This project aims to create an end-to-end management system for IoT deployments. Cebren will alow users to write queries to be performed on the data as well as set constraints for performace, energy use and cost (leasing cloud resources). The sytem will then automatically deploy and optimise the IoT components and cloud elements to maintain the user defined constraints, scale the sytem to meet demand and relocate computation to save energy and network bandwidth.
Cebren is currently in early development with an active team from Newcastle Univeristy's Digital Institute. As part of that team I am responsible for the real time monitoring components and also several of the simulation modules that allow cebren to estimate the impact of incoming workload on the IoT system.
We are working towards our 0.1 release and I will post links when that happens.
For my Master dissertation project I worked with the Newcastle University BeSiDE project to create an autonomous calibration and validation platform for their indoor location system (that’s him on the left).
RALF was designed to take a room plan and be able to fix his position to within a radius of 20cm. After 3 months of development RALF was able to provide a ground truth measurement within 3cm of his true position! RALF represents a solid starting point but unfortunately his movement and stability need improvement and he requires an optimised path planning system to efficiently survey a room. These are all areas I hope to improve upon in the coming months
RALF is built on open source technology, as well as a not insignificant amount of bluetack and gaffa tape. His brain is a Raspberry Pi, his movement is controlled via an Arduino, his sensor package is a laser range finder and electronic compass and his operating code is written in Python.