【作者】 杨杰；

【导师】 蔡超；

【作者基本信息】 华中科技大学 ， 模式识别与智能系统， 2013， 硕士

【摘要】 随着航迹规划技术在军事及航空、航天领域得到广泛的应用，越来越多的研究者开始围绕这一领域展开研究，并提出了多种航迹规划算法，以满足不同飞行约束条件下航迹规划问题的需求。传统的航迹规划算法较少考虑端点方向约束的问题，规划过程中，一旦到达目标点，则认为规划结束。而在实际应用中，为躲避某些威胁因素、提高突防概率，通常要求飞行器能够从目标的特定方向到达。本文就这一问题展开研究。首先对国内外航迹规划领域的研究现状进行了介绍，简要描述了三种具有代表性的航迹规划算法。同时，对A~*算法的基本原理和流程进行了分析，讨论了算法规划结果的最优性与启发函数之间的关系，并就算法的两个关键要素Open表和Close表的作用和用法进行了研究。在此基础上，指出了将A~*算法应用于具有端点方向约束的航迹规划问题的局限性。分析了该局限性产生的具体原因，并通过实验进行了验证。针对A~*算法在进行具有端点方向约束的航迹规划时，由于缺乏方向引导而导致效率低下的问题，提出了动态引导A~*算法。在原有A~*算法的基础上引入了动态变化的引导点，给出了引导点的设置策略，设计了新的代价函数，说明了引导点的切换方法。在航迹规划过程中，在引导点的作用下，该方法能够快速地规划出从特定方向出发，沿指定方向进入的飞行航迹。实验表明，该方法在保证产生满意航迹的同时，还具有较高的规划效率。将动态引导A~*算法应用于航迹网络规划中。使用该算法在具有方向约束的节点之间规划航迹片段生成航迹网络。在进行基于必过区域约束的航迹规划时，可将必过区域转化为引导信息，并用动态引导A~*算法规划出结果。更多还原

【Abstract】 With the wide use of route planning in mailitary, aviation and aerospace fields, more and more researchers begin to conduct researches around this area, and a variety of algorithms have been proposed to meet the different constraints of the route planning problem. The traditional path planning algorithms seldom consider the constraints of directional endpoints. In the planning process, once the path reaches the target in any direction, the planning is considered to have been finished. However, In applications, aircraft often departure in a predetermined direction and arrive along a specified direction to escape certain threats and improve the efficiency of penetration. This paper studies on this issue.In this paper, the research status of the path planning field is introduced, and a brief description of three typical planning algorithms is given. Then the basic principles and processes of the A*algorithm is described in detail. This paper also discussed the optimality effect of the heuristic funciton of the A*algorithm. Then, gives descriptions of the role and implementation of two key elements in the algorithm, the Open table and the Close table. Based on these analysis, it is pointed out that there are some limitations in the A*algorithm when it is applied to the route planning problem with the constraints of directional endpoints. The reasonsof the limitation are given and verified through experiments.In the path planning with constraints of directional endpoints, traditional A*algorithm is inefficient due to lack of direction guide. To solve this problem, this paper proposes a dynamic guiding A*algorithm, which introduces dynamic guide points during the searching process of A*algorithm. The dynamic guiding strategies are given, and then a guiding points generation method is proposed. At the same time, this paper designes a new evaluation function, and describes the switch method of the guide points. In the planning process, the dynamic guiding A*algorithm could find the resultant path, which goes from the specific direction of the start point to the fixed direction of the target point, efficiently. Experimental results show that compared with A*algorithm, this new method could generate a satisfactory path as well, while the planning speed is greatly improved. At last, the dynamic guiding A*algorithm is applied to plan path network. The path network could be generated by planning path between two directional endpoints using this method. In the path planning with the constraints of area that must be passed through, this method is also very applicable using the guide information extracted from the area.更多还原