Effective situation awareness relies on the robust processing of high-dimensional data streams generated by onboard sensors. However, the application of deep generative models to extract features from complex UAV sensor data (e.g., GPS, IMU, and radar feeds) faces two fundamental challenges: critical training instability and the difficulty of representing multi-modal distributions inherent in dynamic flight maneuvers. To address this, this paper proposes a novel unsupervised sensor data processing framework to overcome these issues. Our core innovation is a deep generative model, VAE-WRBM-MDN, specifically engineered for stable feature extraction from non-linear time-series sensor data. We demonstrate that while standard Variational Autoencoders (VAEs) often struggle to converge on this task, our introduction of Weighted-uncertainty Restricted Boltzmann Machines (WRBM) for layer-wise pre-training ensures stable learning. Furthermore, the integration of a Mixture Density Network (MDN) enables the decoder to accurately reconstruct the complex, multi-modal conditional distributions of sensor readings. Comparative experiments validate our approach, achieving 95.69% classification accuracy in identifying situational patterns. The results confirm that our framework provides robust enabling technology for real-time intelligent sensing and raw data interpretation in autonomous systems.
This paper addresses the formation assignment problem for heterogeneous UAV swarms in dynamic mission environments. A constrained optimization model is constructed in which UAV capabilities are divided into shareable and exclusive types; a neighborhood collaboration decay factor captures the locality of capability complementarity; and a Cobb–Douglas production function evaluates position-specific effectiveness under bottleneck constraints. The objective dynamically trades off deployment costs and system risks through threat-adaptive weight adjustment. To solve the model, a Hybrid Adaptive Large Neighborhood Search (HALNS) algorithm is proposed, integrating an adaptive destroy-repair mechanism, a mathematical-programming-based local search, and an incremental re-optimization strategy for rapid dynamic response. Experiments verify that HALNS attains globally optimal solutions on small-scale instances and outperforms mainstream baselines on medium-to-large problems. The collaboration mechanism raises system effectiveness by an average of 34.75% across four mission scenarios. Compared with static re-optimization, the incremental strategy improves dynamic response performance by 58.25% while reducing runtime by up to 56.7%. Sensitivity analyses confirm the robustness of key parameters. This work provides a theoretical and algorithmic foundation for intelligent UAV swarm assignment and reconfiguration.