ID 原文 译文
56248 脉冲星导航是一种极具潜力的深空自主导航技术,通常采用脉冲到达时间(time of arrival,TOA)作为量测信息. Pulsar navigation is a promising deep space autonomous navigation technology, which usually usesthe time of arrival (time of arrival, TOA) as the measurement information.
56249 但脉冲星星历误差和星载原子钟误差等系统误差对导航性能有显著影响. However, system errors such as pulsarephemeris errors and space-borne atomic clock errors have a significant impact on navigation performance.
56250 为了解决上述问题,提出了一种基于TOA和时间差分TOA (TDTOA)的增广状态脉冲星组合导航误差抑制方法,通过将每个脉冲星的星历误差和时钟误差增加到状态向量,并利用TOA和TDTOA量测值对其进行估计和校正. Tosolve the above problems, an error suppression method for augmented state pulsar integrated navigation basedon TOA and time difference TOA (TDTOA) is proposed. By adding the ephemeris error and clock error ofeach pulsar to the state vector, the TOA and TDTOA measurements are used to estimate and correct them.
56251 仿真结果表明,该方法提高了脉冲星星历误差和时钟误差的可观测性,有效地消除了这些系统误差的影响,导航精度相比传统脉冲星导航提高了29%. Simulation results show that this method improves the observability of the pulsar ephemeris and clock errors,eliminates the effect of these systematic errors, and improves navigation accuracy by 29% compared to traditionalpulsar navigation.
56252 随着无人机应用领域的增多,多无人机协同航迹规划问题变得愈发重要. With the increase in the application areas of unmanned aerial vehicles (UAVs), the problem ofcoordinated path planning for multi-UAVs becomes more and more significant.
56253 然而,现存的多无人机协同航迹规划问题大多将多个目标加权转换为单目标问题进行优化,为减少多目标加权的主观性,本文提出一种基于高维多目标优化的多无人机协同航迹规划模型,此模型可以同时优化多无人机航迹距离代价、多无人机航迹威胁代价、多无人机航迹能耗代价,以及多无人机协同性能. However, most of the existingmethods optimize this problem by converting multi-objective weighting into a single-objective problem. Toreduce the subjectivity of multi-objective weighting, a coordinated path planning model based on many-objectiveoptimization is proposed to optimize the multi-UAVs’ track distance cost, track threat cost, track energy cost,and coordination performance. In the many-objective evolutionary algorithm (MaOEA), the convergence anddiversity of the individuals are both important indicators to evaluate the performance of individuals.
56254 同时,为提高高维多目标优化算法在解决此模型时的性能,提出一种基于个体评估交叉策略的NSGA-Ⅲ算法(NSGAⅢ-ICO),可以根据算法运行代数综合评估个体优劣并指导种群交叉操作. To solve theproblem that individuals in the algorithm are not easy to evaluate, an NSGA-III algorithm based on the individualevaluation mating strategy (NSGAIII-ICO) is proposed, which can comprehensively evaluate the performance ofindividuals according to the generation of the algorithm and guide the mating operation.
56255 仿真结果证明,此模型可以有效地提供多无人机协同航迹,且通过与其他高维多目标优化算法的比较,可以证明此改进算法可以有效地提高多无人机协同航迹规划的多种性能. Simulation resultsshow that this model can effectively provide coordinated tracks for multi-UAVs, and by comparing it with otherMaOEAs, it can be proved that this proposed algorithm can effectively improve the performance of coordinatedpath planning for multi-UAVs.
56256 高速列车牵引系统的微小渐变故障诊断是保障高速列车安全可靠运行的一项重要任务,传统的数据驱动方法通常难以在故障初期实现微小渐变故障的诊断. The incipient fault diagnosis of high-speed train traction systems is a fundamental but essential taskto guarantee the safety and reliability of high-speed trains. Traditional intelligent data-driven methods requirecomprehensive expertise and cannot diagnose incipient faults in the early period.
56257 本文考虑到传感器时序信号的非平稳特性和微小渐变故障具有的慢时变特性,提出了一种基于改进长短期记忆网络（long short-term memory,LSTM）的微小渐变故障无监督诊断方法. This paper focuses on thefault diagnosis approach for incipient time-varying faults and unknown faults using sensor signals. Consideringthat the sequential signal has non-stationary characteristics and the incipient faults are slow time-varying, animproved LSTM (long short-term memory) network that is an unsupervised recurrent neural network, is developed.