Collaboratively administrate empowered markets via plug-and-play networks.
Adding Knowledge to IoT
IoT connectivity and Artificial Intelligence services provide almost unlimited new ways of capturing and analysing a wealth of structured & unstructured data. By using a single, common, knowledge automation tool as the brain of the data ecosystem, real business value can be achieved in the form of expert automated decisions.
In this sample
This sample demonstrates the execution of XpertRule on both an edge device for local decisions (Raspberry Pi) & running in the cloud for augmented decisions (IBM Bluemix). The specific technologies include:
XpertRule Decision Author (Design Environment)
XpertRule Node.js Runtime Environment (installed on a BlueMix Instance & Raspberry Pi)
IBM Watson IoT Platform (for communication with IoT hardware)
Raspberry Pi running IBM IoT client software
IBM Alchemy Vision (to count number of people in an image)
IBM Insights for Weather (for current weather conditions)
The premise is to take IoT sensor values, along with face recognition and weather report, to control a fictional heating, ventilation & cooling (HVAC) system.
1 | Retrieve the current EDGE ANALYTICS decision (knowledge running on the edge/hub)
2 | Retrieve the current IoT sensor values
3 | Use Alchemy Vision to detect the number of people in the image
4 | Use Insights for Weather to get the current external temperature
5 | Apply cloud knowledge to (possibly) override the EDGE ANALYTICS HVAC setting with an enhanced (given weather/people data) CLOUD ANALYTICS decision
The main decision tree takes the current environmental state and decides which state the HVAC system should be set to.
Notice that we use sub-trees to separate the logical environmental state from the actual sensor values. These sub-trees can be seen below:
Room Temp Decision Tree
Mapping from the thermometer to a logical state
Room Lighting Decision Tree
Mapping from the light meter to a logical state
Thermostat OnOff Decision Tree
Compare the thermostat to the temperature
The main decision tree takes augmented cloud results (number of people in the image & current weather report) along with the HUB decision to decide whether to override the HUB decision. n.b. An Empty leaf signifies that the decision should NOT be altered.
A custom XpertRule library has been created to allow for simple integration with the Alchemy Vision and Insights for Weather BlueMix services.
Here is the circuit diagram showing how the sensors are connected to the Pi
There is a python script running on the Pi which reads the sensors, executes local knowledge and updates the cloud.
Click here to see the actual python script
Edge vs Cloud