Code and data for "A Unified Receding-Horizon Controller for Safe Multi-Robot Search and Rescue, Pursuit, and Perimeter Defense"

<p dir="ltr">Companion package of the IEEE Access article "A Unified Receding-Horizon Controller for Safe Multi-Robot Search and Rescue, Pursuit, and Perimeter Defense".</p><p dir="ltr"><br></p><p dir="ltr">It contains (1) the MATLAB R2025b simulator of the three tasks (cooperative pursuit, perimeter defense, search and rescue), the receding-horizon controller and its coordination layer, the adversary models, the baselines (PSO, DE, GWO, L-SHADE, MPPI, APF, DWA), the random-layout generator, the continuous-time contact check, the experiment definitions and the resumable batch harness; (2) the per-episode results of every run reported in the article, one CSV per experiment; (3) Python scripts that recompute the tables of the Results section from those results; and (4) the document, dated and hashed by the authors before the runs that test them, that fixed the six confirmatory hypotheses (H1-H6), with its SHA-256.</p><p dir="ltr"><br></p><p dir="ltr">The implementation of the SAM-SOMA solver is not redistributed; Appendix A of the article specifies it completely.

Every SAM-SOMA run is included as per-episode results, so the statistics can be recomputed without it. README.md describes the folders and the experiment modes.</p>