This week, Gatormatix has been focusing on getting ready to present for Prototype Inspection Day (PID) on Tuesday. Getting our hands on a drone body, LiDAR, and other hardware components has given us the ability to have a semi-functional working prototype to present. Preliminary testing has begun on each individual component, to ensure the team understands how those sub-systems work individually before trying to tie them all together.

One of the team members has made great strides testing a Visual SLAM algorithm with a single camera on Jetson Nano. Although a single camera was used in these tests, increasing the number of cameras will increase the accuracy of the data and hence, also the algorithm. Based on the knowledge that we will be able to use this flight computer for our project, we have gone ahead and ordered a Jetson Orin Nano for ourselves, our first official purchase! We are starting to move past the research stage and want to get some hands-on testing with the parts we will be using. We are aware that the task at hand is not simple, so preliminary testing and ordering parts early will be essential for us to meet key expectations by the deadlines. We are putting together the prototype over the next few days and recording videos to show at PID. The team is eager to keep working on the drone together and get it up and running!

This week, Gatormatix met and determined an improved plan for developing our sensing system. Our original vision was a little hazy and involved working on the sensing, mapping, and drone hardware one after the other, but upon reconsidering the needs of our sponsors, we have decided to focus our development around two parallel and independently-functional paths:
Developing the drone hardware itself and creating a handheld “rig” that will allow for fine tuning our sensor setup and mapping pipeline without needing to wait for the drone itself to be functional. Similarly, the drone prototype will be constructed with weight ranges in mind for the sensing components but will not have to wait for the rig to be completed.

By dividing the work along these boundaries, we hope to enable more rapid development and let our team members work within their specializations – our mechanical and aerospace engineers can focus on making sure the drone can easily fly and maneuver, the electrical and computer engineers can focus on interfacing the various components, and our computer scientists can select and refine our sensing algorithms. In this way, we hope to rapidly improve on our design and manage the complexity of integrating a project of this nature. We have updated our PDR to reflect this new approach and will present it for sponsor approval within the coming week.

Gatormatix has been working on their preliminary design architecture, making great strides towards a prototype goal in preparation for the Preliminary Design Review Peer review coming up next week. During the research process, the team was able to home in on some good solutions for systems they will need to design and integrate together onto the drone platform. Further research is required to understand how this integration will happen, especially depending on which hardware pieces we select.

They are committed to prototyping quickly, breaking things, and trying again, as it will provide them with a baseline platform to design and test their sensors on before next semester. This is most likely going to be the most challenging portion of their project. 3D modeling and navigation is no joke, and there are many small details such as safety and reliability of the base platform that need to be flushed out before a final design is created next semester. Back to work!

As our project gets more complex, we need to become better at managing our tasks and assignments. As such, the team has begun filling out tasks for each other, allowing them to become much more efficient and independent in their work. The team has developed an initial draft of the Preliminary Design Review, and has also been brainstorming about different technologies that can be used in the scope of UF regulations.
Looking forward, the team is planning to design the overall system architecture for their drone platform, describing parts and functionality, as it will allow them to break the project into multiple smaller achievable pieces. As the team is following an Agile development approach, they seek to having a working ‘prototype’ with added value at the end of each of their 2 week sprints. The team is very excited to begin work on this prototype to be showcased in a couple of months!

Week 4 featured another exciting meeting for the Gatormatix team and coach with Stephen and Ted Belser. The team has decided to pursue a LiDAR-based navigation approach, while also using it as the primary method to generate 3D models of an indoor scenario. In addition to this, cameras will be used to map real images onto a 3D model, making it look real. As the team continues to research, they will continue to look into ways of creating a primarily Made in America design, ensuring safety of critical components in the future platform.

The team completed the Product Design Specifications, which allows us to prioritize needs, set target values, and set technical performance measures that they hope to achieve. The team has begun working on the Project Roadmap and will be organizing tasks on Jira for week 5 to reflect these goals. In addition, the Meet the Team page is ready!

Gatormatix is born! We decided to name our team Gatormatix, combining ‘Gator’ for UF, with ‘Geomatics’, the study of geospatial data. It combines the two primary aspects of our identity as a team and our project. We finally had our first meeting with our liaisons, which was very productive, and we have gained good insight into the project we have ahead of us. In the coming days we are planning to look into the current available approaches to 3D modeling and compare their strategies with our use case and budget. For now, we will spend time brainstorming ideas on how to approach the problem and see what we can come up with for JWME.