RobotCore: An Open Architecture for Hardware Acceleration in ROS 2

Source: Acceleration robotics, May 8, 2022 [1]
Hardware acceleration can revolutionize robotics,enabling new applications by speeding up robot response times while remaining power-efficient. However, the diversity of acceleration options makes it difficult for roboticists to easily deploy accelerated systems without expertise in each specific hardware platform. In this work, we address this challenge with
RobotCore, an architecture to integrate hardware acceleration in the widely-used ROS 2 robotics software framework. This architecture is target-agnostic (supports edge, workstation, data center, or cloud targets) and accelerator-agnostic (supports both FPGAs and GPUs). It builds on top of the common ROS 2 build system and tools and is easily portable across different research and commercial solutions through a new firmware layer. We also leverage the Linux Tracing Toolkit next generation (LTTng) for low-overhead real-time tracing and benchmarking. To demonstrate the acceleration enabled by this architecture, we use it to deploy a ROS 2 perception computational graph on a CPU and FPGA. We employ our integrated tracing and benchmarking to analyze bottlenecks, uncovering insights that guide us to improve FPGA communication efficiency. In particular, we design an intra-FPGA ROS 2 node communication queue to enable faster data flows, and use it in conjunction with FPGA-accelerated nodes to achieve a 24.42% speedup over a CPU.

RISC-V CEO seeks 'world domination' by winning over the likes of Intel

Source: The Register, May 7, 2022 [2]
The CEO of RISC-V's governing body says she wants to nothing less than "world domination" for the rising open-source processor technology, but to do that, the nonprofit needs buy-in from a variety of organizations, even those steeped in dominant, proprietary architectures, such as x86 giant Intel.

Experiencing NVIDIA Jetson AGX Orin: The Fascinating Future Of AI-Empowered Robotics

Source: Forbes, May 4, 2022 [3]
NVIDIA has been developing a myriad of solutions for with its Jetson portfolio for years now. Robotics applications rely on machine vision to navigate terrain and avoid collisions. Many robots of the future will need to infer and ingest important information from the world around them, interact seamlessly with humans and make decisions autonomously.

Qualcomm unveils RB6 platform and RB5 AMR reference design

Source: The Robot Report, May 10, 2022 [4]
Qualcomm announced the availability of its Robotics RB6 Platform and the RB5 Reference Design. The two products help to bring advanced artificial intelligence (AI) and 5G to robotics, drones and intelligence machines.

Intel sticks with FPGAs and ASICs for next-gen IPUs

Source: The Register, May 10, 2022 [5]
Chip giant sticking with multi-architecture ASIC/FPGA approach through 2024.

AMD to offer CPUs with Xilinx AI engine in 2023

Source: The Register, May 4, 2022 [6]
AMD plans to introduce processors next year that integrate an AI engine from the company's recently acquired Xilinx FPGA business unit, which helped the chip designer deliver high sales growth in the first quarter along with the company's traditional PC and server businesses.

A Journey from Microcontrollers to Microprocessors with AUTOSAR & ROS

Source: Apex.AI, May 11, 2022 [7]
Over the last few decades, automotive software has become increasingly centralized. Instead of having a separate microcontroller with software for each vehicle function, modern vehicles are moving to powerful microprocessors where a single chip runs the software functions previously handled by many discrete microcontrollers.

SLAMcore closes Series A to develop spatial understanding for robots

Source: The Robot Report, May 3, 2022 [8]
SLAMcore uses deep learning to help robots and drones identify physical space, objects and people around them. The company's algorithms allow robots, including everything from domestic vacuum cleaners to industrial vehicles, to precisely map the physical environment around them through a map-making software platform.

Unlocking the potential of robotics in nuclear

Source: Nuclear Engineering International, May 5, 2022 [9]
Nuclear sites are amongst the most challenging environments for robots. Technology used on nuclear sites must be at a high readiness and robustness level. This requirement often prohibits experiments in adopting emerging technology. How can we overcome these barriers to unlock the potential of robotics for nuclear?

Intel acquires graphics tech biz founded by ex-AMD, Qualcomm engineers

Source: The Register, May 3, 2022 [10]
Intel has acquired a graphics technology firm founded by ex-Qualcomm mobile GPU engineers whose previous company, Bitboys, was once thought of as a front-runner of desktop graphics.

Qualcomm Advances Development of Smarter and Safer Autonomous Robots for Logistics, Industry 4.0, and Urban Aerial Mobility with Next-Generation 5G and AI Robotics Solutions

Source: Qualcomm, May 10, 2022 [11]
As 5G advances beyond the smartphone, premium AI-enabled robotics, drones and intelligent machine solutions will empower more productive, intelligent, and advanced robots. These include autonomous mobile robots (AMRs), delivery robots, highly automated manufacturing robots and urban air mobility (UAM aircrafts).

NVIDIA Releases Open-Source GPU Kernel Modules

Source: NVIDIA Technical Blog, May 11, 2022 [12]
NVIDIA is now publishing Linux GPU kernel modules as open source with dual GPL/MIT license, starting with the R515 driver release. You can find the source code for these kernel modules in the NVIDIA Open GPU Kernel Modules repo on GitHub.

Edit: A ROS community member highlighted that this open source effort from NVIDIA might be mostly about marketing and that there might be no good faith behind it. See

San Francisco Police Are Using Driverless Cars as Mobile Surveillance Cameras

Source: VICE, May 11, 2022 [13]
“Autonomous vehicles are recording their surroundings continuously and have the potential to help with investigative leads,” an internal training document states.

Migration guide from ROS 1

Source: Open Robotics, April 12, 2022 [14]
ROS 2 Humble Hawksbill is going to be a big release, and we believe we are approaching the peak of the migration.
It really is time to start thinking about migrating your ROS 1 package to ROS2.


Robot Operating System 2: Design, architecture, and uses in the wild

Source: Science | AAAS, May 11, 2022 [15]
The next chapter of the robotics revolution is well underway with the deployment of robots for a broad range of commercial use cases. Even in a myriad of applications and environments, there exists a common vocabulary of components that robots share—the need for a modular, scalable, and reliable architecture; sensing; planning; mobility; and autonomy.

HEROv2: Full-Stack Open-Source Research Platform for Heterogeneous Computing

Source: Cornell University, January 11, 2022 [16]
Heterogeneous computers integrate general-purpose host processors with domain-specific accelerators to combine versatility
with efficiency and high performance. To realize the full potential of heterogeneous computers, however, many hardware and software design challenges have to be overcome. While architectural and system simulators can be used to analyze heterogeneous computers, they are faced with unavoidable compromises between simulation speed and performance modeling accuracy.

Previous Hardware Acceleration in Robotics Newsletters

Past ROS 2 Hardware Acceleration Working Group meetings

  1. (2022, May 8). e-Print archive. ↩︎

  2. Martin, D. (2022, May 7). RISC-V CEO: Winning over giants like Intel key to growth. The Register: Enterprise Technology News and Analysis. ↩︎

  3. Altavilla, D. (2022, May 4). Experiencing NVIDIA Jetson AGX Orin: The Fascinating Future Of AI-Empowered Robotics. Forbes. ↩︎

  4. Wessling, B. (2022, May 10). Qualcomm unveils RB6 platform and RB5 AMR reference design. The Robot Report. ↩︎

  5. Mann, T. (2022, May 10). Intel teases upcoming IPUs. The Register: Enterprise Technology News and Analysis. ↩︎

  6. Martin, D. (2022, May 4). AMD teases CPUs with Xilinx AI engine for 2023. The Register: Enterprise Technology News and Analysis. ↩︎

  7. Stark, T. (2022, May 11). A journey from Microcontrollers to microprocessors with AUTOSAR & ROS. Apex.AI. ↩︎

  8. Crowe, S. (2022, May 3). SLAMcore closes series a to develop spatial understanding for robots. The Robot Report. ↩︎

  9. Pilon, C. (2022, May 5). Unlocking the potential of robotics in nuclear. Nuclear Engineering International. ↩︎

  10. Martin, D. (2022, May 3). Intel acquires graphics tech firm Siru innovations. The Register: Enterprise Technology News and Analysis. ↩︎

  11. Qualcomm advances development of smarter and safer autonomous robots for logistics, industry 4.0, and urban aerial mobility with next-generation 5G and AI robotics solutions. (2022, May 10). Qualcomm. ↩︎

  12. Cherukuri, R., Baskaran, S., Ritger, A., Oh, F., & Swoboda, D. (2022, May 11). NVIDIA releases open-source GPU kernel modules. NVIDIA Technical Blog. ↩︎

  13. Gordon, A. (2022, May 11). San Francisco police are using driverless cars as mobile surveillance cameras. VICE - VICE is the definitive guide to enlightening information. ↩︎

  14. Migration guide from ROS 1 — ROS 2 documentation: Foxy documentation. (2022, April 11). ROS Documentation. ↩︎

  15. Macenski, S., Foote, T., Gerkey, B., Lalancette, C., & Woodall, W. (2022, May 11). Robot Operating System 2: Design, architecture, and uses in the wild. Science | AAAS. ↩︎

  16. Kurth, A., Forsberg, B., & Benini, L. (2022, January 11). HEROv2: Full-Stack Open-Source Research Platform for Heterogeneous Computing. Cornell University. ↩︎