After the team had a working setup, we approached a problem.
Picked up by the team as a PS that would be done and over with within 6 days, we spent 1.5 months on it, with 20 days straight day-night. Our codebase lies here. We spent first working on a simulation ground, preparing that in order to “be able to tackle” the original problem, but in mid of that ditched the idea.
Moving back to actually building a drone, making it convenient enough for regular flights, including Autopilot handling, then mounting an Odroid XU4 for inducing autonomy, took us 1 month. Then back to the PS, we made sensor handling modules, position control module, and a flight planner as an FSM to detect an object, align with it, pickup with an electromagnet, go to the dropzone, turn off the electromagnet, repeat once more, and land back. We were unable to fully complete the PS due to barometer of our FC glitching out, but the whole pipeline creation was worth the huge time investment.
project
iitk
hardware
aerial-robotics