Volumetric motion detection for multi-drone swarm engagement

Abstract

Coordinated drone swarms pose increasing threats to military installations and critical infrastructure, demanding sophisticated counter-UAS detection capabilities. Existing defense simulation tools face fundamental limitations in multi-target processing due to sequential architectures and inadequate 3D spatial modeling for volumetric threat scenarios. We implement an integrated simulation pipeline combining Bayesian occupancy grids, space carving reconstruction, and multi-view triangulation within a unified evaluation framework optimized for counter-UAS applications. The system employs CPU-based parallel processing, enabling deployment on standard military research platforms without specialized hardware requirements. Validation demonstrates that hybrid algorithm coordination maintains detection capability while achieving computational efficiency across diverse operational scenarios. The framework enables systematic evaluation of volumetric detection methods with comprehensive environmental modeling and reproducible protocols. This methodology advances defense simulation capabilities by providing accessible, rigorous evaluation tools for next-generation counter-UAS systems development and operational readiness assessment. Performance was validated across 50 Monte Carlo scenarios using an 8-camera visible-spectrum optical sensor array with bootstrap confidence intervals (n = 384 measurements) for all reported metrics.