eDOTS - An Opportunistic Indoor Tracking Solution

About

eDOTS is an indoor tracking solution currently being developed in the department of Computer and Information Science at Indiana University-Purdue University Indianapolis (IUPUI). The goal of this project is to make use of existing sensors for the purpose of opportunistically tracking an object, accurately, as it moves through an indoor environment. Through this work we are attempting to provide a solution for indoor tracking that prevents the need for installing a costly/invasive tracking infrastructure all the while provide a high degree of accuracy in real-time to the end user. We feel that this project would benefit many application domains including that of health care and emergency rescue.

Framework

In our framework the eDOTS client application collects data from various different sensor devices (e.g. Wi-Fi, Vision-based, RFID) that are available and willing to participate in the tracking exercise within a given indoor environment. This exchange of information is made available to our client via JINI Lookup Services. Our use of JINI/Apache River is primarily based upon the JINI/Apache River concept of leasing of services. Due to the nature of mobile sensors this inclusion is a key component of our tracking infrastructure. This sensor data can then be fused (currently through two techniques: simple averaging based fusion, or a Kalman-based technique) with the end result of the fused data provided to the user via a graphical user interface. By providing such opportunistic discovery and communication we do not need to rely on a specific class of sensor or a preconfigured infrastructure - instead we make use of any devices, static or mobile, that we can discover, and our willing to participate, during our tracking process.

eDOTS

Sample eDOTS Display

Project History

This project, started in 2008 behind the faculty advisers: Dr. Rajeev Raje and Dr. Mihran Tuceryan, was created in an effort to create an inexpensive distributed indoor tracking system. Since then many other students have worked on the project including: Girish Joshi (MS 2008), Neha Talavdekar (MS 2009), Omkar Tilak (PhD 2011), Aboli Phadke (MS 2013), John Dennert (MS 2015), and Ryan Rybarczyk (2009 - Current).

Current Work

Current work on the project revolves around fully integrating both Wi-Fi Access Points and Active RFID tags/readers into the existing framework, thus making it truely heterogeneous in nature. Work is also being carried out with respect to making use of Android enabled phones for the purpose of mobile tracking and analysis of the frameworks impact on resource constrained devices. Other areas of active research include further investigating multi-sensor data fusion and sensor selection in a dynamic and agile environment. Specific technical challenges such as the integration of trust and reliability, as seperate parameters, at the various different sensor layers with the goal of providing improved sensor subset selection is ongoing research.

Publications & Posters

Awards/Honors

People

Faculty:

Funding

Contact Us

Ryan Rybarczyk
Email : rrybarcz@cs.iupui.edu