Personal Implementation of Learning Drone Race with Interactive Learning
[TODO]
Code is adopted from: https://zenodo.org/records/7955278
Kaufmann, E., Bauersfeld, L., Loquercio, A., Müller, M., Koltun, V. and Scaramuzza, D., 2023. Champion-level drone racing using deep reinforcement learning. Nature, 620(7976), pp.982-987.
This pseudocode accompanies the manuscript. It consists of three parts:
drift_estimator: pseudocode for estimating VIO drift from gate detections and VIO readings.gate_detector: pseudocode for gate detection from images recorded from the onboard camera.swift_policy: pseudocode for policy training and residual model identification.