ID 原文 译文
38606 最后,开发一个利用遗传算法求解规划模型的图形化界面计算工具,利用该工具计算对比了2种目标函数和3种转向方式对于舰载机出动架次决策的影响。 Finally, developing a using genetic algorithm and calculate the graphical interface programming model, the tool is used to calculate compared the two methods for the objective function and three to for aircraft sortie decisions.
38607 计算结果表明,模型和工具可行,剩余航程等分点转向方式总是优于原有的转向方式。 The results show that the model and the tool is feasible and the remaining distance etc equinoctial steering mode is always better than the original method.
38608 针对强海杂波背景下微弱动目标信号提取困难、雷达检测性能差的问题,在稀疏表示理论的基础上,提出利用字典学习算法抑制海杂波、重构目标信号。 In view of the strong sea clutter background weak moving target signal extraction difficult, radar detection problem of poor performance, on the basis of the theory of sparse representation, use a dictionary to study algorithm inhibition of sea clutter, reconstruct the target signal.
38609 该算法通过K类奇异值分解(K-singular value decomposition,K-SVD)算法学习海杂波和目标的稀疏域主成分特征,得到相应的学习字典,抑制海杂波并对目标信号稀疏重建,解决了以往固定字典与高海况下雷达回波匹配度低、目标信号提取效果差的问题;并通过算法参数的分析和优化,进一步提高了算法性能和工程实用性。 The algorithm by K class singular value decomposition (K - singular value decomposition, K - SVD) algorithm study of sea clutter and target sparse principal component characteristics, get the corresponding learning dictionary, inhibition of sea clutter and thinning of the target signal reconstruction, solved the previous fixed dictionary radar echo matching degree is low and high sea condition, target signal extraction effect is poor;And through the algorithm analysis and optimization of the parameters, further improves the algorithm performance and practicability.
38610 基于实测数据的实验结果表明,相比传统检测方法,所提算法能够有效检测高海况下微弱动目标,显著提升检测性能。 Based on the measured data of the experiment results show that compared with traditional detection method, the proposed algorithm can effectively detect dim moving target, high sea condition significantly improve detection performance.
38611 快速扩展随机树(rapidly-exploring random tree,RRT)无人机航迹规划方法能够快速获得满足约束要求的可行航迹,但是无法获得接近最短航迹的较优航迹。 Rapid expanded random tree (rapidly - exploring the random tree, RRT) unmanned aerial vehicle (uav) flight path planning method can quickly obtain satisfies the requirement of the constraint of feasible path, but unable to get close to the optimal path of the shortest path.
38612 针对航迹的最优性问题,提出了混合种群RRT无人机航迹规划方法。 In view of the track is the optimality problem, proposed the hybrid population uavs RRT path planning method.
38613 在基于环境势场的RRT算法的基础上,设计了一种种群优化方法,通过引入自优化种群和协同优化种群改善航迹段,使算法同时具有局部和全局寻优能力。 In RRT algorithm based on potential field environment, on the basis of design a population optimization method, by introducing the optimum population and collaborative optimization to improve the track section, the algorithm of both local and global optimization ability.
38614 在得到航迹节点的基础上,采用B样条曲线的平滑方法生成曲率连续的可跟踪航迹。 In track nodes, on the basis of B spline smoothing method is used to generate the curvature can be track.
38615 仿真结果表明,所提算法能够综合考虑无人机航程代价和雷达威胁代价,快速地收敛得到接近最优且满足无人机动力学约束的可行航迹,在不同环境下也能有满意的收敛效率。 The simulation results show that the proposed algorithm can comprehensive consideration of unmanned aerial vehicle (uav) flight cost price and radar threat, fast convergence to get close to the optimal and meet the feasible path of uav dynamic constraints, in a different environment can have a satisfactory convergence efficiency.