To achieve future autonomous mobility for Unmanned Aerial Vehicles (UAVs), reliable runtime trust assurance is crucial. This paper presents a trust-assurance method for autonomous drones utilizing runtime compliance checking via a Digital Twin (DT). The DT incorporates drone-specific metrics from AirSim simulations, including sensors’ health and network centrality for real-time behavior analysis. Drones collaborate in swarms, sharing predictive and actual behaviors for trust assessment. The model uses Random Forest (RF), Support Vector Regression (SVR), and Convolutional Neural Network (CNN) to estimate swarm coordination rates as trust indicators, with each iteration classifying drones as Trusted or Malicious. The model was tested in an autonomous drone delivery system against various trust-related attacks, demonstrating SVR’s effectiveness in estimating coordination rates and RF and SVM’s role in trust classification.