This session will describe and demo methods to connect the Intel Edison to Amazon AWS in order to create a versatile IoT structure. The Intel Edison is a powerful system on chip module, the size of a postage stamp with powerful on board processing. It can be used as a sensor hub to gather data, a control board for actuators, and a gateway to connect to the cloud. When combined with the powerful services offered by AWS it can form the basis for many IoT solutions.
AWS DevDay San Francisco, June 21, 2016.
Presenter: Martin Kronberg, Intel oT Evengelist
6. 1. IDC
2. MC/EDC: The Digital Universe of
Opportunities
3. Goldman Sachs
4. IMS Research
The Internet of Things is …
Home
Mobile
Network
Industrial
Gateway
DC/Cloud
44ZETABYTES2
3
COST OF
SENSORS 2X
PAST 10
YEARS
COST OF
BANDWIDTH 40X
PAST 10
YEARS
COST OF
PROCESSING 60X
PAST 10
YEARS
50BDEVICES1
85%UNCONNECTED4
21
212BSensors
7. 7
1. Research & Development Roadmap for Emerging HVAC Technologies. U.S. Department of Energy (October 2014).
2. The ‘Internet of Things’ Is Now, Connecting the Real Economy. Morgan Stanley (April 3, 2014).
3. How much electricity is lost in transmission and distribution in the United States? U.S. Energy Information Administration (April 6, 2016).
The Internet of Things (IoT) is fueling
innovation across a range of industries to
optimize processes and increase efficiency.
4. "How Manufacturers Use IoT for Operational Efficiencies." Industry Week (October 21, 2015).
5. Tech Today Video Series Episode 3: Vnomics. Intel (March 4, 2014).
Need for IoT
TRANSPORTATION
Fleet management, freight tracking,
urban congestion management
Enhanced driver coaching can save
nearly 7% on fuel costs.5
RETAIL
Point of sale, vending machines,
supply chain
Supermarket cost-to-sales ratios can be
reduced 2% by automatic checkout.2
SMART BUILDING
Energy use monitoring; sensors in heaters
and chillers to find inefficiencies
Greater efficiency for HVAC systems, which
account for 41% of U.S. building energy use.1
INDUSTRIAL &
MANUFACTURING
Assembly-line equipment reporting,
inventory management; automation
Manufacturers using IoT report 82% increased
efficiency and 49% fewer defects.4
ENERGY
Environmental data logging, substation
monitoring, grid efficiencies
IoT can optimize energy grids, which lose
~6% during transmission and distribution.3
10. Yocto* Project
Linux* OS
image
Supports Intel® Edison, Intel® Galileo boards
as well as Intel® IoT Gateway
Middleware libraries for IO and sensors /
actuators
IDEs and tools to create, run, debug and
optimize IoT solutions
C/C++, JavaScript*, Python* and Java*
programming language support
Advanced power efficiency/performance
optimization tools
Examples with WindRiver* Helix* Cloud
IoT Cloud and Edge Analytics for data
collection, visualization and analytics
Intel® IoT
Developer Kit OS
Image (Intel®
Edison, Intel®
Galileo boards)
IoT OS
*Other names and brands may be claimed as the property of others.
Intel® IoT Developer Kit Overview
For more details go to https://software.intel.com/iot
Intel Edison
Board
IO Library
(SPI, GPIO, I2C, AIO,
UART)
HardWareOS
Image
CloudEdge
Analytics
Intel Galileo
Board
Intel® IoT Gateway
Middleware
Communication Stack
(Wi-Fi, Bluetooth®
Technology, ZigBee*, etc.)
Manageability, Security,
RealSense SDK, etc.
API Bindings (C/C++, Node.JS*, Python*, Java*)
Sensor & Actuator
Libraries
IDE/Tools
Prototype Evaluate/ Prototype / Product
IBM* BlueMix*, Microsoft* Azure*, Amazon* AWS*, General Electric Predix* and more...
Intel® D2000 Sensors / Actuators
10
12. Almost all IoT solutions start with a proof of concept
• Fast and Easy to
develop
• Familiar Tools
• Low cost
• Mature and stable offering
• Reliable and scalable
• Flexible and secure
13. What is a System On
Module (SOM)?
• Falls between an MCU and Single-
Board Computer
• Includes processor, RAM, I/O
• Lacks standard I/O peripherals
• Commonly requires carrier board
• Offers densely packed, highly
integrated components
• Typically low-energy
• Primary goal: Accelerated time to
market, reduced cost
14. SOM Strengths and Weaknesses
Strengths:
• Easier to upgrade compute to next
generation by swapping on carrier
• Makes no assumptions about
required I/O, customer can tailor
carrier to their needs
• Faster time to market, especially if
software and tools are provided to
program the core unit
• FCC and global pre-certification
saves time and money
Weaknesses:
• Module must be compatible with
carriers across upgrades for
swappable value proposition to
emerge
• Connection points can be source
of system fragility
• Assumes the need for a carrier,
which can take up more space
than a “board-down”
implementation
16. Intel® Edison Module
Key Benefits
• Many supported communication
protocols
• High density 70-pin connector
allows small footprint
• Programmable with C++,
JavaScript, Python, and Java
• Extensive Sensor Library
• High compute power with x86
• Low Power consumption
• FCC Wireless certification for Wi-Fi*
and Bluetooth® worldwide
• Simple integration with cloud
services such as AWS*
21. Sensor Support and Control Protocols
200+ sensor drivers for maker and industry
Leverages MRAA communication layers
Adds ZigBee, Z-Wave, Bluetooth, Modbus, BACnet,
RS422/485
Industrial Sensors: Aeon Labs*, Bosch*, Comet*, Maxim
Integrated*, Omega*, Veris*, and others
2
22. Wireless Certification
FCC Certification (United States) is
complex, time consuming and
expensive
Can cost $250,000 and take 6+ months
to certify a wireless module globally
Worldwide certification can take months
and cost hundreds of thousands of
dollars
SOMs allow you to avoid this cost
27. Case Study: KINETIC*
Objective: Provide workers with a
solution to reduce lifting-related injuries
by unsafe lifts in the workplace.
Project: Wearable waist and wrist-
based lift trackers
Team Size: ~10
Location: New York, NY
Website: www.wearkinetic.com
28. KINETIC* Product
Prototyped on Intel® Edison board,
replacing alternate SoC
Used I2C, Bluetooth® Low Energy, Wi-
Fi*, GPIO
Acquired 2,000+ field experience with
Intel Edison
Python* micro-service for code updates
and deployment
1500mAH LiPo battery
Cost: $300
29. Case Study:
Dronesmith Technologies*
Objective: Enable drone developers
with an Intel® Edison compute module
based prototyping and experimentation
platform “Luci”
Project: Luci flight controller, QK II
drone
Team Size: ~10
Location: Las Vegas, Nevada
Website: www.dronesmith.io
31. Case Study:
Robodub, Inc.
Objective: Live-action streaming for
robots and drones, gamification of
robotics
Project: RamBot
Team Size: ~12
Location: Bellevue, Washington
Website: www.robodub.com