I’m working with a team on how to equip quad-copters for indoor navigation. The application is STEM students exploring robotics.
The hope is that a multi copter can be equipped for basic navigation without the addition of optical sensors.
One idea is to see if the copter’s lateral movement could be “calibrated” so that the amount of motion (distance travelled) could be estimated. With such an calibration, then perhaps a LUA script could be written to instruct the copter to navigate a proscribed path.
The idea is that if a small pitch or roll change is introduced to a hovering copter (at a constant altitude) and the distance traversed were noted - a calibration might be achieved.
I would not expect that we’re the first people to think of this idea - so if anyone knows of work done in this area, we’d greatly appreciate any suggested references that we could learn from.
I have read that recently. I’d like to see what can be done without optical sensors.
I’m looking to take more of a dead-reckoning approach. I acknowledge that there will be limited accuracy.
We want STEM students to be able to program simple navigational task - such as move 5 feet left, 10 fee forward, and then 2 feet to the right.
We’re building inexpensive quad-copters for these programs - that have only the basic equipment. Adding optical sensors would make this equipment too expensive.
Then use search here on the forum for 'dead reckoning". You will find some comprehensive information that generally concludes modest at best results with a Flight Controller that has a ~$400 IMU
So, get yourself a CUAV X7-Pro or a Qio-Tek H743 Aero and put those kids to work.