News

ROS 2 Hardware Acceleration Working Group #11 meeting

Source: Acceleration Robotics, Septemeber 13, 2022 [1]
Join the #11th meeting of the ROS 2 Hardware Acceleration Working Group to discuss progress on the Robotic Processing Unit, a hardware ROS 2 microcontroller unit (MCU) and more!

We have exciting updates coming up including a) Robotic Processing Unit updates and release, b) Hardware acceleration in ROS 2 and Gazebo community survey updates and release schedule, c) REP-2008 updates, d) ROSCon and IROS hardware acceleration talks and e) a new WG subproject (roscore-v) that will design an open source software and hardware ROS 2 microcontroller unit (MCU) powered by RISC-V.

PRESS RELEASE: Acceleration Robotics collaborating with AMD to design next-generation robotic compute architectures with ROS

Source: News Acceleration Robotics, Septemeber 19, 2022 [2]
Acceleration Robotics, who leads the ROS 2 Hardware Acceleration community efforts, will collaborate with AMD to develop new robotic capabilities with its Adaptive SoCs and SOMs

Upset by high prices, GM's Cruise develops its own chips for self-driving cars

Source: Reuters, Septemeber 14, 2022 [3]
General Motors' autonomous driving unit Cruise has developed its own chips for self-driving cars to be deployed by 2025, as they aim to bring down costs and scale up volume.

What will be the impact of ROS 2 on commercialized robotics?

Source: Formant, Septemeber 14, 2022 [4]
Open Robotics' (ROS) second release, Humble Hawksbill, comes with long term support until May of 2027. With readily available ROS 2 components, it's easier than ever to build a working prototype ready for production. This faster time-to-market will break down typical barriers to creating a robotic product. Robotics companies should consider making the switch to ROS 2. The increased capabilities of ROS 2 for safety and support have the ability to make devices supported by the system more enticing. If you're not using ROS 2, you don't need to feel the pressure to switch.

SiFive seeks to fuel next-gen designs with automotive RISC-V cores

Source: allaboutcircuits, Septemeber 16, 2022 [5]
With its new portfolio of automotive RISC-V processor cores, SiFive aims to solve challenges in the design of evolving digital cars.

iWave Systems doubles down with latest SoMs using Xilinx and Intel SoCs

Source: allaboutcircuits, Septemeber 18, 2022 [6]
The world of embedded systems keeps expanding as iWave Systems employs the FPGAs and SoCs from leading manufacturers to launch small form factor systems on modules (SoMs).

Sensor breakdown: how robot vacuums navigate

Source: The Robot Report, Septemeber 14, 2022 [7]
They often missed key areas on the floor and became stuck on thick rugs. Today's machines use sensors and motor controllers in combination with open-source software and drivers to improve navigation and cleaning.

How To compare chips

Source: Semiconductor Engineering, Septemeber 15, 2022 [8]
Traditional metrics for semiconductors are becoming much less meaningful in the most advanced designs. Number of transistors packed into a square centimeter only matters if they can be utilized. Performance per watt is irrelevant if sufficient power cannot be delivered to all of the transistors. While transistor density continues to increase, it's no longer doubling at each new process node.

NVIDIA launches Lovelace GPU, cloud services, ships H100 GPUs, new Drive Thor

Source: Forbes, Septemeber 20, 2022 [9]
NVIDIA made a slew of technology and customer announcements at the Fall GTC. Highlights included an H100 update, new NeMo LLM services, IGX for Medical Devices, Jetson Orin Nano, Isaac SIM. The highly anticipated headliner was the Lovelace GPU for gamers.

NVIDIA Jetson Orin Nano sets new standard for entry-level edge AI and robotics with 80x performance LEAP

Source: NVIDIA Newsroom, Septemeber 20, 2022 [10]
System-on-modules deliver up to 80x the performance over the prior generation. Jetson family spans six Orin-based production modules to support a full range of edge AI and robotics applications. This includes the Orin Nano, which delivers up to 40 trillion operations per second (TOPS) of AI performance.

NVIDIA Isaac Nova Orin opens new era of innovation for autonomous mobile robots

Source: NVIDIA Blog, Septemeber 20, 2022 [11]
The Nova Orin platform is designed to improve reliability and reduce development costs for building and deploying AMRs. AMRs are like self-driving cars but for unstructured environments. They don't need fixed, preprogrammed tracks and are capable of avoiding obstacles.

NVIDIA robotics software jumps to the cloud, enabling collaborative, accelerated development of robots

Source: NVIDIA Blog, Septemeber 20, 2022 [12]
The Isaac Sim robotics simulation platform is now available on the cloud. It's available now on AWS RoboMaker, a cloud-based simulation service for robotics development and testing. Mobile robotics market is expected to grow 9x worldwide from $13 billion in 2021 to over $123 billion in 2030.

NVIDIA unveils DRIVE Thor — Centralized car computer unifying cluster, infotainment, automated driving, and parking in a single, cost-saving system

Source: NVIDIA Newsroom, Septemeber 20, 2022 [13]
DRIVE Thor unifies intelligent functions for greater efficiency and lower overall system cost. Automakers can efficiently consolidate many functions on a single system-on-a-chip (SoC). DRIVE Thor is the first AV platform to incorporate an inference transformer engine, a new component of the Tensor Cores within NVIDIA GPUs. DRIVE Thor features 2,000 teraflops of FP8 precision and 8-bit floating point capability. The superchip also uses the latest NVLink®-C2C chip interconnect technology. DRIVE Thor replaces NVIDIA DRIVE Atlan and will be the follow-on to DRIVE Orin.

Papers

Shared vector register of RISC-V for the future hardware acceleration

Source: Carrv, June 19, 2022 [14]
In this study, we present a vector register sharing mechanism that directly shares vector registers inside the processor with,the acceleration circuitry. Because this mechanism can share the value of a vector register at a given time, high-speed communication is expected, particularly in SoC FPGAs. To
validate this mechanism, this study designs and implements a processor with vector registers to obtain a preliminary estimation.

Auto-ViT-Acc: an FPGA-Aware automatic acceleration framework for vision transformer with mixed-scheme quantization

Source: Simon Fraser University, August 16, 2022 [15]
Vision transformers (ViTs) are emerging with significantly improved accuracy in computer vision tasks. This work proposes an FPGA-aware automatic ViT acceleration framework based on the proposed mixedscheme quantization. Compared with the 32-bit floatingpoint baseline, our accelerator achieves around 5.6× improvement on the frame rate and 0.71% accuracy drop on ImageNet dataset.

Previous Hardware Acceleration in Robotics Newsletters

Past ROS 2 Hardware Acceleration Working Group meetings

  1. https://www.linkedin.com/events/ros2hardwareaccelerationworking6975460776610291713/comments/ ↩︎

  2. https://news.accelerationrobotics.com/acceleration-robotics-collaborating-with-amd-to-design-next-generation-robotic-compute-architectures-with-ros/ ↩︎

  3. Lee, J. L., & Jin, J. (2022, September 14). Upset by high prices, GM's cruise develops its own chips for self-driving cars. Reuters. https://www.reuters.com/business/autos-transportation/upset-by-high-prices-gms-cruise-develops-its-own-chips-self-driving-cars-2022-09-14/ ↩︎

  4. Sixt, A. (2022, September 14). What will be the impact of ROS 2 on commercialized robotics? Formant. https://formant.io/news-and-blog/2022/09/14/community/what-will-be-the-impact-of-ros-2-on-commercialized-robotics/ ↩︎

  5. Hertz, J. (2022, September 16). SiFive seeks to fuel next-gen designs with automotive RISC-V cores. allaboutcircuits. https://www.allaboutcircuits.com/news/sifive-seeks-to-fuel-next-gen-designs-with-automotive-risc-v-cores/ ↩︎

  6. Oyekunle, A. (2022, September 18). iWave Systems doubles down with latest SoMs using Xilinx and Intel SoCs. allaboutcircuits. https://www.allaboutcircuits.com/news/iwave-systems-doubles-down-with-latest-soms-incorporating-xilinx-and-intel-socs/ ↩︎

  7. Hartwell, P. (2022, September 14). Sensor breakdown: How robot vacuums navigate. The Robot Report. https://www.therobotreport.com/sensor-breakdown-how-robot-vacuums-navigate-and-clean/?spMailingID=75612&puid=2502618&E=2502618&utm_source=newsletter&utm_medium=email&utm_campaign=75612 ↩︎

  8. Sperling, E. (2022, September 15). How to compare chips. Semiconductor Engineering. https://semiengineering.com/how-to-compare-chips/?cmid=e55c5b75-8f13-4203-a2f1-3b8330fb988a ↩︎

  9. Freund, K. (2022, September 20). NVIDIA launches Lovelace GPU, cloud services, ships H100 GPUs, new Drive Thor, and .... Forbes. https://www.forbes.com/sites/karlfreund/2022/09/20/nvidia-launches-cloud-services-ships-h100-gpus-announces-new-drive-thor-and-/?sh=690c56ffb4ae ↩︎

  10. NVIDIA Jetson Orin Nano sets new standard for entry-level edge AI and robotics with 80x performance LEAP. (2022, September 20). NVIDIA Newsroom. https://nvidianews.nvidia.com/news/nvidia-jetson-orin-nano-sets-new-standard-for-entry-level-edge-ai-and-robotics-with-80x-performance-leap ↩︎

  11. Goel, A. (2022, September 20). NVIDIA Isaac Nova Orin opens new era of innovation for autonomous mobile robots. NVIDIA Blog. https://blogs.nvidia.com/blog/2022/09/20/nvidia-isaac-nova-orin/ ↩︎

  12. Andrews, G. (2022, September 20). NVIDIA robotics software jumps to the cloud, enabling collaborative, accelerated development of robots. NVIDIA Blog. https://blogs.nvidia.com/blog/2022/09/20/nvidia-isaac-sim-robotics-simulation/ ↩︎

  13. NVIDIA unveils DRIVE Thor — Centralized car computer unifying cluster, infotainment, automated driving, and parking in a single, cost-saving system. (2022, September 20). NVIDIA Newsroom. https://nvidianews.nvidia.com/news/nvidia-unveils-drive-thor-centralized-car-computer-unifying-cluster-infotainment-automated-driving-and-parking-in-a-single-cost-saving-system ↩︎

  14. Tomoaki, T., Ryosuke, H., Hidetaro, T., Takefumi, M., Yasunori, O., Jubee, T., Kiyofumi, T., & Hironori, N. (2022, June 19). Shared vector register of RISC-V for the future hardware acceleration. carrv. https://carrv.github.io/2022/papers/CARRV2022_paper_10_Tanaka.pdf ↩︎

  15. Zhengang, L., Mengshu, S., Alec, L., Haoyu, M., Geng, Y., Yanyue, X., Hao, T., Yanyu, L., Leeser, M., Zhangyang, W., Xue, L., & Zhenman, F. (2022, August 16). Auto-ViT-Acc: an FPGA-Aware automatic acceleration framework for vision transformer with mixed-scheme quantization. Simon Fraser University. https://www.sfu.ca/~zhenman/files/C25-FPL2022-Auto-ViT-Acc.pdf ↩︎