UAV-Coverage

Simulations for planar area coverage by a swarm of UAVs equipped with downwards facing cameras.

Usage

Included are simulations taking into account different sensing patterns, coverage quality functions and region importance functions, each one being a different .m file. The initial network configuration and all simulation parameters are set inside these files which contain comments on their exact usage. The following simulations are included:

• UAV_coverage_uniform_quality.m simulates a network with circular sensing patterns and uniform coverage quality on a region whose points have equal importance.
• UAV_coverage_decreasing_quality.m simulates a network with with circular sensing patterns and decreasing coverage quality (paraboloid) on a region whose points have equal importance.
• UAV_coverage_decreasing_quality.m simulates a network with circular sensing patterns and uniform coverage quality on a region whose points are assigned importance weights by a function phi.
• UAV_coverage_anisotropic.m simulates a network with with elliptical sensing patterns and uniform coverage quality on a region whose points have equal importance.

Included are also some .m files used for plotting the results of the aforementioned simulations, namely FIGURE_simulation_results.m, FIGURE_simulation_comparison.m and FIGURE_simulation_results_anisotropic.m.

Each simulation is structured as follows:

Initialize region
Initialize network
Initialize simulation
while simulation is running {
    for each agent {
        Create sensed region
        Calculate coverage quality
    }
    for each agent {
        Calculate its assigned sensed region
    }
    Calculate the value of the optimization objective H
    Plot the network state
    for each agent {
        Calculate the control inputs
        Simulate movement with an ODE solver
    }
}
Save results to timestamped .mat file

Screenshots

The following screenshots are from the UAV_coverage_uniform_quality.m simulation.

The following screenshots are from the UAV_coverage_anisotropic.m simulation.

References

[1] Sotiris Papatheodorou, Anthony Tzes, Yiannis Stergiopoulos, Collaborative Visual Area Coverage, Robotics and Autonomous Systems, ISSN 0921-8890, Elsevier [Accepted for publication]

[2] Sotiris Papatheodorou, Anthony Tzes, Cooperative Visual Convex Area Coverage using a Tessellation-free strategy, In Proceedings of the 56th IEEE Conference on Decision and Control (CDC) 2017, 12-15 December 2017, Melbourne, Australia [Review Pending]

License

Distributed under the Apache License Version 2.0.
Copyright © 2016 Sotiris Papatheodorou