Flying Predetermined Course in GPS-Denied Environments

todd45040

We just received our new M500 and are trying to get up to speed on using the platform. I would like to have the M500 autonomously fly a simple predetermined rectangular course 6'x30' and do some simple (single item) object detection of something like rubber duckies along the way. My thought is to physically model the course and then use the flight logs to setup up and new, autonomous mission plan and complete the course searching for the objects.

Does this approach seem reasonable, or is there a better way to approach the task?

Thanks,

-Todd

modaltb

Hey @todd45040 ,

Sorry for the lag, and you may be past this stage by now, but yes that looks decent!

I'm myself not so involved on the object detection front but can comment first on the navigation part. Here's a simple example using GSP denied nav (VIO via IMU and tracking sensor) with python in a docker container on the VOXL

--> https://gitlab.com/voxl-public/voxl-docker-images/voxl-docker-mavsdk-python/-/blob/master/offboard_position_ned2.py#L29

We're basically setting waypoints in X,Y,Z, which could be a good starting point.

todd45040

Thanks for the info., it's not too late. We will work the task in the upcoming week.

We'll make a few attempts and report back our outcome.

Thanks again,

-Todd