Lightweight bounded-error propagation for resource-limited UAV cooperative positioning networks

The rapid proliferation of Unmanned Aerial Vehicles (UAVs) has revolutionized applications from infrastructure inspection to disaster response, where precise positioning forms the operational cornerstone. However, in GNSS-denied environments, cooperative positioning becomes essential, yet resource limitations of lightweight UAV platforms pose significant challenges. Conventional algorithms often require substantial computational power and communication bandwidth, exceeding the capabilities of such systems. To address these issues, this paper proposes a Lightweight Bounded-Error Cooperative Positioning (LBE-CP) algorithm for resource-limited UAV networks. By integrating information geometry with Control Lyapunov-Barrier Functions (CLF-CBF), the algorithm characterizes error propagation on Riemannian manifolds and establishes formal error bounds. The nonlinear error propagation model for cooperative nodes is derived, making the gradient of the manifold more accurate. Furthermore, the distributed CLF-CBF bounded-error framework ensures formal error bounds and system stability. The derived model guarantees that over 95% of positioning errors remain below the bound under certain conditions, with provable convergence properties. A fully distributed framework is developed, requiring only local communication and featuring incremental gradient updates on manifolds with sparse covariance propagation. Simulation results demonstrate that the proposed algorithm achieves superior positioning accuracy and reduced computational complexity compared to existing methods. The validation on the dataset also shows that the algorithm has excellent positioning performance.