Object Identification Under Known Dynamics: A PIRNN Approach for UAV Classification

Abstract--This work addresses object identification under known dynamics in unmanned aerial vehicle applications, where learning and classification are combined through a physics-informed residual neural network. The proposed framework leverages physics-informed learning for state mapping and state-derivative prediction, while a softmax layer enables multi-class confidence estimation. Quadcopter, fixed-wing, and helicopter aerial vehicles are considered as case studies. The results demonstrate high classification accuracy with reduced training time, offering a promising solution for system identification problems in domains where the underlying dynamics are well understood. I. INTRODUCTION The increasing deployment of unmanned aerial vehicles (UA Vs) in civilian and defense sectors has elevated the importance of dynamic modeling and intent inference for tasks such as control, classification, and anomaly detection. Traditional approaches to UA V identification rely primarily on visual, radio-frequency, or pattern-based features, which are vulnerable in contested or adversarial environments [1], [2].