Improved particle swarm optimization with reverse learning and neighbor adjustment for space surveillance network task scheduling

Space Surveillance Network (SSN) plays a crucial role in Space Domain Awareness (SDA) as it helps to maintain the catalog of Resident Space Objects (RSOs). Howe`ver, as the population of RSOs continues to grow, the contradiction between the increasing number of RSOs and limited sensor resources will be further apparent. How to effectively utilize existing sensor resources becomes a vital concern. This paper focuses on the Multi-sensor Collaborative Observation Scheduling (MCOS) in SSN. The workflow of SSN is summarized, and a Mathematical Programming Model (MPM) is established to maximize observation profit. Given the complexity of MPM, a modified Particle Swarm Optimization is proposed by combining Reverse learning, Neighbor adjustment, and Particle perturbation (RNP-PSO). In the MPM, multiple constraints such as observation time for each RSO, observation frequency, and sensor capability are considered, enabling a comprehensive description of the MCOS problem for the first time. The RNP-PSO incorporates a reverse learning algorithm to improve the quality of the initial population. Additionally, continuous mapping is added to the neighbor adjustment process to enhance the algorithm's search capabilities. Furthermore, a particle perturbation strategy is proposed to increase the diversity of solutions and avoid local optima. A series of experiments have been conducted to evaluate the performance of the RNP-PSO algorithm in solving the MCOS problem. The results demonstrate that compared to several other effective algorithms, the RNP-PSO algorithm significantly improves the Task Completion Rate (TCR) by approximately 17.01%, 19.1%, 13.53%, and 13.99%, while simultaneously reduces the Resource Consumption Rate (RCR) by about 55.48% and 36.38%. The experiments also include strategy and parameter sensitivity verification, which thoroughly validate the performance of the RNP-PSO algorithm in various aspects.