UP Squared RoboMaker Pro Kit

The new UP Squared RoboMaker Pro Kit, powered by Intel®, Amazon Web Services and Cogniteam, provides an easy approach to create a machine vision application for an autonomous mobile robot (AMR) in a single day.
With Intel® Movidius™ Myriad™ X vision technology, deep learning capabilities at the edge and pre-configured ROS tools from AWS RoboMaker, robots can perform more intelligent tasks, such as enabling avoidance of dynamic obstacles within one second & replanning the path.
This seamless integration of embedded system and AWS RoboMaker delivers an unprecedented experience of AI simulation-to-reality while allowing developers to speed up the time to production.

Overview　|　Key Features　|　Kit Details　|　Tech Specs　|　Community Support

Overview

Powerful vision. Powered by Intel®

With Intel® x7-E3950 processor, UP Squared is designed with low-power high-performance components, UP Squared is a favorite among developers of mobile applications.

Featuring Intel® Movidius™ Myriad™ X, it is capable of speeds up to 105 fps (80 typical). This outperforming VPU is dedicated hardware-based acceleration for deep neural networks for powering AI edge computing processes.

Developers can take a shortcut by using off-the-shelf tech, building an AMR in one day.

Intelligent Simulation-to-reality

AWS RoboMaker is pre-integrated with popular AWS analytics, machine learning, and monitoring services. It’s very easy to add functions like video streaming, face and object recognition, voice command and response, or metrics and logs collection to your robotics application.
AWS RoboMaker services are exposed as ROS packages so that developers can easily use them to build intelligent functions into their robotics applications without having to learn a new framework or programming language.

Vision from edge to cloud

Intel® RealSense™ depth camera D435i with an inertial measurement unit (IMU) inside, which opens the door for rudimentary SLAM and tracking applications allowing better point-cloud alignment. Additionally, IMU allows your application to refine its depth awareness in any situation where the camera moves.
OpenVINO™ (Open visual inference and neural network optimization) toolkit provides a ROS–adapted runtime framework of neural network. By leveraging Intel® OpenVINO™ toolkit and corresponding libraries, it can reinforce the efficiency of deploying applications for vision inference.

Key Features

What can you achieve with this kit?

Enable your AMR to see in 3D and make autonomous decisions.

Utilizing the 4 Key components in AWS RoboMaker and executing SLAM efficiently

Integrated Development Environment: All dependencies are pre-packed and ready to go, so developers can easily spin up a ROS environment.
Simulation: Accelerate testing of your ROS applications with a simple API call, developers can spin up their ROS applications into a simulation environment, and access ROS simulation tools such as Gazebo, rqt, rviz (to view the robot’s SLAM), Terminal.
Fleet Management: Deploy your robot applications to robots in production over-the-air and it leverages AWS Greengrass daemon that would run on your robots.
Cloud Extensions for ROS: Enable developers to extend their ROS applications into the cloud and leverage cloud services from AWS, i.e. Boto3 and ECS SDK. Other cloud extensions automatically create connections and make API calls to AWS services, such as Amazon Kinesis, Amazon Rekognition, Amazon Lex, and more.

❝ The UP Squared RoboMaker Pro Kit includes vision compute hardware on board.
This enables us to free our CPU for decision making while processing data and building the world model of the robot in an optimized manner, by having production ready boards integrated with Intel® Movidius™ Myriad™ X VPU and an Intel® RealSense™ depth camera enables to jump from low-level control of the robot to a fully autonomous robot with cloud connectivity within days. ❞

Dr. Yehuda Elmaliah

CEO of Cogniteam

See its possibilities

By leveraging UP Squared RoboMaker Pro Kit, developers can easily upgrade their PoC and scale to production.

Play Video

Kit Details

Hardware overview

UP Squared board with Intel® Atom® X7-E3950 processor, onboard 8GB DDR4, 64GB eMMC
Intel® RealSense™ depth camera D435i (IMU inside)
UP AI Core X – Mini-PCIe card with the Intel® Movidius™ Myriad™ X
Intel® AC9260 WiFi Kit (via M.2 2230)
Servo and DC motor with encoder
Cogniteam Motor Control Board
Wheels

Preinstalled Software

Ubuntu* 18.04 desktop
ROS1 – melodic/ROS2 – dashing
Intel® Distribution of OpenVINO™ toolkit 2019 R1.1 release
Intel® Media SDK
Drivers for Intel® VTune™ Amplifier, Intel® Energy Profiler, Intel® Graphics Performance Analyzers
MRAA and UPM I/O and sensor libraries for C++, Python*, Java*, and JavaScript*

Tech Specifications

UP Squared RoboMaker Pro Kit
Hardware - UP Squared
CPU	Intel Atom® x7-E3950 (up to 2.00 GHz)
Graphics	Intel® HD Graphics 505, supporting 4K Codec Decode and Encode for HEVC4, H.264, VP8
RAM	8GB (dual channel) LPDDR4
AI Accelerator	UP AI Core X with 1x Intel® Movidius™ Myriad™ X VPU 2485 (via Mini-PCIe)
Storage capacity	64GB eMMC
OS	Pre-load Ubuntu image
Ethernet	2x Gb Ethernet (full speed) RJ-45
WiFi & Bluetooth	Intel® AC9260 card (via M.2 2230)
Video output	HDMI+DP
Display interface	1x eDP
Camera Interface	1x MIPI-CSI (2 lane) & 1x MIPI-CSI (4 lane)
USB	2x USB 2.0 pin header 1x USB3.0 OTG 3x USB3.0 (Type A)
Expansion	1x 40 pin GP-bus enabled by Altera MAX 10 FPGA 1x 60 pin EXHAT 1x Mini-PCIe (used by AI Accelerator) 1x M.2 2230 E-key (used by WiFi & Bluetooth) 1x SATA3
Dimension	85.6 mm × 90 mm (3.37" x 3.54")
Power input	5V@6A with DC jack 5.5/2.1mm
Certificate	CE/FCC Class A, RoHS compliant, REACH, Microsoft Azure certified
Hardware - Intel® RealSense™ depth camera D435i
Depth Field of View (FOV)	87°±3° x 58°±1° x 95°±3°
Depth Output Resolution & Frame Rate	Up to 1280 x 720 active stereo depth resolution. Up to 90 fps.
Image Sensor Technology	Global Shutter, 3μm x 3μm pixel size
Minimum Depth Distance (Min-Z)	0.11 m
Inertial Measurement Unit (IMU)	Yes
Click here to see more information about RealSense™ camera D435i

Battery (not included in the kit)
Battery prerequisite (not included in the kit)	Lipo battery with T connector 7.4V 2 cell
Battery charger (not included in the kit)	Charger 2S-3S 7.4-11.1 V for Lipo Battery
Software - Cogniteam
Get started	Connecting all parts within the kit
Setup	- WiFi
Simulation	Simulation example

UP Squared RoboMaker Pro Kit supports ROS & ROS2
Features	ROS	ROS2
Platforms	Tested on Ubuntu Maintained on other Linux flavors as well as OS X	ROS2 is currently being CI tested and supported on Ubuntu Xenial, OS X El Capitan as well as Windows 10
C++	C++03 // don’t use C++11 features in its API	Mainly uses C++11 Start and plan to use C++14 & C++17
Python	Target Python 2	>= Python 3.5
Middleware	Custom serialization format (transport protocol + central discovery mechanism)	Currently, all implementations of this interface are based on the DDS standard.
Unify duration and time types	The duration and time types are defined in the client libraries, they are in C++ and Python	In ROS2 these types are defined as messages and therefore are consistent across languages.
Components with life cycle	In ROS every node usually has its own main function.	The life cycle can be used by tools like roslaunch to start a system composed of many components in a deterministic way.
Threading model	In ROS the developer can only choose between single-threaded execution or multi-threaded execution.	In ROS2 more granular execution models are available and custom executors can be implemented easily.
Multiple nodes	In ROS it is not possible to create more than one node in a process.	In ROS2 it is possible to create multiple nodes in a process.
roslaunch	In ROS roslaunch files are defined in XML with very limited capabilities.	In ROS2 launch files are written in Python which enables to use more complex logic like conditionals etc.
AWS Robomaker		Cross-Compiling Applications with Colcon Capturing log data with rosout