Explainable Adaptive Path Reconfiguration for Obstacle Avoidance in UAV Networks Using Transformer-Based Predictive Modeling and Bio-Inspired Optimization

Critical challenges for Unmanned Aerial Vehicle (UAV) networks in FANETs operation include collision avoidance, energy efficiency, and decision transparency in highly dynamic aerial conditions. This paper proposes an Explainable Adaptive Path Reconfiguration framework by integrating Transformer-based trajectory prediction, Secretary Bird Optimization Algorithm (SBOA), and SHAP-driven interpretability, enabling predictive, self-aware UAV navigation. The Transformer model predicts future UAV and obstacle trajectories under multi-head self-attention to proactively reconfigure before any potential collision risk. Then, the SBOA further optimizes the trajectories by minimizing a multi-objective cost function that jointly considers collision risk, energy consumption, and path deviation. It estimates the quantitative contribution of key features distance, velocity, and orientation-to each decision made, guaranteeing human-interpretable autonomy. Extensive simulations for diverse UAV densities and obstacle environments prove that EAPR achieves 96.2% collision avoidance, 82.8% energy efficiency, and 0.128 path deviation with improvements of 15%~25% over PSO, FHOA, and PSOA. Robustness tests further confirm resilience in case of communication loss and dense obstacle scenarios. This paper establishes a unified paradigm of predictive, optimized, and explainable UAV swarm navigation for safe and transparent autonomous operations in next-generation aerial networks.