Upload
abdul-mujeeb-dalal
View
56
Download
2
Embed Size (px)
Citation preview
RECENT MACHINE LEARNING APPLICATIONS IN IOT ”THE REAL VALUE OF IOT LIES NOT IN CONNECTING DEVICES (ALTHOUGH THAT IS IMPORTANT) BUT IN ANALYTICS”.ABDUL MUJEEB DALAL CSE/09/13DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERINGNATIONAL INSTITUTE OF TECHNOLOGY SRINAGAR 06-01-2017
PRESENTATION OUTLINEBRIEF INTRODUCTION TO IOTBRIEF INTRODUCTION TO MACHINE LEARNINGWHY MACHINE LEARNING & IOT (TOGETHER)INDUSTRIAL INTERNETLIVING & HEALTHENERGY & RESOURCESAPPLICATIONS IN INDUSTRYCONCLUSION
INTERNET OF THINGSTHE TERM INTERNET OF THINGS WAS PROPOSED BY KEVIN ASHTON IN 1999.THE INTERNET OF THINGS (IOT) IS THE NETWORK OF PHYSICAL OBJECTS OR "THINGS" EMBEDDED WITH ELECTRONICS, SOFTWARE, SENSORS, AND NETWORK CONNECTIVITY, WHICH ENABLES THESE OBJECTS TO COLLECT AND EXCHANGE DATA.
THE IEEE DEFINITION OF THE IOT IS, “A NETWORK OF ITEMS — EACH EMBEDDED WITH SENSORS — WHICH ARE CONNECTED TO THE INTERNET” IEEE, 2015. THE IOT HAS ALSO BEEN REFERRED TO AS CYBER PHYSICAL SYSTEMS (CPS), M2M (MACHINE TO MACHINE), AND SIMPLY INDUSTRIAL INTERNET AND CONNECTED DEVICES.
IOT ALLOWS OBJECTS TO BE SENSED AND CONTROLLED REMOTELY ACROSS EXISTING NETWORK INFRASTRUCTURE, CREATING OPPORTUNITIES FOR MORE DIRECT INTEGRATION BETWEEN THE PHYSICAL WORLD AND COMPUTER-BASED SYSTEMS, AND RESULTING IN IMPROVED EFFICIENCY, ACCURACY AND ECONOMIC BENEFIT.
IOT TOPOLOGYFigure depicts the generic topology of the IoT viewed in layers to include the Datacentre, Gateway, IoT Devices and Sensors
5
THE NEXT BIG INTERNETInternet of Things
Trillion nodes
Fringe InternetBillion nodes
Core InternetMillion nodes
Smart metering
Transportation
Logistics
Industrial
Automation
Personal sensors
Phones
Building Automation
KNOWLEDGE HIERARCHY Turning data into wisdom. And the key point to IoT is that the knowledge hierarchy can be automatically summarized by machines.Here machine learningComes into play.
MACHINE LEARNINGARTHUR LEE SAMUEL :MACHINE LEARNING IS THE SUBFIELD OF COMPUTER SCIENCE THAT "GIVES COMPUTERS THE ABILITY TO LEARN WITHOUT BEING EXPLICITLY PROGRAMMED" HERBERT ALEXANDER SIMON: “LEARNING IS ANY PROCESS BY WHICH A SYSTEM IMPROVES PERFORMANCE FROM EXPERIENCE.”“MACHINE LEARNING IS CONCERNED WITH COMPUTER PROGRAMSTHAT AUTOMATICALLY IMPROVE THEIR PERFORMANCE THROUGH EXPERIENCE.“EVOLVED FROM THE STUDY OF PATTERN RECOGNITION AND COMPUTATIONAL LEARNING THEORY IN ARTIFICIAL INTELLIGENCE,MACHINE LEARNING EXPLORES THE STUDY AND CONSTRUCTION OF ALGORITHMS THAT CAN LEARN FROM AND MAKE PREDICTIONS ON DATA – SUCH ALGORITHMS MAKE DATA DRIVEN
PREDICTIONS OR DECISIONS, THROUGH BUILDING A MODEL FROM SAMPLE INPUTS.
Herbert Simon Turing Award 1975Nobel Prize in Economics 1978
Arthur SamuelCoined the term "machinelearning" 1959. Computer Pioneer Award1987
Flood of available data. IoT will produce a treasure trove of big data . We need to improve the speed and accuracy of big data analysis in order for IoT to live up to its promise. The only way to keep up with this IoT-generated data and gain the hidden insight it holds is with machine learning.
Machine Learning can be employed in security-related use cases, such as determining safe device behaviour and general usage patterns, which can subsequently help to spot and block abnormal activity and potentially harmful behaviour in IoT networks.
Increasing computational power. Taking advantage of limited functionalities of IoT devices. IoT devices are designed to carry out a limited set of functions. The solution includes a pebble-like device that gets installed in the home network, a mobile app that allows the user to control the device and monitor the network status and a cloud service where the data is consolidated and analysed using proprietary statistical tech and mathematical models coupled with machine learning algorithms.(MLaaS)
Efficient resource and energy consumption. Growing progress in available algorithms and theory developed by
researchers. Too often, machine learning requires a massive investment of time and
terabytes of data before it can deliver meaningful insights. Here we’ve an already built platform.
Increasing support from industries
WHY MACHINE LEARNING AND IOT (TOGETHER)
INDUSTRIAL INTERNETThe Industrial Internet is a term coined by GE and refers to the integration of complex physical machinery with networked sensors and software.The industrial Internet draws together fields such as machine learning, big data, the Internet of things and machine-to-machine communication to ingest data from machines, analyse it (often in real-time), and use it to adjust operations.
EXAMPLES1. The Google driverless car takes in environmental data from roof-mountedLIDAR, uses machine-vision techniques to identify road geometry and obstacles,and controls the car’s throttle, brakes and steering mechanism in real-time.
2. The Union Pacific Railroad mounts infrared thermometers, microphonesand ultrasound scanners alongside its tracks. These sensors scan every train asit passes and send readings to therailroad's data centres, wherepattern-matching software identifies equipment at risk of failure.
Falling prices for computing power and networked sensors mean that similar techniques can be applied to small, common devices like machine tools.
LIVING AND HEALTH
As the inclusion of more devices and appliances within the IoT ecosystem increases, methodologies for lowering their energy consumption impact are appearing. These include : To make appliances as ACs,
refrigerators etc. energy efficient.
Energy monitoring and Smart Metering.
ENERGY AND RESOURCES
EXAMPLES
Rachio Smart Garden Sprinkler
INDUSTRIAL APPLICATIONS
WORKFLOW IN MODERN MANUFACTURING
Company DescriptionTachyus End-to-end solutions for the oil and gas industrySentrian Remote patient monitoring platform.Maana Machine learning and data mining of massive amounts of data
in industriesVeros Systems Uses machine learning to isolate electrical problems in motor-
driven devicesNeura App that helps connected devices adapt to user behaviour
Augury Systems Uses signal-processing and machine learning to diagnose a machine’s health
Glassbeam Attempts to automate the converting of raw machine data into usable format
Building Robotics Localized air temperature regulation in commercial buildingsmnubo Data analytics platform for agriculture, smart homes etc.C-B4 data compression, pattern recognition, machine learning to
analyse IoT dataPointGrab Uses deep-learning to provide analytics and insights about
energy saving.Tellmeplus Develops software that can be used in the IoT industry for
predictive analyticsMoov Fitness wearable with an AI-based personal coach
Sentenai Cloud platform using ML techniques to automate ingestion and sorting
Imagimob AI-based motion intelligence systemFocusmotion Applies ML to data from body sensors and wearables.
MoBagel Uses machine learning to predict product trends and forecast sales
The future of IoT is more fascinating than this where billions of things will be talking to each other and human intervention will become least. IoT will bring macro shift in the way we live and work.Machine learning changes the machine understanding to human, With the capability of dealing Big Data , those application are becoming essential to our life & those application just can make the prediction based on the previous history, the future is still somehow uncertain, that's the magic. So the future realization of IoT’s promise is dependent on machine learning to find the patterns, correlations and anomalies that have the potential of enabling improvements in almost every facet of our daily lives.
“It’s not about ideas. It’s about making ideas happen.”
CONCLUSION