uav

Drone/UAV path planning using Reinforcement Learning and Simulator

install and Setup

1. create data/ and model/ directory for storing datasets and model checkpoints.
2. pip install -r requirements.txt
3. Set environment variable: export AIRSIMHOST=<gcloud-external-ip>

Tensorboard for visualization

pip install tensorboard tensorboard --logdir data/<model_tensorboard>

usage

to use the drone object:

from sim.conn import client
client.move...

General info

airsim uses NED coords: +X North, +Y EAST, +Z DOWN