Mainflux is open source, Apache 2.0 licensed high performance and scalable IoT middleware platform. Mainflux is designed to handle and process data coming out of your IoT devices, edge gateways, LoRa gateways and deployments.
Mainflux is developed using microservice architecture and Go programming language. Thanks to its design and technology choices, Mainflux can be deployed on gateway type device, on a single server or it can be scaled using kubernetes to multidata center systems. Mainflux is cloud agnostic and can be set up on any public or private cloud. Mainflux IoT Platform has several components:
This session will cover main features of Mainflux IoT platform and show how to create full featured IoT solution using Mainflux IoT middleware platform. We will explain:
The holy grail of the Internet of Things is the ability to easily distribute the intelligence of your application across the Cloud and the Edge. Being able to run analytics, AI or store data at the Edge addresses many common and key enterprise IoT scenarios. Come learn how to easily create deployments for IoT devices that include AI, Machine Learning, Stream Analytics, as well as your own custom code on devices smaller than a Raspberry PI.Speaker(s):
While IoT brings great business value and opportunities to the organizations, it also brings unseen security threats & risk at the same time. Though the traditional security controls do have a major role to play in IoT security, but it does not solve all the challenges resource-constraint-unsecure IoT devices bring with it. Due to sheer number of these IoT devices, constrained environment they operate in, and lack of inter-operability and security standards, it's a need of hour to address IoT security challenges with the different set of approach and purpose-built security controls to get the long term security benefit and right ROI over longer run. To address these IoT challenges, Intel's built in hardware security features together with SW ecosystem plays a very important role in building & deploying the secured IoT devices and system for risk-free* operations.Speaker(s):
The session provides an overview of the development and intelligent voice data analysis from a machine learning perspective; a historical, state-of-the-art view and a view on some future trends in the field of artificial intelligence. The session describes some areas within voice recognition domain which seem to be important for applying machine learning in medical diagnosis. This describes a recently developed method of detecting respiratory problems quickly by recognising the changes in voice over time. Machine learning algorithms are applied here.Speaker(s):
Whether an industrial equipment company is building equipment and devices for manufacturing, energy, medical, transportation or another segment, they are experiencing a dramatic digital transformation from a focus on hardware technology to a new modern focus on software technology that allows the consolidation of industrial systems. Two software technologies are helping to bring this digital transformation to the design and building of the latest generation of industrial equipment and devices: virtualization and containerization. Attend a Wind River session and discover how virtualization and containerization technologies are enabling the consolidation of industrial systems and driving the digital transformation of the manufacturing, medical and rail transportation market segment.
Abstract: "Applications at the Edge" — so what does that truly mean? Since the introduction of Apple Siri, voice interfaces have been commonplace in today’s consumer electronics. There’s even a new product category for consumer electronics called, "smart speakers", which is a household device (most notably, the Amazon Echo) that allows the user to interact with the unit completely with their voice. Voice interfaces work when a mobile phone or smart speaker captures the user’s voice and then subsequently transmit the audio data to the cloud for text transcription (converting the voice data to text). After the audio has been converted to text, some action is performed on behalf of the user or an audio response is given. For a truly hands-free experience, these devices are capable of allowing the local microphone to be active 100% of the time, and the unit will give you a response when it hears a “wake word”, such “Hey Siri”, “Alexa”, or “Ok Google”. In that case, the speech is recognized locally, and all subsequent audio data is sent to the cloud for processing as before.
Now, there are several companies that offer cloud-based speech recognition and NLP (natural language processing) tasks as a cloud-based service. However, what if you wanted to create a voice interface for either a desktop application or for an IoT device and you don’t want to pay a 3rd party service indefinitely during the lifetime of your product? What if you're an automaker, and you want your drivers to control various features of the car, without needing reading the (dreaded) user manual? What if you needed to voice-enable a software application that works with sensitive personal data, such as a hospital with medical data or a bank with financial data? All of these situations are real-world scenarios where using a cloud-based speech and NLP service is neither practical nor pragmatic.
Therefore, the purpose of this talk is to explain challenges that developers will face when they need build a fully-offline voice interface. The participant will not need to have any experience in Natural Language Processing or linguistics. The talk will assume that the user already has access to a platform or service that provides speech-to-text services since this is a commodity service provided by multiple companies in the industry. Therefore after explaining the problem statement and use cases, we will deep-dive and explain how to use Intel NLP Architect to process statements in Natural Language and perform "intent extraction" on text statements.
While Industry 4.0 (I4.0) is adopted by the global players with a high pace, there is a risk that the SMEs, which are the backbone for industrial innovations, are left behind. In Germany, as an example, the SMEs are responsible for 97% of all exports, 60% of all manufacturing related employments and originate 1300 of the world wide Hidden Champions.
Service robotics includes but not limited logistic systems robots, defense robots, field robots (milking robots), public relations robots, powered human exoskeletons, medical robots. The market for service robotics is experiencing a perfect storm of circumstances that are driving growth. These circumstances include market readiness, technology maturity, delivered value, cost and availability, and the maturity of related technologies.
Generally, two or more SoC are used for service robotics. One SoC is running with Robot Operating System (ROS) to collect the information of environment and control motion of Robot. Other Soc are responsible for hosting GUI to interact with customers. Due to the rich app ecosystem, Android is always selected as GUI OS. Rosbridage or RosJava is used as the communication solution between Ros OS and Android OS.
Current two SOC solution bring some issues to customers. Firstly, two SOC solution means high cost and maintain effort. Secondly, TCP/IP based comminution brings performance overhead. Lastly, without security checking mechanism bring more risks of DDOS attach on some core nodes of ROS.
Our novel solution, named RosRoid, is trying to reduce the paint point by consolidating ROS OS with Android in single computing platform. Two mixed OS are running in single kernel and isolated by each other. Moreover, binder based communication solution reduces performance overhead and brings security check mechanism to avoid DDOS attack.Speaker(s):