RobotPerf Benchmarks compare robotics computing performance

Listen to this article

Voiced by Amazon Polly

RobotPerf chart illustrating performance factors.

Acceleration Robotics announced the beta release of RobotPerf Benchmarks in collaboration with global academia and industry leaders including AMD, Intel, Ford, Harvard, Klagenfurt University, Georgia Institute of Technology, Boston University, Johannes Kepler University Linz, Barnard College, Columbia University and Carnegie Mellon University.

This pioneering benchmarking suite is designed specifically to assess robotics computing performance, using ROS 2 as its foundation. A scientific paper on this collaboration is available at https://accelerationrobotics.com/pdf/robotperf.pdf.

The report benchmarks measurable elements such as latency, throughput, and power for various compute operations that control robotic devices. The use cases tested include perception, localization, control, and manipulation.

RobotPerf is a vendor-agnostic benchmarking suite designed to evaluate robotics computing performance across a diverse range of hardware platforms using ROS 2 as its common baseline. The suite encompasses ROS 2 packages covering the full robotics pipeline. It integrates two distinct benchmarking approaches: black-box testing, which measures performance by eliminating upper layers and replacing them with a test application, and grey-box testing, an application-specific measure that observes internal system states with minimal interference.

The test suite is intended to offer a standard methodology for evaluating performance differences between hardware architectures and between newer versions of a hardware platform.

Acceleration Robotics is a robotics semiconductor firm that offers services consultancy in robotics architecture and produces custom robotic brains by using hardware acceleration. The company also offers its own designs. Hardware (or IP cores ) for robots that speed up their response time and improve other features, including a reduction in power consumption.

Credit: Source link

Comments are closed.