Team Assessments

This semester the team has mostly been focused on trying and eliminating different processes, technologies, and software before committing to one idea next semester. In terms of hardware the team has prepared for two different scenarios. A hardware supply list was made in the case that the group succeeds in receiving funding from L3Harris. This includes multiple 3D printing filament spools of different thermoplastics, differing cameras of different frame rates, FOVs, and supported resolutions, and sensors such as lidars and wheel encoders. These will allow us the freedom to experiment with differing hardware to see what the most optimal and price efficient kit we can put together is. We have also prepared for the worst case scenario, which is supplies needed to fulfill our MVP goals in the chance that we do not receive additional funding. This includes a scale backed amount of electronics of lower quality.

In terms of software the team has made great strides in narrowing our focus. We have decided on using ROS 2 and a focusing on a decentralized swarm environment rather than a centralized ROS 1 swarm. ROS 2 is poorly documented and will take a long time to build out the proper communications for our package. Thus, the team has decided to focus on making our final software product a ROS 2 package that will implement a model-based learning approach for the demonstration rather than an instance-based learning approach. This will allow us to focus on primarily building tools out for other developers to use rather than wasting time learning machine techniques. We want to focus on robotics and computer engineering, not other fields. A sample of sending a message between two Raspberry Pis on the same network was achieved during the semester using ROS 2. Additionally, we settled on using Docker to speed up set up time of additional testing robots since all system packages will be included in the image.