I’m Abhyudit, a Masters student in Mechanical Engineering. I worked on the conductor, arm controller, drive controller, and path planner modules, and was also responsible for implementing the A* planning algorithm. For me, the most difficult part was transferring results (like moving the arm and planning collision-free paths) from simulation to the actual locobot. However, this project gave me a great opportunity to learn more about ROS, MoveIt, and work with real robotic systems.

I’m Bianca, a co-term student in Bioengineering. For this project, I worked on our initial testing with the physical locobot, writing a single launch file that could launch either the simulation or real robot, and multi-robot simulation. The main challenge I faced was the multi-robot simulation, a task that seems straightforward but proved to be more complicated under the hood. I attempted to reverse-engineer example code from interbotix, however each file referenced another file which referenced another file… and despite my best efforts (and TA efforts) to adapt that code for our own project, there was not enough time to fully functionalize multi-robot simulation. However, through working through these challenges I gained a deeper appreciation for the complexity behind the ROS network and the information flow that must be orchestrated in order to simulate robot perception and motion.

I’m Connor, a co-term in Mechanical Engineering. I worked on designing, implementing, and troubleshooting many of the modules including the conductor, occupancy grid, station tracker, and drive controller. There were many challenges developing such an integrated system of moving parts from scratch, but the hardest part for me was troubleshooting the Interbotix packages to get the system functioning on the real robot. This project taught me about integration testing, creating robust autonomous systems, and designing algorithms that intuit and act on the belief states of collaborators. 

My name is Mateo and I’m a co-term student in Electrical Engineering. I was responsible for the unit testing module and the Rapidly-exploring Random Tree path planning algorithm. I also set up launch files with the appropriate Locobot flags. In general, most time was spent troubleshooting technical issues with the physical Locobot over the last few weeks of the quarter. I learned a lot about ROS topics, path planning algorithms, and odometry pose transforms.