 7, 2022 
Despite the increased use of embedded devices, development techniques remain in the past. Hardware devices need to have more ubiquitous connectivity and code should be able to be reused. This article will make a case for adopting modern software development and applying that to edge and embedded development – based on an open-source approach.
Source: Tech Xplore, November 2, 2022 
A team of researchers from University of Illinois at Urbana-Champaign and Stanford University led by Prof. Katie Driggs-Campbell, have recently developed a new deep reinforcement learning-based method that could improve the ability of mobile robots to safely navigate crowded spaces. Their method, introduced in a paper pre-published on arXiv, is based on the idea of using people in the robot's surroundings as indicators of potential obstacles.
Source: Google Scholar, September 27, 2022 
In the new Internet of Things (IoT) era, embedded Field-Programmable Gate Array (FPGA) is enabling the deployment of custom-tailored embedded IoT solutions. This article presents the FAC-V coprocessor, which is an FPGA-based solution for an RISC-V processor.
Source: arXiv, September 22, 2022 
Transformer is a deep learning language model used for natural language processing (NLG) services in datacenters. DFX is a multi-FPGA acceleration appliance that executes GPT-2 model inference end-to-end with low latency and high throughput. It achieves 5.58× speedup and 3.99× energy efficiency over four NVIDIA V100 GPUs on the NLP model.
Hardware Acceleration in Robotics Newsletters
- Hardware Acceleration in Robotics #32 - Acceleration Robotics expands to India and takes over TechnoYantra to grow in Asia, An open architecture for hardware acceleration in robotics and ROS 2 - ROSCon 2022 and more
- Hardware Acceleration in Robotics #31 - ROS 2 HAWG #12 meeting, Google announces new open-source OS for RISC-V chips, How digital twin technology can bridge America’s chip manufacturing gap and more
- Hardware Acceleration in Robotics #30 - The semiconductor industry is ‘near its limit,’ says Nvidia CEO, Intel and Google collaborate on computing ASIC for data centers and more
- Hardware Acceleration in Robotics #29 - 2022 Hardware acceleration report in robotics, ROS 2 HAWG #12, REP-2008 - ROS 2 hardware acceleration architecture and conventions and more
- Hardware Acceleration in Robotics #28 - ROS2 HAWG #11, About Tesla's Optimus robot brain, a robotics hardware and software computer architecture perspective and more
- Hardware Acceleration in Robotics #27 - Acceleration Robotics announce and open sources ROBOTCORE®, Acceleration Robotics, AMD partner to design robotic compute architectures with ROS and more
- Hardware Acceleration in Robotics #26 - ROS2 HAWG #11 meeting, Acceleration Robotics collaborating with AMD to design next-generation robotic compute architectures with ROS and more
- Hardware Acceleration in Robotics #25 - Hard vs. Soft: Software may be eating the world, but hardware monetizes better, ROS2 HAWG #11 and more
- Hardware Acceleration in Robotics #24 - ROS Robotics Companies, SiFive RISC-V CPU cores to power NASA's next spaceflight computer and more
ROS 2 Hardware Acceleration Working Group meetings
- Hardware Acceleration WG, meeting #12
- Hardware Acceleration WG, meeting #11
- Hardware Acceleration WG, meeting #10
- Hardware Acceleration WG, meeting #9
- Hardware Acceleration WG, meeting #8
- Hardware Acceleration WG, meeting #7
- Hardware Acceleration WG, meeting #6
- Hardware Acceleration WG, meeting #5
- Hardware Acceleration WG, meeting #4
- Hardware Acceleration WG, meeting #3
- Hardware Acceleration WG, meeting #2
- Hardware Acceleration WG, meeting #1
Hardware Acceleration WG, meeting #12. https://discourse.ros.org/t/hardware-acceleration-wg-meeting-12/27802/3 ↩︎
Shih, W. (2022, November 3). Why are some companies still short semiconductor chips when others are swimming in them? Forbes. https://www.forbes.com/sites/willyshih/2022/11/03/why-are-some-companies-still-short-semiconductor-chips-when-others-are-swimming-in-them/amp/ ↩︎
Flaherty, N. (2022, November 7). AMD launches cost optimised Xilinx FPGAs. EENewsEurope. https://www.eenewseurope.com/en/amd-launches-cost-optimised-xilinx-fpgas/ ↩︎
Patel, D. (2022, November 10). AMD Genoa detailed – Architecture makes Xeon look like a dinosaur. SemiAnalysis | Dylan Patel | Substack. https://www.semianalysis.com/p/amd-genoa-detailed-architecture-makes ↩︎
New design center in Spain for next Gen semiconductors. (2022, November 10). Cisco Newsroom. https://newsroom.cisco.com/c/r/newsroom/en/us/a/y2022/m11/cisco-to-launch-new-design-center-in-spain-for-next-generation-semiconductor-devices.html ↩︎
Bursky, D. (2022, November 4). Pushing acceleration to the edge. Semiwiki. Retrieved November 9, 2022, from https://semiwiki.com/eda/siemens-eda/320270-pushing-acceleration-to-the-edge/ ↩︎
Martin, D. (2022, November 8). China conflicts prompt chip manufacturing battles in Europe. The Register: Enterprise Technology News and Analysis. https://www.theregister.com/2022/11/08/china_at_heart_of_conflict/?utm_source=daily&utm_medium=newsletter&utm_content=article ↩︎
Clark, D. (2022, November 3). Chipmakers, once in high demand, confront sudden challenges. The Economic Times. https://economictimes.indiatimes.com/small-biz/trade/exports/insights/chipmakers-once-in-high-demand-confront-sudden-challenges/articleshow/95267433.cms ↩︎
Carman, A. (2022, November 9). NASA’s EMIT mission taps FPGA processing to measure surface minerals. allaboutcircuits. https://www.allaboutcircuits.com/news/nasas-emit-mission-taps-fpga-processing-to-measure-surface-minerals/ ↩︎
Robinson, D. (2022, November 4). Providers of chipmaking kit face ban while others do not. The Register: Enterprise Technology News and Analysis. https://www.theregister.com/2022/11/04/us_chip_industry_china_ban/?utm_source=daily&utm_medium=newsletter&utm_content=article ↩︎
'Like the iPhone did, robots will open a new market'. (2022, November 7). THE INVESTOR. https://m.theinvestor.co.kr/view.php?ud=20221107000152 ↩︎
Rabault, N. (2022, November 7). How microservices enhance agility in embedded systems development. embedded. https://www.embedded.com/how-microservices-enhance-agility-in-embedded-systems-development/ ↩︎
Fadelli, I. (2022, November 2). Improving the autonomous navigation of mobile robots in crowded spaces using people as sensors. Tech Xplore - Technology and Engineering news. https://techxplore.com/news/2022-11-autonomous-mobile-robots-crowded-spaces.html ↩︎
Gomes, T., Sousa, P., Silva, M., Ekpanyapong, M., & Pinto, S. (2022, September 27). FAC-V: an FPGA-based AES coprocessor for RISC-V. Google Scholar. https://scholar.google.es/scholar_url?url=https://www.mdpi.com/2079-9268/12/4/50/pdf%3Fversion%3D1664283511&hl=es&sa=X&d=13701883750163396031&ei=-o43Y9n-FeSTy9YPmsyrQA&scisig=AAGBfm1Z7QgOwC_86i5BDRt8PYPy6h-ZEQ&oi=scholaralrt&hist=fHCcxNkAAAAJ:12793294858605781301:AAGBfm2rf1tEU9i1lasczMbBKPNB8lseKw&html=&pos=1&folt=kw-top ↩︎
Seongmin, H., Seungjae, M., Junsoo, K., Sungjae, L., Minsub, K., Dongsoo, L., & Younf, K. J. (September 22). DFX: a low-latency multi-FPGA appliance for accelerating transformer-based text generation. arXiv. https://arxiv.org/pdf/2209.10797.pdf ↩︎