Other products

ROBOTCORE Cloud

Tools to accelerate your robotic
computations with/in the Cloud

ROBOTCORE Cloud helps roboticists launch parts of their ROS 2 computational graphs into the cloud while addressing hardware acceleration, interoperability and scalability issues. It supports the most popular cloud providers including Azure, GCP and AWS.

Get ROBOTCORE® Cloud Need robotics consulting?
ROBOTCORE CLOUD

How does it work?

ROBOTCORE Cloud extends ROS 2 to a) provision cloud machines from either the ROS 2 CLI or the ROS 2 launch system, b) set up a secure private network and address interoperability issues and c) deploy the graphs as specified in the ROS 2 launch system extensions. Best of all, it aligns with ROBOTCORE® Framework, empowering developers with a unique flow to deploy their accelerators into cloud instances in a compute substrate-agnostic manner (across CPUs, GPUs and FPGAs).

Benchmarks

Launch your ROS
graphs into the cloud

ROBOTCORE Cloud extends the ROS 2 launch system to provision, set up, configure and launch computational graphs into remote machines in the Cloud. Supercharge your ROS graphs with the cloud while keeping the same exact development flow and using the common ROS syntax in launch files.

Supporting the top
cloud service providers

ROBOTCORE Cloud offers support for the top cloud service providers including Microsoft Azure, Amazon Web Services (AWS) and Google Cloud Platform (GCP).

Hardware acceleration in
the cloud for ROS

Leveraging the cloud provides roboticists with unlimited resources to further accelerate computations. Besides lots of CPU, cloud computing providers such as GCP, Azure or AWS offer instances that provide big FPGAs and GPUs for on-cloud hardware acceleration. This means once the ROS graph is partially in the cloud, architects can use custom accelerators to reduce and optimize robotic computations. But tapping into all that power while aligned with common ROS and robotics development flows is non-trivial.

ROBOTCORE Cloud helps robotic architects bridge the gap and simplify the use of hardware acceleration in the cloud for ROS. It allows to easily build IP cores for robots that target cloud instances and automate build processes, while aligning to the unified APIs for cloud provisioning, set up, deployment and launch derived from the standard ROS 2 launch system.

In partnership with
cloud robotics experts

ROBOTCORE Cloud results from cooperating with researchers from the UC Berkeley Automation Lab. A leading center for research in robotics and automation sciences.

Read paper

Developer-ready documentation and support

ROBOTCORE Cloud is served by seasoned ROS developers and for ROS development. It ships as a complement to ROBOTCORE, a one-stop shop for hardware acceleration in robotics. ROBOTCORE Cloud includes documentation, examples, reference designs and the possibility of various levels of support.

Ask about support levels

Benchmarks

(plots are interactive)

ROS 2 NODES

ORB-SLAM2 Simultaneous Localization and Mapping (SLAM) Node runtime (s)
(Measured the mean per-frame runtime obtained from the ORB-SLAM2 Node while running in two scenarios: 1) Default ROS 2 running on the edge with an Intel NUC with an Intel® Pentium® Silver J5005 CPU @ 1.50 GHz with 2 cores enabled and with a 10 Mbps network connection and 2) ROBOTCORE® Cloud running in the cloud with a 36-core cloud computer provisioned.)

Node speedup - ORB-SLAM2 SLAM Node runtime

4x

OTHER

Grasp Planning with Dex-Net compute runtime (s)
(Measured the mean compute runtime obtained over 10 trials while using a a Dex-Net Grasp Quality Convolutional Neural Network to compute grasps from raw RGBD image observations. Two scenarios are considered: 1) Edge - running on the edge with an Intel NUC with an Intel® Pentium® Silver J5005 CPU @ 1.50 GHz with 2 cores enabled and with a 10 Mbps network connection and 2) ROBOTCORE® Cloud - the same edge machine to collect raw image observations sent to a cloud computer equiped with an Nvidia Tesla T4 GPU. )

Grasp Planning speedup - Dex-Net computation total runtime (including network)

11.7x

Motion Planning Templates (MPT) compute runtime (s)
(Measuring the mean compute runtime while running multi-core motion planners from the Motion Planning Templates (MPT) on reference planning problems from the Open Motion Planning Library (OMPL). Two scenarios are considered: 1) Edge - running on the edge with an Intel NUC with an Intel® Pentium® Silver J5005 CPU @ 1.50 GHz with 2 cores enabled and with a 10 Mbps network connection and 2) ROBOTCORE® Cloud - the same edge offloads computations to a 96-core cloud computer. )

Motion planning speedup - Motion Planning Templates (MPT) compute runtime (including network)

28.9x

Do you have any questions?

Get in touch with our team.

Let's talk Case studies