【作者】 李娜； 张波； 姬广淼；

【Author】 LI Na;ZHANG Bo;JI Guangmiao;School of Electronic and Information Engineering, Beihang University;

【机构】 北京航空航天大学电子信息工程学院；

【摘要】 中地球轨道搜救(mediumearthorbitsearchandrescue,MEOSAR)系统中的到达时间差(time difference of arrival, TDOA)定位是一个典型的非线性优化问题,粒子群算法(particle swarm optimization,PSO)由于具有全局搜索、不依赖梯度、简单易实现等特性在处理这类问题时往往很高效。但卫星被动定位问题受限于地球表面约束条件,使用经典PSO算法会产生无效解,因此提出一种改进PSO算法,将优化目标从笛卡尔坐标系转换到空间大地坐标系,利用约束条件限制粒子搜索空间,从而将约束问题转为非约束问题。仿真结果表明,相较于标准PSO算法、海鸥算法(seagull optimization algorithm, SOA)和鲸鱼优化算法(whale optimization algorithm, WOA),改进的PSO算法在定位误差上分别提升了72.5%、61.9%和40.7%,收敛精度较SOA和WOA分别提升83.8%、38.2%,验证了该算法在MEOSAR系统的TDOA定位问题中的有效性。更多还原

【Abstract】 Time difference of arrival(TDOA) positioning in the medium earth orbit search and rescue(MEOSAR)system is a typical nonlinear optimization problem. The particle swarm optimization(PSO) algorithm is effective in addressing such problems due to its global search capability, gradient-free nature, and implementation simplicity.However, satellite passive positioning is constrained by the Earth’s surface condition, causing classical PSO to yield invalid solutions. To address this, an improved PSO algorithm is proposed in this paper. The optimization framework is transformed from Cartesian coordinates to the geodetic coordinate system. This conversion inherently restricts the particle search space using the constraint condition, thereby converting the constrained problem into an unconstrained one. Simulation results demonstrate that compared with the standard PSO algorithm, seagull optimization algorithm(SOA), and whale optimization algorithm(WOA), the improved PSO algorithm achieves positioning error reductions of 72.5%, 61.9%, and 40.7%, respectively. Convergence accuracy is enhanced by 83.8% and 38.2% relative to SOA and WOA respectively, verifying its effectiveness in TDOA positioning for MEOSAR system.更多还原