基于改进冲突搜索的多智能体路径规划算法

多智能体路径规划问题在航空航天领域的多机任务中应用广泛但求解困难。基于改进冲突搜索的算法被设计用来快速求解多智能体路径规划问题。全局路径规划方面，首先设计综合考虑路径代价总和以及最大完工时间的多目标代价函数，其次提出基于唯一最短路径的冲突分类及消解方案，降低多智能体路径规划的计算量。在线冲突消解方面，利用速度障碍法在线检测和消解智能体与动态障碍物间的突发冲突。仿真结果表明，本文算法在全局路径规划方面保留基于冲突搜索算法的最优性并且降低了算法计算量，同时本文算法能够有效实现在线冲突检测与消解。     

基于蚁群算法的无人机三维路径规划与冲突解脱

针对传统无人机路径规划方法难以满足无人机在复杂山区环境应用这一缺点,提出一种基于蚁群算法并利用插值算法和容斥算法优化的无人机三维路径规划算法。通过无人机飞行约束条件和对蚁群算法的优化在所建立的三维空间中得到最优路径。利用二次、三次样条插值、容斥算法等优化算法对无人机路径优化,规划出一条更加平滑的便于无人机执行的最佳路径,并建立了无人机冲突约束,解决了无人机冲突。     

结合跳点引导的无人机随机搜索避撞决策方法

针对无人机在城市空域环境和密集交通流下的避撞决策问题，提出马尔科夫决策过程（MDP）和蒙特卡洛树搜索（MCTS）算法对该问题进行建模求解。蒙特卡洛树搜索算法在求解过程中为保证实时性而使其搜索深度受限，容易陷入局部最优，导致在含有静态障碍的场景中无法实现避撞的同时保证全局航迹最优。因此结合跳点搜索算法在全局规划上的优势，建立离散路径点引导无人机并改进奖励函数来权衡飞行路线，在进行动态避撞的同时实现对静态障碍的全局避撞。经过多个实验场景仿真，其结果表明改进后的算法均能在不同场景中获得更好的性能表现。特别是在凹形限飞区空域仿真模型中，改进后的算法相对于原始的蒙特卡洛树搜索算法，其冲突概率降低了36%并且飞行时间缩短47.8%。     

低空无人机路径规划算法综述

随着无人机技术的发展,无人机在低空的应用场景越来越多,复杂的低空环境对无人机路径规划算法提出了新的要求。本文总结了近年来常用的无人机路径规划算法,包括图搜索算法,线性规划算法,智能优化算法（遗传算法、粒子群算法、蚁群算法）,强化学习算法;对这些算法的原理、适用场景及其优缺点进行了归纳分析;并基于无人机发展现状对无人机路径规划算法进行了展望。     

基于HTGVNS算法的卡车与无人机协同配送方法

在现代物流中,针对最后一公里配送中的成本高昂、耗时长、道路不易通行等难题,首次引入卡车与无人机协同配送模式,并以总配送时间最小化为目标,建立混合整数规划模型(MIP)。提出一种新的混合禁忌可变邻域搜索算法(HTGVNS)进行求解。算法利用旅行商问题(TSP)求解器和系统性邻域变化的探索能力求解卡车路径,将客户分配给无人机形成聚类,再使用禁忌搜索过程来优化无人机路径。通过求解不同规模算例验证了所建模型的可行性和算法的有效性。结果表明卡车与无人机协同配送模式能有效地减少配送时间,提高配送效率。     

一种适用于中期冲突探测的概率型算法

借鉴飞机沿航迹误差服从正态分布这一经典结论,对飞机沿航迹方向运动进行建模,根据同高度两架飞机的相遇几何约束,建立适当的直角坐标系表示两架飞机的位置,构造椭圆形冲突区域,并在此基础上提出了一种针对同高度两条交叉航路的中期冲突探测的概率型算法,然后结合仿真曲线分析了各影响因素与冲突概率之间的关系.结果表明:该算法对于15～20 min预测时间范围的冲突探测可以取得良好地效果;相同条件下,两条航线夹角越大,探测精度越高.同时与以往的随机化算法相比,该算法具有计算方法简单、计算速度快、计算结果精度高等优点.最后将该算法推广至可适用于由多个航路段组成的航路之间的冲突探测,从而使该算法具有更广泛的应用性.     

移动机器人路径规划方法简介

综述了移动机器人历史,路径规划的分类。重点介绍了移动机器人路径规划的环境模型和搜索算法,以及全局路径规划和局部路径规划,比较各种方法的优缺点,最后介绍了其应用范围。     

基于蜂群算法的飞机滑行路径优化

研究了飞机场面滑行路径动态规划问题,将三种滑行冲突作为约束条件,建立了场面运行模块化模型。基于蜂群算法给出了问题的优化算法,并进行了计算机仿真实验,结果表明可以大大减少滑行时间。算法既可以用于滑行路径的动态规划,也可以为繁忙机场的安全运行提供决策支持。     

一种无人机局部路径重规划算法研究

提出了一种突发威胁体下无人机局部路径重规划的算法。首先根据不同威胁体的分布情况构造无人机的可飞航路集,用“改进型V orono i图”表示出来,采用D ijkstra算法求解初始粗略最短路径。在无人机飞行过程中,通过基于混合动态贝叶斯网络的切换线性动态系统模型感知环境,应用V iterb i解码算法确定突发威胁体的实时位置及威胁等级,再依据局部路径重规划原则进行寻优,最后应用三次平滑及序列二次规划方法获得实际可飞路径,并用M atlab仿真验证了算法的有效性。     

考虑运动学约束的不规则目标遗传避碰规划算法

 针对复杂环境下不规则目标的路径规划问题,提出了一种带有运动学约束的遗传避碰规划算法。以舰载机在航母甲板上的路径规划问题作为研究对象,并且该算法可推广至其他具有此类约束的路径规划问题中,它较好地解决了目标形状复杂、障碍环境复杂、目标运动时带有回转半径约束等特殊问题。在传统遗传路径规划算法的基础上,针对性地设计了三维位置和姿态混合编码、三段法路径解码、轨迹包围盒的碰撞检测及距离计算等方法,并在遗传操作中引入惩罚项和修补策略来辅助算法寻优。最后,为得出复杂环境下的最优路径,基于VC++平台对算法进行了仿真验证。结果表明,在复杂障碍环境下,本文提出的算法可求得最优避碰路径,并满足预先设定的目标回转半径约束,能够有效地解决此类目标的避碰路径规划问题。     

Hierarchical path computation approach for large graphs

Time is a critical factor in several path planning problems such as flood emergency rescue operations, escape planning from fires and chemical warfare agents dispersed in large buildings, evacuation from urban areas during natural disasters such as earthquakes, and military personnel movement. We propose a hierarchical path planning algorithm (HIPLA) for real time path planning problems where the computational time is of critical significance. The main idea of HIPLA is to significantly reduce the search space for path computation by searching in a high-level abstraction graph, whose nodes are associated with precomputed risk estimates. The cumulative risk associated with all nodes along a path determines the quality of a path. We present a detailed experimental analysis of HIPLA by comparing it with two well-known approaches viz., shortest path algorithm (SPAH) [1] and Dijkstra's algorithm with pruning [2] for large node-weighted graphs.     

A hybrid genetic approach for airborne sensor vehicle routing in real-time reconnaissance missions

Past initiatives to address surveillance and reconnaissance mission planning mainly focused on low-level control aspects such as real-time path planning and collision avoidance algorithms in limited environment. However, few efforts have been spent on high-level real-time task allocation. It is believed that automated decision capabilities supporting real-time resource allocation for sensor control and interactions might significantly reduce user workload, focusing attention on alternate tasks and objectives while assigning hard computational tasks to artificial agents. In this paper, we propose a new hybrid genetic algorithm to solve the dynamic vehicle routing problem with time windows, in which a group of airborne sensors are engaged in a reconnaissance mission evolving in a dynamic uncertain environment involving known and unknown targets/threats. In that context, visiting a target may consist in carrying out a collection of subtasks such as search, detect, recognize and confirm suspected targets, discover and confirm new ones. The approach consists in concurrently evolving two populations of solutions to minimize total travel time and temporal constraint violation using genetic operators combining variations of key concepts inspired from routing techniques and search strategies. A least commitment principle in servicing scheduled customers is also exploited to potentially improve solution quality.     

Study on the resolution of multi-aircraft flight conflicts based on an IDQN

With the rapid growth of flight flow, the workload of controllers is increasing daily, and handling flight conflicts is the main workload. Therefore, it is necessary to provide more efficient conflict resolution decision-making support for controllers. Due to the limitations of existing methods, they have not been widely used. In this paper, a Deep Reinforcement Learning(DRL) algorithm is proposed to resolve multi-aircraft flight conflict with high solving efficiency. First, the characteristics ...     

Satellite range scheduling with the priority constraint: An improved genetic algorithm using a station ID encoding method

Satellite range scheduling with the priority constraint is one of the most important problems in the field of satellite operation.This paper proposes a station coding based genetic algorithm to solve this problem,which adopts a new chromosome encoding method that arranges tasks according to the ground station ID.The new encoding method contributes to reducing the complexity in conflict checking and resolving,and helps to improve the ability to find optimal resolutions.Three different selection operators are designed to match the new encoding strategy,namely random selection,greedy selection,and roulette selection.To demonstrate the benefits of the improved genetic algorithm,a basic genetic algorithm is designed in which two cross operators are presented,a single-point crossover and a multi-point crossover.For the purpose of algorithm test and analysis,a problem-generating program is designed,which can simulate problems by modeling features encountered in real-world problems.Based on the problem generator,computational results and analysis are made and illustrated for the scheduling of multiple ground stations.     

微小卫星低可观测外形飞行姿态规划

为提高在轨微小卫星的使用效能及生存能力,提出一种微小卫星低可观测外形飞行姿态规划算法.根据微小卫星雷达散射截面(RCS)、轨道及雷达威胁特性,建立了可进行长时间内最佳飞行姿态规划的数学模型,设计了低计算复杂度的链表式个体结构及进化规划策略,并实现了算法对高威胁区优化规划的能力.同时,算法低迭代步长下的快速收敛特性以及进...     

重叠网格洞面优化技术的改进与应用

针对传统割补法在壁面狭缝处网格重叠失败的问题,提出了以单元顶点状态进行洞面优化的vertex方法及以单元大小进行洞面优化的volume方法的两种改进方法。这两种方法将传统割补法中切割和填补两步搜索方法改为一步搜索方法进行重叠网格的洞面优化。通过数值试验证明:两种洞面优化方法都能很好地解决壁面狭缝网格重叠问题,网格重叠区域流场数值传递正确,计算结果与试验值吻合,优化效率高,收敛速度快。volume优化方法与vertex优化方法相比可以减少由网格大小不对等引起的数值误差,计算结果和计算精度均优于vertex优化方法。     

有限姿控能力的低RCS微小卫星姿态实时规划

 为提高在轨微小卫星使用效能及生存能力,提出一种受姿控能源消耗及驱动能力约束的低雷达散射截面（RCS）微小卫星姿态实时规划算法。算法应用曲面像素法、时域有限差分法及假设检验,在三维空间内对微小卫星的RCS及雷达探测水平进行建模,并结合雷达分布模型构建了相应的威胁评估函数及规划代价评估函数。同时,为提高算法的实时性能,采用了粒子群优化(PSO)算法以降低计算复杂度。仿真结果表明,在有限计算量的基础上,算法能够以较小规划代价有效降低卫星威胁方向的RCS及雷达探测概率,满足对微小卫星飞行姿态实时规划的需要。     

基于局部几何特征的稠密点云配准方法

针对现有的稠密点云配准方法依赖初始位置设定、计算成本高、配准成功率不高等问题，提出了一种基于点云局部几何特征的稠密点云配准方法。采用深度卷积网络模型提取点云的局部几何特征，从而减少了三维点云数据的噪声、低分辨率和不完备性等带来的影响。在此基础上，使用K维树搜索完成局部几何特征描述子的关联工作。最后，通过随机采样一致算法对点云的相对位姿进行鲁棒的估计。通过对开源数据集上5个典型场景中的数据测试表明，该方法的配准成功率达到92.5%，配准精度达到0.0434m，配准时间相对最邻点迭代配准算法缩短了74.7%，实验结果验证了该方法的有效性、实时性和鲁棒性。     

相关色噪声下无冗余累积量稀疏表示DOA估计

针对传统均匀线阵中四阶累积量计算复杂度大、对快拍数敏感的问题,提出了一种快速去冗余的高分辨波达方向估计新方法。该方法首先通过构造选择矩阵对四阶累积量矩阵进行第1次降维处理,摒弃传统四阶累积量中大量冗余数据,然后对无冗余累积量矩阵进行矢量化并通过二次降维得到统计性能更优的向量观测模型,最后在相应的过完备基下建立观测模型的稀疏表示进行波达方向(Direction of Arrival,DOA)估计。同时将方法推广到L型阵列2维DOA估计,扩展了其应用范围。与传统的四阶累积量方法相比,该方法大大地减小了计算量,对快拍数要求不高,并且能够有效地抑制相关色噪声。理论分析和仿真实验验证了该方法对1维和2维DOA估计都具有较高的估计精度和分辨率。     

A multi-objective multi-memetic algorithm for network-wide conflict-free 4D flight trajectories planning

Under the demand of strategic air traffic flow management and the concept of trajectory based operations (TBO),the network-wide 4D flight trajectories planning (N4DFTP) problem has been investigated with the purpose of safely and efficiently allocating 4D trajectories (4DTs) (3D position and time) for all the flights in the whole airway network.Considering that the introduction of large-scale 4DTs inevitably increases the problem complexity,an efficient model for strategic level conflict management is developed in this paper.Specifically,a bi-objective N4DFTP problem that aims to minimize both potential conflicts and the trajectory cost is formulated.In consideration of the large-scale,high-complexity,and multi-objective characteristics of the N4DFTP problem,a multi-objective multi-memetic algorithm (MOMMA) that incorporates an evolutionary global search framework together with three problem-specific local search operators is implemented.It is capable of rapidly and effectively allocating 4DTs via rerouting,target time controlling,and flight level changing.Additionally,to balance the ability of exploitation and exploration of the algorithm,a special hybridization scheme is adopted for the integration of local and global search.Empirical studies using real air traffic data in China with different network complexities show that the pro posed MOMMA is effective to solve the N4DFTP problem.The solutions achieved are competitive for elaborate decision support under a TBO environment.