AuMAV is an entry-point into a set of techniques and packages used and/or implemented by the drone lab at CAS/RPL.
In its most basic form it is a ROS launch file with which a drone can be made to run autonomously in different configurations.
Planned features:
- Support for multiple localization and mapping solutions
- HectorSLAM for LiDAR-based SLAM
- ORB-SLAM2 for visual SLAM
- Collision avoidance pass-through controller
- Extensible Python state machine for high-level planning
- Recovery from tracking errors
- Autonomous exploration capabilites
Please refer to COROUTINES.md
for an explanation of how planning is handled.