ROS 2 Hardware Acceleration Working Group

ROS 2 Hardware Acceleration Working Group - meeting #11

Source: ROS Discourse, Septemeber 29, 2022 [1]
We discussed Robotic Processing Unit updates and release, hardware acceleration in ROS 2 and Gazebo community survey updates and release schedule, REP-2008 updates, ROSCon and IROS hardware acceleration talks and a new WG subproject (roscore-v) that will design an open source software and hardware ROS 2 microcontroller unit (MCU) powered by RISC-V.

News

About Tesla's Optimus robot brain, a robotics hardware and software computer architecture perspective

Source: Acceleration Robotics, October 5, 2022 [2]
Optimus robot isn't anything new but the team behind it believes that "their brain" is the key to conquer humanoids. This article takes a closer look at the public information available on their hardware and software computer architecture, and the evolution behind it. When compared to Optimus' latest "Bot Brain", existing open source software and hardware alternatives present a much more appealing starting point for building humanoid robots.

Chipmakers cut output, investment – but government bucks never go out of style

Source: The Register, Septemeber 30, 2022 [3]
Micron and Kioxia announce production cuts as demand for semiconductors continues to fall. Micron's capex investment in production in 2023 will be about $8 billion, down more than 30 percent from this year. Research outfit TrendForce predicted a 15 to 20 percent cut in NAND flash pricing during this quarter.

Advances in computer vision propel transportation autonomy

Source: Forbes, October 4, 2022 [4]
Computer vision as an academic discipline took off in the 1960s, primarily at universities engaged in the emerging field of artificial intelligence (AI) and machine learning. Recent advances in deep learning and artificial intelligence have further accelerated the application of computer vision to provide real-time, low latency perception and cognition.

Meta launches AI software tools to ease switching between Nvidia, AMD chips

Source: Reuters, October 3, 2022 [5]
Facebook parent Meta Platforms Inc (META.O) has launched a new set of free software tools for artificial intelligence applications. Software could make it easier for developers to switch back and forth between different underlying chips. It is based on an open-source machine learning framework called PyTorch.

Chipmakers still shoveling cash into new fabs as demand slows

Source: The Register, Septemeber 28, 2022 [6]
Investment in semiconductor fab equipment to hit $99 billion by the end of 2022, SEMI projects. SEMI: Industry has already increased capacity this year and will do so again in 2023. ASML reported a record $8.7 billion in orders for the second quarter of this year.

This new computer chip is ideal for AI

Source: Tech Xplore, October 3, 2022 [7]
Artificial intelligence presents a major challenge to conventional computing architecture. Researchers have zeroed in on hardware redesign to deliver required improvements in speed, agility and energy usage. A team of researchers from the University of Pennsylvania's School of Engineering and Applied Science has introduced a computing architecture ideal for AI.

Simplifying CUDA upgrades for NVIDIA Jetson users

Source: NVIDIA Technical Blog, October 4, 2022 [8]
NVIDIA JetPack provides a full development environment for hardware-accelerated AI-at-the-edge on Jetson platforms. NVIDIA JetPack is released on a rolling cadence with a single version of CUDA. Users on JetPack 5.0 and later can upgrade to the latest CUDA release without updating the NVIDIA jetPack version or Jetson Linux BSP (Board Support Package).

Detecting objects in point clouds using ROS 2 and TAO-pointpillars

Source: NVIDIA Technical Blog, October 4, 2022 [9]
This post presents a ROS 2 node for detecting objects in point clouds using a pretrained model from NVIDIA TAO Toolkit based on PointPillars. The node takes point clouds as input from real or simulated lidar scans and performs TensorRT-optimized inference to detect objects.

Papers

A survey on FPGA cybersecurity design strategies

Source: ACM Transactions on Reconfigurable Technology and Systems, September 15, 2022 [10]
Field programmable gate array (FPGA) devices present unique challenges to cybersecurity. A single software flaw could open up a breach that might allow an attacker to extract the FPGA's configuration data. This paper aims to provide the groundwork to build a solid threat-based cybersecurity design strategy.

An FPGA framework for interferometric vision-based navigation (iVisNav)

Source: NASA/ADS, August 7, 2022 [11]
iVisNav employs laser emitting structured beacons and precisely characterizes six degrees of freedom relative motion rates. iVisNav's embedded package must efficiently process high frequency dynamics for robust sensing and estimation. The resulting system is capable of interfacing with the photonics and implement the estimation algorithm in a field-programmable gate array.

Previous Hardware Acceleration in Robotics Newsletters

Past ROS 2 Hardware Acceleration Working Group meetings

  1. Hardware Acceleration WG, meeting #11. https://discourse.ros.org/t/hardware-acceleration-wg-meeting-11/27300/4 ↩︎

  2. Mayoral, V. (2022, October 5). Tesla's Optimus brain hardware and software architecture. Hardware Acceleration in Robotics. https://news.accelerationrobotics.com/tesla-optimus-robot-brain-computer-architecture-hardware-software/ ↩︎

  3. Robinson, D. (2022, September 30). Chipmakers cut output, investment as demand. The Register: Enterprise Technology News and Analysis. https://www.theregister.com/2022/09/30/chipmakers_cut_output/?utm_source=daily&utm_medium=newsletter&utm_content=article ↩︎

  4. Rangwala, S. (2022, October 4). Advances in computer vision propel transportation autonomy. Forbes. https://www.forbes.com/sites/sabbirrangwala/2022/10/04/advances-in-computer-vision-propel-transportation-autonomy/?sh=6f0061e7147f ↩︎

  5. Lee, J. L., & Nellis, S. (2022, October 3). Meta launches AI software tools to ease switching between Nvidia, AMD chips. Reuters. https://www.reuters.com/technology/meta-launches-ai-software-tools-help-speed-up-work-blog-2022-10-03/ ↩︎

  6. Robinson, D. (2022, September 28). Chipmakers still investing in new fab kit as demand slows. The Register: Enterprise Technology News and Analysis. https://www.theregister.com/2022/09/28/chip_fabrication_investment_semi/?utm_source=daily&utm_medium=newsletter&utm_content=article ↩︎

  7. Fischler, D. (2022, October 3). This new computer chip is ideal for AI. Tech Xplore - Technology and Engineering news. https://techxplore.com/news/2022-10-chip-ideal-ai.html ↩︎

  8. Mukund, R., Sheshadri, S. H., Kini, Y., & Oh, F. (2022, October 4). Simplifying CUDA upgrades for NVIDIA Jetson users. NVIDIA Technical Blog. https://developer.nvidia.com/blog/simplifying-cuda-upgrades-for-nvidia-jetson-users/ ↩︎

  9. Bhide, A. (2022, October 4). Detecting objects in point clouds using ROS 2 and TAO-pointpillars. NVIDIA Technical Blog. https://developer.nvidia.com/blog/detecting-objects-in-point-clouds-using-ros-2-and-tao-pointpillars/ ↩︎

  10. Proulx, A., Chouniard, J. Y., Fortier, P., & Miled, A. (2022, September 15). ACM Transactions on Reconfigurable Technology and Systems. https://dl.acm.org/doi/abs/10.1145/3561515 ↩︎

  11. Ramchander, R. B., Kookjin, S., & Manoranjan, M. (2022, August 7). An FPGA framework for interferometric vision-based navigation (iVisNav). NASA/ADS. https://ui.adsabs.harvard.edu/abs/2022arXiv220803605R/abstract ↩︎