Posts by scis-admin

Smart HelperBot



Focuses on the quality of life and security while providing solutions that increase comfort and entertainment.

ISHA: Intelligent System for Home Automation

Intelligent System for Home Automation

Home automation solutions – such as centralized control over lights, air conditioning and appliances – seek to improve the quality of life of the residents by giving them control over the different systems in the house with just the push of a button. Taking advantage of this technology, the intelligent home goes even one step further; automating the decision making process and operating the house systems based on information obtained from the environment.


ISHA: Intelligent System for Home Automation

Our intelligent home will be implemented as a wireless home network, with every device in the house being a node (or connection point) of this network. Even the home residents, by carrying wearable sensors, will belong to the home network. The nodes will be continually collecting data from the environment and ISHA will react based on the information obtained.

Users will be able to interact with the intelligent home by using a computer, smartphone or tablet, as well as by giving the house direct voice or gesture commands.


Node Localization

One of the novel features to be implemented in ISHA is that of node localization, which will allow our system to know the location of all the devices in the house. This means that our intelligent home will not only be collecting data from its surroundings, but it will also know where this data is being obtained and it will react differently depending on the location. Node location information will, essentially, make the intelligent home even more intelligent. How? A simple example will illustrate it better.
You are sitting in your living room watching television. ISHA detects that your body temperature is raising and your heartbeat is getting faster. The house detects that you are in the living room and turns on the air conditioning only in that room.
Not only has the house made you more comfortable by lowering the temperature, but it is also being energy-efficient by only turning on the AC on the specific room where you currently are.
For this to work, the location information has to be very precise; the kind of precision that is usually very expensive. To overcome this, our proposed approach uses multilateration – the same technology used in GPS – to obtain the coordinates of the nodes inside our home network.
However, GPS measurements have errors ranging from 4 meters to 10 meters. That could be the difference between your home and your neighbor’s home! While GPS uses 3 reference satellites to calculate your location when you are driving, our system will reduce the error by using most of the available reference nodes to calculate a target’s location.

Benefits of ISHA


Past Projects

– Statistics Game (Untitled)

Education is essential to the growth of a young mind. Mathematics is often used as a tool to reinforce critical thinking and sharpen skill sets.This web-based, shopping simulation game will teach basic statistics principles to middle school students.
Advisor: Dr. Nagarajan Prabakar
Student Members: Jose Maldonado, Martin Di Diego, Kennet Postigo
– Virtual Reality Research
Virtual Reality is a simulated experience that shows its user a surreal environment created in real-time. The Discovery Lab dedicates research to further understand Virtual Reality concepts such as virtual motion, virtual
object placement, user interactions within a virtual space and effective scenario simulation. This will be applied into building a virtual reality set prototype.
Advisor: Dr. Nagarajan Prabakar
Student Members: Jose Maldonado, Felipe Z. Carlier, Shadeah Ferris-Francis
Swarm Controlled UAV
The use of Unmanned Aerial Vehicles in the area of applied research is accelerating at an astonishing rate. With the advances made in technology to extend the control of these vehicles, the opportunity to control multiple aircraft at a time presents itself as a natural progression towards autonomy. Using a wireless link to control the aircraft, the objective is to create scouting vehicles that can cover a large area in a short amount of time.
Advisor: Dr. Nagarajan Prabakar
Student Members: Deon Wilkins, Fernando Campo

FIU Discovery Lab Members Build a Decentralized Banking System of Top of Bitcoin


The four entrepreneurs have combined training in hyper-scale distributed financial systems, decentralized Bitcoin applications, artificial intelligence algorithms, and developing machine learning. The group was drawn from the two universities — Stone Hill College and Florida International University (FIU) — and their company, Master Branch is a startup that competed against 20 other companies for the final prize of US$100,000. For the competition, they named their team Decentral Bank to reflect their ideas.

The team was the only university group in the competition, as well as the only team to present a Bitcoin solution that will allow banks to accept Bitcoin without actually managing cryptocurrencies. They accomplished the task by implementing multisig technology. You can view the white paper here……

5th Burn your Brain Event with Discovery Lab Quad-Copter Team Workshop

On Saturday, May 24, Dr. Jong-Hoon Kim and his staff were the main event of our final Burn Your Brain event of the semester.  They presented the Discovery Lab’s QuadCopter project, showing the 25 attendees the technical ins and outs of building robotic flying devices. The students also learned how to write code to interpret realtime data from the same types of motion sensors used in robotic flying devices. Kim was assisted by Deon Wilkins, Fernando Campo, and Jaime Acevedo from the Discovery Lab. This was the 5th and final Burn Your Brain event of the semester, which has now reached approximately 120 computer students from Dade and Broward counties.

Discovery Lab: National Robotics – Open House event

Thank you for all student members, volunteers, and staff for hosting an open house event as part of the National Robotics Week for an enthusiastic 230 registered attendees.

Efficient Concurrent Operations of Telepresence Avatars

Efficient Concurrent Operations of Telepresence Avatars

Jong-Hoon Kim, Nagarajan Prabakar, Cynthia Tope


Telepresence robot extends operation in remote locations as avatars. The remote physical presence application will enable individuals with mobility limitations such as disabled veteran or police officer to perform regular duties through telepresence robots known as avatars or TeleBots. Certain scenarios require many officers to operate on several TeleBots (1:1, 1:many, many:1, or many:many) in the same service area. Thus, we need an intelligent, fault tolerant management system to support these scenarios.

This system requires dynamic resource allocation such as communication bandwidth, battery power, spatial proximity, etc. In addition, conflicts among the concurrent telepresence operations need to be resolved efficiently. We propose a novel architecture that provides a fault tolerant multi telepresence robot Management System. This system will allow us to coordinate the assignment, scheduling, monitoring, and administrating multi telepresence robots and multi operators efficiently.  Further, we present strategies to resolve conflicts among concurrent operations.

Full paper : ISR-13_Concurrent_Avatars.pdf

Standardized Linearization and Vectorization Algorithm for Arm Motion Control of A Humanoid Telepresence Robot

Standardized Linearization and Vectorization Algorithm for Arm Motion Control of A Humanoid Telepresence Robot

Michael Waddell, Joel Villasuso, Daniela ChavezGuevera, and Jong-Hoon Kim


Recently, the new motion sensor: the Kinect is being used for natural motion retrieval with no additional equipment on the operator, less computational demand, and it is cost effective. But many restrictions apply because the results retrieved from the Kinect include noisy data which disturbs precise and smooth robot motion control. We propose a simple linearization algorithm to improve the accuracy of the data retrieved from the Kinect and designed a vectorization algorithm for converting positions of an operator’s skeleton in three dimensions to robot motions. In this paper, we provide the algorithm and its implementation. Furthermore, we demonstrate the performance of the algorithm using the prototype robot.

Full paper : ISR-13-ArmMotion.pdf