Hybrid UAV Data
Description
This dataset is generated from 60 Monte Carlo trials, each consisting of 370 sampling points with a sampling interval of 0.1 seconds. The true target trajectory follows a sequential pattern of motion modes: Constant Acceleration (CA) → Counter‑Clockwise Turn (CT1) → Constant Velocity (CV) → Clockwise Turn (CT2) → Constant Velocity (CV), with random perturbations at the switching points. The dataset stores three‑dimensional arrays: true_data (true x‑y positions), measurement_data (noisy 7‑dimensional observations including position, velocity, acceleration, and turn rate), and mode_data (labels corresponding to the four motion modes). It is intended for training a discriminative network to recognise different motion patterns and to fuse predictions from individual models (CV, CA, CT1, CT2) for improved target tracking performance.