April 24 – April 28

I am back again and whole year older! This week I will be celebrating my birthday and the closing of this internship. This has been a great experience. I hope to continue learing about robotics as I matriculate through my final year at the University as well as in my future career.

This week we discussed the closing stages of the internship. Our report for CRA/CREU is due on  May 5th and we are continuing the revisions on the Swarmathon technical pages as we see fit. In short we will be continuing the testing process of our code through sims and physical application. This will in turn help the future interns in the next robotics team.

April 16 – April 22

I am finally giving another summary. This one is from “Cooperative Control Design for Uninhabited Air Vehicles (UAVs)”

The objective given in the abstract is that the research was geared towards the development and evaluation of the performance of strategies for cooperative control of autonomous air vehicles . As you can already tell  this is directly related to the research we are doing for the Swarmathon competition. As with the UAVs that were under observation in this research our robots are tasked with observing the environment they are put in through sensory.  As with all technology there are a number of problems that appear and a few solutions are given on how to optimize the results of the UAVs are optimal search theory, exhaustive geographic search , and derivative-free optimization methods, just to name a few. Finding out and understanding how these theories were used to help with the efficiency of the UAVs can help us with understanding how to fix and optimize the efficiency of our rovers. The paper went on to discuss advances in distributed computing and wireless communications which have enabled the design of distributed agent systems.

April 9 – April 15

Daphne gave us a few pointers for the upcoming Swarmathon competition. She informed us that aside from the physical competition, we will have different people asking us questions about our contributions, what we enjoyed or what was most challenging to us. I unfortunately will not be attending Swarmathon due to a few personal family issues and I do not want to miss another week of school knowing how difficult it was for me to catch up on work after the CIAA basketball tournament. The semester is coming to a close and it is imperative I remained focused for final exams. I know I am missing out on a great opportunity however I had to make the difficult decision to stay home and go to class.

April 2 – April 8

Daphne’s diligent work towards the composition of the Technical Report was superb and it has been successfully submitted! Prior to  submitting, the final copy of the report was given to every member of the team, as well as our advisor.  The team discussed their level of satisfaction towards the final product and came to a consensus that the report was satisfactory and that we were ready to continue progressing and move on to other objectives .

The team will now use this time to continue to revise our code and get it to the desired level of effectiveness by finding improvements to the search and detection algorithms. Localization issues are still at the forefront after ironing out the kinks the team’s goal is to get consistent test results and add our own flare to the existing algorithms.

March 27-March 31

This week the group finalized and submitted code ,March 28, 2017. We also did some last minute testing using sims and physical rovers. I helped run the sims trying to find why the rovers wold sometimes enter the Center. I noticed that rovers were also having difficulty seeing picking up cubes. I suspect that our devices do not have computational power or graphic capabilities to reflect real-time performance and more physical testing is needed to troubleshoot these issue.

Our attainable short term goal is to complete the paper and submit before the new deadline of April 4.

Some of our long term goals include producing code that will effectively search for, detect, identify, collect, and return target objects to a home base, aided by April Tags on the objects and at home base, improve code, and attend the Swarmathon competition. I will not be attending because I must focus on my studies at the university being that I already took a week off for the CIAA Tournament in February.

March 19 – March 25

We are back from the break and ready to spring into action.  I am unfortunately not able to participate in the High School Outreach due to the same issue that hindered me from attending the regular meetings. However I am up-to-date on the tasks that are being completed and discussed with in the group. My goals moving forward are to contribute as much as I can to the group as we move on to make submissions for our Code and reports. I will be reading papers that are directly related to our work with the robots and creating blogs about what I learned and how I can relate that to what the team is doing for the competition.

March 13- March 18

This week is spring break and I have been in Georgia visiting family and catching up on regular school work. With all that being said I was able to get in touch with one of the other team members and they provided me with these updates in regards to our progress as a team. Members of the team met to review robot behavior due to several inconsistent responses when collecting and returning cubes. It took some of the robots multiple attempts to pick up a cube and others had issues with the detection of the cubes period. Once collected, a rover not in visual range of “home” should make a calculation based on their perception of the coordinates and then begin the journey to home.  We tested several versions of code and make revisions as needed. According to the sims that are being ran, the issue odd robotic behavior is present is just a matter of capturing it on the sim.

March 5- March 11

As I have mentioned in multiple of my posts I am unable to attend regular meetings due to a mandatory lab I am taking for my AFROTC program. The way I was going to get back on track with the team and stay involved was to participate in the High School Outreach Program. This program was offered to high school students from Terry Sanford and Jack Britt, schools local to the Fayetteville area.  To prepare for the workshops we would host the team members were tasked to  complete NetLogo modules. I look forward to working with the students and continue my learning.

February 27- March 4

I’m back in action! As I mentioned in a previous blog I was tasked to read and summarize papers or articles relevant to the tasks and research the rest of the group is doing for competition. The first of these papers is Coverage For Robotics- A survey of recent results.

The purpose of this survey is to provide insight on several methods of coverage paths for robots. The methods are classified as either heuristic or complete. The survey was divided into four categories: heuristic, approximate, partial approximate, and exact cellular decomposition. The heuristic and randomized approach allows for the robot to chose its own path at random until it is able to retrieve an object. This method does not guarantee complete coverage of the given region  however it can be more cost effective because the robot does not need localization sensors. In addition, the robot does not need a prior knowledge of the given free space. The cellular decomposition methods all rely on a grid-based representation of the free space the robot is sweeping. In approximate cellular decomposition the cells sizes and shapes are identical however the combination of the cells is just an approximation of the target area. In semi-approximate cellular decomposition the cells are fixed in width however the ceiling and floor can have any shape. This algorithm calls for the robot to move in a zigzag motion and uses diversion inlets to cover any overlooked/undetected areas. Trapezoidal decomposition was the highlight of the exact decomposition section. This method allows for total coverage because of the simple back and forth motions the robot makes to visit the cells. The cells are often merged or split in order for the robot to continue performing these back and forth motions without intersecting an obstacle. Finally, multi- robot coverage is discussed. This task is comparable to that of animals. When ants find food they often leave a pheromone trail to communicate/alert other ants about where to find the given resource. Similarly robots are able to communicate and leave traces to perform tasks. This allows for the robots optimize the time it takes to cover the region of space given and reduces the dead-reckoning error because the robots use each other as beacons.

February 20- February 26

Hello everyone! This week I was able to travel to Charlotte, North Carolina and participate in the Central Intercollegiate Athletic Association (CIAA) Basketball Tournament as a cheerleader on the 2017 traveling team. The experience was great and I am so glad I was able to embark on this journey to the championship. Unfortunately we lost to Bowie State in the finals but the hard work and dedication each team put out on the court every day was worth the long nights and early mornings. As a cheerleader CIAA is one of the many things we look forward to as a closing to our cheerleading season (most of us are on the team from football season through basketball). Out of the 24 members of the Bronco Cheeleading team only 14 are able to participate during the Super Saturday Exhibition, a showcase of all 12 CIAA schools cheerleading squads, and I was lucky enough to be one of them. Here are some photos taken during the week . Enjoy !