Multi-UAV Cooperative Observation and Tracking Method Driven by Heterogeneous Sensors

When conducting cooperative tracking of ground moving targets from beyond the defensive zone, unmanned aerial vehicle swarms face multiple challenges, including target state uncertainty, strict safety distance constraints, and difficulties in heterogeneous sensor coordination. To address these issues, a multi-unmanned aerial vehicle cooperative tracking method that integrates receding horizon control (RHC) with heuristic optimization was proposed. By analytically characterizing the measurement properties of heterogeneous sensors, the overall Fisher Information Matrix of the multi-unmanned aerial vehicle system was systematically derived, and the optimal observation configuration for a heterogeneous-sensor multi-unmanned aerial vehicle system was analytically obtained. On this basis, a high-dimensional non-convex trajectory optimization problem was formulated with the objective of maximizing the determinant of the cumulative Fisher information matrix over the prediction horizon. An improved snake optimizer was introduced as the solver for the receding horizon control framework, incorporating both encircling consistency maintenance and smoothing mechanisms, thereby enabling autonomous guidance of the heterogeneous unmanned aerial vehicle swarm to form and maintain a near-optimal observation configuration. Simulation results demonstrate that, compared with the Lyapunov Vector Field method, the proposed approach significantly enhances overall positioning accuracy and observation robustness in tracking ground moving targets.