A hitchhiker’s guide to Ambient Intelligence
Blog Posts:Mu Sigma
Published On: 08 December 2014
You live in an enriched environment today powered by smart devices, sensors, actuators, microprocessors and microcontrollers everywhere. These electronic environments are sensitive and responsive to you. Are you ready for the Ambient Intelligence around you?
Ambient Intelligence (AmI) refers to electronic environments that are sensitive and responsive to the presence of people. The underlying premise behind the evolution of AmI is that, when an environment is enriched by networked technologies (sensors, processors, actuators, displays), a system can be built to collect and process real-time and historical data and make decisions benefiting the users of that environment.
AmI is not a disruptive phenomenon. Rather, it is an evolution and convergence of academic research in multi-disciplinary fields. While this research has been going on in controlled environments for over two decades, its implementation in real-world scenarios has been fairly limited to date. Smartphones, with their ever-increasing plethora of sensors (gyroscope, temperature, humidity, heartbeat and more) and rapidly multiplying processing power, are finally providing the required conduit for AmI to support realistic scenarios.
Why? Aside from their embedded sensors, smartphones are context aware (think driving directions), adaptive (voice recognition, favorites lists), and anticipative (autocorrect). Wearables such as Apple iWatch and Google Glass sport many of these same features and thus are also great venues for AmI. Another example might be Google’s driverless cars.
At Mu Sigma, we of course are always curious about how trends like AmI impact Decision Sciences.
We have worked on several interesting projects that help highlight the potential of AmI in the decision sciences realm.
For an Internet of Things project involving management and monitoring of live electrical equipment, we had to bridge machine data and legacy protocols with Wi-Fi and Bluetooth technologies for monitoring and managing industrial devices. We built a custom mobile application to conduct real-time monitoring of machine operational parameters. By allowing users to save this data to their device as snapshots, we provided in-mobile analysis that helped inform basic preventive maintenance.
As an extension to this project, we are enabling AmI by allowing field service personnel to automatically detect the devices on site and provide device specifications and maintenance instructions via smart devices and wearables. We are also working on streaming real-time sensor data onto Google Glass. By comparing real-time data with predictive engines deployed in the cloud, field service personnel can determine variances or anomalies in operational efficiency, and make decisions about preventive or corrective actions on the spot.
By combining AmI and machine learning, Mu Sigma is helping optimize field service maintenance and eliminating the need for field staff to carry test kits, connectivity cables and laptops. Additionally, by using geo-location, field service personnel can easily be guided to under-performing units in a large industrial or residential sites, thus reducing the time and cost of maintenance.
Is your organization applying AmI techniques in order to improve efficiency? Share your stories below.