BUILD A STORY - Peter Camamis 

Faculty Mentor: Dr. Daniela Rosca

ABSTRACT

As more and more parents struggle to provide for their children, they frequently take on jobs and cannot always be around for their children. Build a Story was created to help connect remote parents with their young ones. In addition to providing an application to author interactive bedtime stories, the results may be shared remotely with the community of parents, including interactive feedback to give the author a sense of satisfaction. Authors may build their stories including associated media capable of attaching to words during composition. Formal documentations and methods have been processed including task estimations and wireframes. The application is written in Swift 2.2 for iOS 8+ and in particular for iPad tablet computers. It uses Amazon Web Services for network capabilities. Amazon services used are Cognito for authentication, DynamoDB for remote data storage, S3 for remote associated media, Route 53 for internet infrastructure and SNS for mobile notifications. Service analytics are powered by Crashlytics providing crash reporting, mobile analytics, mobile identity as well as growth, monetization, infrastructure and UX analysis.

SWARM SIMULATION - Philip DiMarco

Faculty Mentor: Dr. William Tepfenhart

ABSTRACT

From birds, and fish to ants and bees, swarming plays a critical role in how animals interact and live. The emergent behavior that results from swarming allows complex problems to be solved without any centralized control structure. Swarming, as it applies to control of groups of robots is particularly exciting.  This research focused on the simulation of multi agent systems, where the movement of each agent is determined by the surrounding local environment. Stated in a more concrete way, we can think about this problem space as consisting of a group of lawn mowers trying to cut the entire area of a lawn, where each lawn mower is locally attracted to high grass, and its movement is determined by the relative positions of all surrounding lawn mowers.  The behaviors of these lawn mowers is designed to mimic the behavior of herds of sheep.

Each swarm member operates on the following principles:

1. Every agent can sense the relative distance and angle of all other agents around it.

2. Agents are able to coordinate their movement through the use of an attraction/repulsion force (attraction over far distances, repulsion over short distances, 0 at some equilibrium point).

3. Agents are attracted to high grass.

The combination of attraction/repulsion forces and attraction to high grass allow the agents to move as a swarm and cut the entire area of a lawn.

The research performed here examines various functional mechanisms for computing attractive and repulsive forces, the resultant behaviors dictated by how forces are computed, and the efficiency of the swarm in mowing a lawn.

XERCISE FITNESS - Nico Flora, Kyle Blazier, Ryan Aucone, Ryan O’Rourke and Yaqoub Alshatti

Faculty Mentor: Dr. Daniela Rosca

ABSTRACT

Xercise Fitness is a personalized fitness iOS application offering users a fully immersive personal health experience. There are two main focuses for the Xercise Fitness application, and they involve exercises/workouts and macronutrients. The main feature of the application consists of the ability to randomly generate a single exercise or entire workout for the user based on two simple criteria, a main muscle group and a sub muscle group. Personalization is another key feature for Xercise Fitness, so users have the ability to create, save, and share their own personal exercises/workouts or those created by other users. Finally, users can track their own macronutrients including carbohydrates, fats, and proteins. They are also provided with the ability to set a goal, track a running total, and see their progress for the day compared to the goal they set for themselves. The objective was to target a wide range of users, those who may need a jumpstart in the gym, those who may be looking for something new in their workouts, or those who want to track a healthy lifestyle. The team followed agile software development practices in order to achieve the goals set for this project.  Through agile development, the team incremented work through iterative sprints of planned work, which differs from the traditional functions of waterfall development.  The group developed requirements specifications, detailed mock designs, a project management plan, and test documentation in order to facilitate the success of the application while in development.  Thus far, our results from Xercise Fitness have directly correlated to the goals we set out to achieve. Making minor adjustments during the course of development, we never once steered away from the achievable goals that we set out to do in the start of the project.

AUTOMATED REPEATABLE RELEASABLE HOLDBACK BAR (AARHB) TEST FIXTURE IMMERSIVE UNITY OPERATIONAL/MAINTENANCE TRAINER - Jason Gaglione, Ryan Curto, Rich Allindogan and Christopher Mott

Faculty Mentor: Dr. Daniela Rosca

ABSTRACT

The project sponsor, supporter, and acquirer is the United States Navy.  Prospective users will be United States Navy trainees. The purpose of the Automated Repeatable Releasable Holdback Bar (AARHB) Unity 3D project is to simulate the process of operating the AARHB in a virtual environment to serve as a virtual reality training simulator. We have created a simulation of the computer graphical user interface (GUI) to further simulate the actions that are done in the AARHB laboratory. The 3D simulation begins with the trainee being dropped into the AARHB testing fixture laboratory with the ability to walk around freely. The user will be able to walk up to the testing fixture, computer or any walls within the laboratory. During the Fall 2015 semester our team worked on documentation such as the Software Requirements Specification (SRS), Software Architecture Document (SAD) and Software Project Plan (SPP). We are in constant communication with the team (our client) at the naval base, making sure to keep them in the loop and understanding where we stand in the development of the simulation. During the Spring 2016 semester our team is participating in an agile process for our development phase. 

Our agile process is broken down into two week sprints in which we dedicated each sprint to a specific aspect of the project. We planned out five two-week sprints which allows the whole team to look forward to a finished simulation on April 9th, 2016. We also use a System/Subsystem Design Description (SSDD) document to help organize our sprints and really understand what we wanted our end goal of each sprint to be.

ENHANCING NAMED-ENTITY RECOGNITION WITH DBpedia - Ramya Kalathingal and Maurani Saha

Faculty Mentor: Dr. Richard Scherl

ABSTRACT

We are using a state-of-the-art parser to process natural language texts such as news articles. The parser uses a named-entity recognizer to identify phrases as names of people, organizations, and locations. DBpedia is a semantically organized database that is automatically created from the structured content of Wikipedia. In accordance with the principles of the Semantic Web, DBpedia has an ontology that further categorizes people, organizations, and locations. By accessing the latest archives of DBpedia, we are experimenting with matching the named entities found in the text with those categorized by DBpedia to obtain a finer-grained categorization. For example, this technique allows us to identify people as scientists or politicians, and organizations as companies or universities. Our goal is to use this information to improve the performance of text clustering and classification methods.

A ROBOT TO IMPROVE SOCIAL SKILLS OF CHILDREN WITH AUTISM - Ramya Kalathingal, Maurani Saha, and Himabindu Yalamanchili, Teena Gopi

and Swethana Gopisetti

Faculty Mentors: Dr. Patrizia Bonaventura and Dr. Richard Scherl

ABSTRACT

The presentation describes the first stage of a proposed research project on testing a zoomorphic robot as a therapy tool for children with autism.  The diagnosis of Autism Spectrum Disorders (ASD) is currently on the rise, with 1 in 68 children currently affected by this disorder. ASD affects social communication and causes lack of appropriate social interaction of the children with their peers and caregivers, inability to decode and express emotions, lack of ability to communicate verbally, and aggressive behavior.

Interaction with robots has been shown to improve communication skills by children with autism, but issues related to ethical and social acceptability of these technological tools are still debated. Previous research on the most acceptable appearance of the robot shows that zoomorphic robots elicit better social interaction of the children with autism than human-like robots.

However, there are very few studies that have tested the robots with real ASD patients. The present exploratory research, which has not entered data collection, aims to bridge this gap and test the effect of a zoomorphic robot on verbal and non-verbal interaction with children with autism.

The poster reports the first stage of the research: the construction of a relatively simple robot using the inexpensive EV3 platform. The robot is programmed to move up to the child and say several sentences.  An infrared sensor is used to calculate the exact position of the child, as the robot moves towards the child. The robot starts talking when the sensor indicates that the child is within a distance threshold.

The data to be gathered concerning the children’s interaction with the robot will be important for increasing our understanding of ASD and developing better therapies. Future work will involve recording and analyzing the children’s speech in response to the robot, and using a more sophisticated robot.

T-TRAIN: A TYPING PLATFORM COOPERATED WITH MULTI-DISCIPLINES - Lotachukwu E. Onwumelu and Yuxi Yang

Faculty Mentor: Dr. Cui Yu

ABSTRACT

T-Train is a special typing platform that can be cooperated with multiple disciplines in the university. It provides functions to train typing, practice spelling, simulate flash-card learning, save and share writing, and even support cognitive exercises that involve reading (listening), understanding, remembering and repeating. It is a flexible platform that users can choose how to use for various needs. This project is an example demonstrating how a computer system can exploit simple tools and empower them with approaches and findings from multiple disciplines to become more useful. Take typing tools as the example; there are many typing programs available, but they mostly focus on typing training for positioning, accuracy and speed. In this project, we experiment to add elements from literature language (English) education, engaging-learning, social connection, cognitive neurology theory, and special needs therapy practices. This is the first version of implementation aiming to prove the concept. Future improvement will be forthcoming.

FEED - Justin Schlemm, Abdul Muhsin J. Al-Kandari, Taylor Klodowski and Mary F. Menges

Faculty Mentor: Dr. Daniela Rosca

ABSTRACT

The FEED system is a web application for gathering, organizing, and displaying of interests and ideas. Although a multitude of social media outlets exist that allow users to sort and share content, the goal of this application is to give them a private place to organize their many interests and ideas. The currently available systems can be used to store and organize content, but they often limit the type of content that users can store, lack multiple options for organization, and are based on social media concepts. The FEED system will match what we feel is an area where current software falls short.

Modern computer power users are assaulted with a variety of content (web pages, tweets, things to take note of, photos, videos, lyrics, documents, etc.). Users need to be able to return to information which they have saved in a quick and reliable manner. To solve this issue, we are proposing a system that:

• Saves all these types of items as well as associated metadata

• Provides natural methods of organizing and searching through these items

FEED lets users access their stored items across multiple views including a timeline view and a mind map view. While the timeline view helps users to quickly find items based on the time they were saved, the mind map view allows uses to organize items in a hierarchical format which is particularly useful for brainstorming new ideas.

The FEED system was developed at Monmouth University as a Software Engineering Practicum project during the fall 2015 and spring 2016 semesters. The system runs on an nginx/unicorn/postgres stack, using ansible to manage development, testing, staging, and deployment. The application was written in Ruby on Rails with a Bootstrap and AngularJS user interface. Agile development strategies and test-driven development methods were used throughout the project.