News

The ROS 2 hardware acceleration stack and ROBOTCORE™

Source: Acceleration Robotics, June 20, 2022 [1]
The core of robots resides in their computing systems including both the hardware and the software levels. This entails computer architectures, algorithms but also frameworks for development. With robotics called to be the next major driver for computing demand and ROS 2 the standard framework for robot application development, the ability to build faster ROS 2 graphs with hardware acceleration in an open and vendor agnostic manner will be instrumental for the development of robotic systems. We at Acceleration Robotics discuss the technology enablers that empower this, group them into the ROS 2 Hardware Acceleration Stack and present ROBOTCORE™ as its implementation.

RISC-V shines at embedded world with new specs and processors

Source: allaboutcircuits, June 22, 2022 [2]
At Embedded World 2022 today, RISC-V activity heats up as RISC-V International reveals four new spec approvals and SiFive unveils a new version of its X280 processor.

AMD bests Intel in cloud CPU performance study

Source: The Register, June 16, 2022 [3]
AMD's processors have come out on top in terms of cloud CPU performance across AWS, Microsoft Azure, and Google Cloud Platform. Multi-core x86-64 microprocessors Milan and Rome beat Intel Cascade Lake and Ice Lake instances in tests of performance in the three most popular cloud providers.

Former AMD chip architect says it was wrong to can Arm project

Source: The Register, June 20, 2022 [4]
Jim Keller, who worked on Arm development at AMD, reckons the chipmaker was wrong to halt the project after he left the company in 2016. Speaking at an event in April, Keller discusses how much of the architecture is very similar for Arm and X86.

Qualcomm AI stack aims to achieve parity with NVIDIA Intel and others

Source: ServeTheHome, June 22, 2022 [5]
Today Qualcomm is announcing its AI unification offering. The Qualcomm AI Stack is designed to allow developers to build software that works across the company’s silicon offerings. This is a feature many other companies already have been working towards, but Qualcomm has a unique set of silicon solutions.

Codasip adds Veridify secure boot to RISC-V processors

Source: RealWire, June 22, 2022 [6]
Codasip's RISC-V processors now pre-integrated with Veridify's secure boot function. Integrators can now easily verify the authenticity of the firmware on their embedded systems. Codasip processors will be used in long-life embedded applications such as surveillance cameras and smart meters.

Nvidia patent helps autonomous cars detect emergency vehicles

Source: The Robot Report, June 21, 2022 [7]
Nvidia filed a patent for a system that would help self driving cars identify and react to emergency vehicles. System would use microphones to capture sounds of nearby emergency response vehicles' sirens.

A deep learning framework to estimate the pose of robotic arms and predict their movements

Source: Tech Xplore, June 22, 2022 [8]
Brazilian researchers have created a new deep learning model to estimate the pose of robotic arms and predict their movements. This model is specifically designed to enhance the safety of robots while they are collaborating or interacting with humans. It was developed by researchers at the Universidade Federal de Pernambuco in Brazil.

Imagination launches its first real-time embedded RISC-V CPU

Source: eenewseurope, June 21, 2022 [9]
UK processor designer Imagination Technologies has launched its first real time embedded RISC-V CPU core. The IMG RTXM-2200 follows the announcement of a series of RISC-V cores called the Catapult CPU family back in December. The 32bit core is highly scalable and supports up to 128KB of tightly coupled memories…

Minimizing storage usage on Jetson

Source: NVIDIA Technical Blog, June 17, 2022 [10]
NVIDIA Jetson provides flexible storage options/configurations for development. Some Jetson modules have limited eMMC flash memory storage size for large-scale product deployments. You can reclaim more than 8 GB of storage space on some of the latest NVIDIA JetPack versions.

Improving GPU utilization in Kubernetes

Source: NVIDIA Technical Blog, June 16, 2022 [11]
For scalable data center performance, NVIDIA GPUs have become a must-have. The most compute-intensive applications across diverse industries use them today. Provisioning the right-sized GPU acceleration for each workload is key to improving utilization and reducing operational costs of deployment, whether on-premises or in the cloud.

Papers

An FPGA and ASIC implementation of Cubing architecture

Source: SpringerLink, May 5, 2022 [12]
In this paper proposes a technique for the implementation of cubing. By using the proposed method, the complexity of the multiplication of numbers for cubes is reduced. The proposed architecture is synthesized and simulated using Vivado design suit 2018.3 and implemented on a Kintex-7 FPGA board.

An experience report on challenges in learning the Robot Operating System

Source: RoSE Workshop series, May 9, 2022 [13]
In this work, discuss the challenges and experience of learning and using ROS from the perspective of three novice users with little to no prior experience in robotics. Report on the experiences in learning ROS through a popular commercial training course provided by The Construct Sim. Through our analysis, we identify several common misunderstandings, mistakes, and bugs, and we outline possible improvements to ROS to overcome these challenges.

Previous Hardware Acceleration in Robotics Newsletters

Past ROS 2 Hardware Acceleration Working Group meetings

  1. Mayoral, V. (2022, June 20). The ROS 2 hardware acceleration stack and ROBOTCORE™. Hardware Acceleration in Robotics. https://news.accelerationrobotics.com/ros2-hardware-acceleration-stack/ ↩︎

  2. Child, J. (2022, June 22). RISC-V Shines at Embedded World With New Specs and Processors. allaboutcircuits. https://www.allaboutcircuits.com/news/risc-v-shines-at-embedded-world-with-new-specs-and-processors/ ↩︎

  3. Clark, L. (2022, June 16). AMD beats Intel in cloud CPU performance research. The Register: Enterprise Technology News and Analysis. https://www.theregister.com/2022/06/16/amd_beats_intel_in_cloud/ ↩︎

  4. Robinson, D. (2022, June 20). Former AMD chip architect says it was wrong to can arm CPU. The Register: Enterprise Technology News and Analysis. https://www.theregister.com/2022/06/20/jim_keller_arm_cpu/ ↩︎

  5. Robinson, C. (2022, June 22). Qualcomm AI stack aims to achieve parity with NVIDIA Intel and others. ServeTheHome. https://www.servethehome.com/qualcomm-ai-stack-aims-to-achieve-parity-with-nvidia-intel-amd-and-others/ ↩︎

  6. Codasip adds Veridify secure boot to RISC-V processors. (2022, June 22). RealWire. https://www.realwire.com/releases/Codasip-adds-Veridify-secure-boot-to-RISC-V-processors ↩︎

  7. Wessling, B. (2022, June 21). Nvidia patent helps autonomous cars detect emergency vehicles. The Robot Report. https://www.therobotreport.com/nvidia-patent-helps-autonomous-cars-detect-emergency-vehicles/ ↩︎

  8. Fadelli, I. (2022, June 22). A deep learning framework to estimate the pose of robotic arms and predict their movements. Tech Xplore - Technology and Engineering news. https://techxplore.com/news/2022-06-deep-framework-pose-robotic-arms.html ↩︎

  9. Flaherty, N. (2022, June 21). Imagination launches its first real-time embedded RISC-V CPU. eenewseurope. https://www.eenewseurope.com/en/imagination-launches-its-first-real-time-embedded-risc-v-cpu/ ↩︎

  10. Yato, C. (2022, June 17). Minimizing storage usage on Jetson. NVIDIA Technical Blog. https://developer.nvidia.com/blog/minimizing-storage-usage-on-jetson/ ↩︎

  11. Klues, K., Perelygin, K., & Ramarao, P. (2022, June 16). Improving GPU utilization in Kubernetes. NVIDIA Technical Blog. https://developer.nvidia.com/blog/improving-gpu-utilization-in-kubernetes/ ↩︎

  12. Naresh Kumar Reddy, B., Seetharamulu, B., Krishna, G. S., & Vani, B. V. (2022, May 5). An FPGA and ASIC implementation of cubing architecture. SpringerLink. https://link.springer.com/article/10.1007/s11277-022-09715-w ↩︎

  13. Canelas, P., Tavares, M., Cordeiro, R., Fonseca, A., & Timperley, C. S. (2022, May 9). An Experience Report on Challenges in Learning the Robot Operating System. RoSE Workshop series . https://rose-workshops.github.io/files/rose2022/papers/RoSE22_paper_11.pdf ↩︎