[ ] Given positions of the other drone and poses of our drone, both following arbitrary paths, create artificial bounding boxes.
Face the "camera" towards an arbitrary point so that the other drone might suddenly appear in frame.
This is already done in compute_measurement in `uavf_2024/imaging/particle_filter.py
[ ] Take the bounding boxes, feed them into the particle filtering, and compute the loss between its output and the ground truth generated chosen above.
Description
Add tests in https://github.com/uci-uav-forge/uavf_2024/blob/drone-tracking/tests/imaging/drone_tracker_tests.py to test the Kalman filter prediction. We need to simulate the bounding boxes given arbitrary ground truth paths and test the particle filtering against it.
TODOs
compute_measurementin `uavf_2024/imaging/particle_filter.py