Demo Scenarios

In this repository, we provide illustrative demos of both discrete (binary) and continuous target variable terrain monitoring in the Gazebo-based RotorS environment. The simulation set-up includes a rotorcraft-type UAV equipped with a downward-facing camera (AscTec Firefly), as per the algorithms' problem formulation.

---

Simulation environments<

tmplanner_discrete

For discrete variable mapping, we provide a 4x4m demo environments using AprilTags on the ground to represent "targets". The target occupancy is mapped using the tag pose detections output from the AprilTag detector provided by the apriltags_ros package. To change the target feature for mapping, please modify the updateMapFromPoseArray function in grid_map.cpp accordingly, or create your own map update function following this skeleton.

Note: For now, tmplanner_discrete is only capable of mapping a binary variable (e.g. AprilTag presence) using an occupancy map-based representation. In the future, we plan to extend the software to handle multi-label classifications, and add image data inputs.

tmplanner_continuous

For continuous variable mapping, we provide a 30x30m demo environment with painted green blobs on an a (grey) ground surface. The hue level is mapped using the images received from the on-board camera. To change the target feature for mapping, please modify the updateMapFromImage function in grid_map.cpp accordingly.

---

Instructions

The following steps outline the terminal commands to run an example for mapping a continuous variable. The same procedure can be applied to map discrete variables, by replacing for the tmplanner_discrete package name in the commands.

1. Start the simulation environment by running:

$ roslaunch tmplanner_continuous monitoring_example.launch

2. Start the planner node:

$ roslaunch tmplanner_continuous tmplanner.launch

Note: The appropriate parameters must be loaded in the launch file.

3. Initialize the planning routine via the rosservice:

$ rosservice call /firefly/start_planning

You should now be able to see the UAV moving in the Gazebo window as the finite-horizon planning routine is executed. The planned paths can be visualized using the ROS package rviz by following the instructions here.