ROS Start Robot

Through out my early years in undergrad, I had struggled with the time of planning out a project. My idea was to design a robot with camera vision and remote control. I spent the spring semster planning out the funds I would need to allocate in order to buy parts for a ROS robot. Then the pandemic hit. I was stuck in my house. Most companies had long shipping times, and it became an unreliable and costly resource. Using left over robot kits, a raspberry pi, and a PS3 camera I had at home, I went ahead with my project. I lacked the resources to buy drill bits to make host for a two level chassis, so I used a box to store components.

I’ll admit the chassis itself is not very good, but for the sake of just making a start project, it worked fine for the first iteration. I spent time that I wasn’t working at Indus Manufacturing working on reading ROS documentation online, and connecting with engineers at Canonical and ROS to seek feedback on my project. Being my first robot project using ROS, almost ever issue I encountered involved using ROS answer forums to solve issues. Being able to to receive feedback from the engineers also helped to accelerate the pace of the project, leading me to be able to have the robot run autonomously in a week.

By week two, I had already figured out how to write my own packages for ROS. This was an essential step in the completition of the project. At the time, the nodes required for the project we’re not up to date for the new ROS firmware, so i had to build these old nodes from source. Acquiring this new knowledge enabled me to figure out using the controller with the robot. Then just by curiosity, I wondered if the same could be done for camera vision. Camera vision proved to be even simpler, as I could use my laptop as a viewing point for the camera on the robot.

This project taught me the importance of ignoring the fear of not knowing, and the joy of discovering unexpected answers. Had I not taken a chance, I would have probably convince myself that I could make something like this in the future, without actually making it. I also used this experience to document my work for the first time, and published my documentation on the popular tinkering website hackster.io. I wanted to be able to look back and see my thinking process at the time, and also look for feedback on how to improve my flow of work from the internet.