Hardware Acceleration WG, meeting #12

[Víctor Mayoral-Vilches]

I'm happy to call for the 12th Hardware Acceleration Working Group meeting (LinkedIn event) to be held on [date=2022-11-03 time=18:00:00 timezone=Europe/Madrid] (your browser's time). In case you missed past meetings, here's the ROS 2 Hardware Acceleration Working Group list with past recordings the WG is keeping. As for other working group resources, find them below or in the WG community repo:

ROS 2 Hardware Acceleration Working Group (HAWG) resources

• Objective: Drive creation, maintenance and testing of acceleration kernels on top of open standards (C++ and OpenCL) for optimized ROS 2 and Gazebo interactions over different compute substrates (including FPGAs, GPUs and ASICs).
• Reference hardware platform: Kria K26 Adaptive SOM
• Meetings invite group: ROS 2 Hardware Acceleration WG Google Group
• Instant messaging: Matrix community (Matrix is an open network for secure, decentralized communication).
• Backlog management: https://github.com/ros-acceleration/community/projects/1
• Github organization: ros-acceleration
• Meeting minutes: HAWG minutes
• Discourse tag: wg-acceleration
• Previous meetings:
  • ROS 2 Hardware Accelerating WG, meeting #1
  • ROS 2 Hardware Acceleration WG, meeting #2
  • ROS 2 Hardware Acceleration WG, meeting #3
  • ROS 2 Hardware Acceleration WG, meeting #4
  • ROS 2 Hardware Acceleration WG, meeting #5
  • ROS 2 Hardware Acceleration WG, meeting #6
  • ROS 2 Hardware Acceleration WG, meeting #7
  • ROS 2 Hardware Acceleration WG, meeting #8
  • ROS 2 Hardware Acceleration WG, meeting #9
  • ROS 2 Hardware Acceleration WG, meeting #10
  • ROS 2 Hardware Acceleration WG, meeting #11

This meeting is happening after ROSCon and IROS 2022, so expect exciting updates and results from the interactions during these two events in Japan! In the last meeting in September we had 365 persons registered and we discussed exciting topics. In this meeting we will touch on: a) Updates from ROSCon and IROS on hardware acceleration, b) the Robotic Processing Unit, updates and block diagram, c) Standardization efforts in hardware acceleration updates, REP-2008 and REP-2014, d) 2022 Hardware Acceleration Report in Robotics, e) 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.

The session will also include a guest talk provided by Vijay Janapa Reddi. Professor Vijay Janapa Reddi is an associate professor of engineering and applied sciences at Harvard University. His research interests include machine learning, robotics, and the design of computer systems. He is an expert in low-power, high-performance system-on-chip (SoC) architectures, which he achieves by blending hardware and software via co-design in a unique manner. His long-term objective is to create open-source SoCs for robotics in order to commoditize the technology. Vijay leads a team of computer system architects at Harvard's Edge Computing Lab. He has contributed to the creation of industry benchmarks, such as MLPerf, and non-profit organizations, such as MLCommons, to allow global access to cutting-edge technology. Vijay will tell us more about his experience with benchmarking and how the RobotPerf project can present a relevant resource and an attempt to generate an industry standard for benchmarking robotic computations.

For more and a related discussion, see https://discourse.ros.org/t/hardware-acceleration-wg-meeting-12/27802