Acceleration Robotics, a robotics semiconductor startup based in Vitoria-Gasteiz, has announced ROBOTCORE®, a robot-specific processing unit for the robotics architect. This Robotic Processing Unit is specialized in accelerating the Robot Operating System (ROS) computations and features 16xCPUs, a GPU and an FPGA to empower robots with the ability to react faster, consume less energy, and deliver additional real-time capabilities. The design files of ROBOTCORE® have been disclosed and open sourced at the ROS 2 Hardware Acceleration Working Group.
Víctor Mayoral-Vilches, who is also a former Systems Architect at Xilinx (now part of AMD), says: “This is a fantastic time in robotics and with the advent of ROS standardizing as the common robotics API, we’re starting to see a growing demand for hardware acceleration in ROS, which makes its computations faster and power-efficient. ROBOTCORE® attempts to address this demand by facilitating an open blueprint for architects building robotic brains”.
Robotic Processing Unit (RPUs) are robot brains, processing units for robots that map efficiently robot behaviors (programmed as ROS computational graphs) to underlying compute resources. ROBOTCORE® features multiple CPUs, a GPU and an FPGA interconnected in a common Ethernet databus which allows combining the traditional control-driven approach used in robotics with a data-driven one. When architected appropriately, through acceleration kernels (or robot cores), ROBOTCORE® provides robotics architects with ROS 2 accelerated libraries that deliver faster computations (getting tasks done quickly once started), additional determinism (task happens in exactly the same timeframe, each time) and real time (meeting the time deadlines set for each task).
ROBOTCORE® will be showcased in Japan at ROSCon and IROS 2022 conferences and is custom designed to accelerate three robotic scenarios: autonomous mobility, industrial manipulation and healthcare applications. The robotic processing unit is powered by the technologies of AMD and NVIDIA, packing best-of-class accelerators from top silicon vendors.