Onboard Reconfigurable Holographic Surface-Aided Simultaneous Localization, Mapping, and Moving Object Tracking for Low-Altitude Wireless Networks
Abstract
Emerging to support the burgeoning low-altitude economy, the low-altitude wireless network (LAWN) serves as a mobile infrastructure with unified connectivity and sensing capabilities. As the primary enablers of this network, unmanned aerial vehicles (UAVs) are required to go beyond self-localization to perceive surrounding obstacles and track dynamic targets in real-time. While simultaneous localization, mapping and moving object tracking (SLAMMOT) is well-suited for these tasks, traditional systems typically rely on narrow beams with limited coverage, often leading to tracking failures. This paper presents a UAV-based SLAMMOT system enabled by onboard reconfigurable holographic surfaces (RHSs) whose effective aperture is reconfigurable. Given the stringent energy constraint of UAVs, an inherent resource trade-off exists between tracking and mapping beams. To address this, we formulate a joint optimization problem aiming to enhance both tracking and mapping performance. Subsequently, the fractional programming (FP) method and the successive convex approximation (SCA) algorithm are employed to optimize power allocation, RHS effective aperture, and RHS beamformer. We derive the optimal solution for power allocation, characterize the Pareto boundary of tracking and mapping performance, and analyze the algorithm’s complexity. Simulation results demonstrate that compared to benchmark schemes, the proposed SLAMMOT system can enhance tracking accuracy and achieves a favorable trade-off between mapping and tracking performance.