This is a repository for OpenAI gym environment. The general task is navigation of MAV with limited field of view in unknown and dynamic environment. The accompanying paper can be reviewed here.
To run an experiment, an example command is listed below:
python main.py --gaze_method Oxford --planner Primitive --agent_number 10 --agent_max_speed 20 --agent_radius 15 --drone_max_speed 40 --map_id 1
There are many parameters that can be pre-defined. By running python main.py -h
, a list of all parameters can be seen.
To generate a new map with pre-defined difficulty level, an example command is listed below:
python ./script/fit.py
An GUI will be shown where you can define the survivability difficulty level and generate a map accordingly. The number, radius, and speed of the dynamic obstacles can also be optionally pre-defined.
Another repository is used to calculate the metric in Gazebo. The repository can be found here.