SenSquare is an IoT service middleware able to collect raw sensor data from open sources (e.g. OpenWeather, ARPA, ThingSpeak …) or Mobile Crowdsensing dedicated campaigns and deliver it in the form of aggregate services. Final users can build their own services through a visual programming language and instantiate them in geographic areas. Try the current release version out at http://sensquare.disi.unibo.it. You can also read the main publication.
Single Source of Truth
The platform is currently supported by a MySQL monolith database, which gets updated periodically. We need to explore how to make it dynamical, requesting the data directly to the source and using the database as a local cache.
Official Documentation and Manual
We need the main Github page to be better organized and documented. We also need an online user manual with a walkthrough
Arrowhead systems and tools need a seamless way to be orchestrated in a pipeline. The work involves the use of various systems (RPi, Smartphone, ...) and their automation through the Arrowhead Choreographer in a practical demo.
Web of Things
The Web of Things seeks to counter the fragmentation of the IoT, making it much easier to create applications without the need to master the disparate variety of IoT technologies and standards. Digital twins for sensors, actuators and information services are exposed to consuming applications as local software objects with properties, actions and events, independently of the physical location of devices or the protocols used to access them.
Mobile Crowdsensing (MCS) is the paradigm for which phenomena of common interest, happening mostly in urban scenarios, are being monitored through the aggregation of the contribution of a crowd of mobile sensors and/or smartphones (e.g. Google Traffic API). You can read about MCS in this paper.
CrowdSensing on a CrowdSourcing platform
We will build a system integration between a MCS mobile system and a CrowdSourcing platform such as Amazon Mechanical Turk to handle micro-payments.
The Costs of CrowdSensing
The idea is to compute the cost to deploy and maintain an infrastructure to sense specific data. Then it is possible to determine what is the price someone is allowed to pay to users if the same data is obtained through crowdsensing. The system should leverage open data and real parameters.
Client for MCS on Ethereum
This is an early idea that aims to design a MCS scenario over a commercially available blockchain (Ethereum). Optionally we can think of developing the relative smart contracts, but the main focus is the mobile client.
Activity Recognition & Context Awareness
Here we have miscellaneous works on Activity recognition and Context Awareness. Most of them are LAM-enabled as we will leverage the inner sensors of smartphones.
Path Planner for Bicycles
The aim is to design a path planner (like GMaps) specifically for bicycles, which does not give only teh shortest path, but the best path according to a number of other features (road quality, etc..)
- Thesis by V. Conte (upon request)
Comparison Study on Pedometers
We will compare different pedometers and find a way to feed synthetically sensor data to an android emulator to create a solid test.
Text and Drive
We will use the Android sensors to understand whether a person is in a car and driving (i.e. sitting in the front left seat).
In order to test this you must have access to a car.
Drones and autonomous systems
Applications and system analysis for unmanned aerial vehicles (UAVs)
Simulator development (DROMNeT++)
We want to have an integrated simulator tool to model multi-UAV systems. The idea is to integrate the following simulators/frameworks:
- wireless communication (e.g. OMNeT++);
- mobility control (e.g. Ardupilot);
- 3D flight simulators (e.g. Gazebo, Airsim)
Mobile application for drone swarms
Parrot-Sphinx is a simulation tool developed by Parrot to test the control of a drone. We want to investigate the possibility to control multiple drones (internship). If this is the case, we want to develop an application that is able to easily mange a swarm of drones.