节点文献
低空救援航空器航迹规划算法研究
Research on Algorithm of Aircraft Trajectory Planning in Low-altitude Rescue
【作者】 王磊;
【作者基本信息】 南京航空航天大学 , 交通运输规划与管理, 2015, 硕士
【摘要】 我国是世界上发生自然灾害以及各类公共安全事故灾害最为严重的国家之一,而航空救援作为立体救援有机整体的重要一环,其具有快速、高效、受地理空间限制少等优势,已然成为社会各界高度关注和重视的对象。除此之外,随着《低空空域管理使用规定》发布,低空开放政策日渐成熟,也让航空器低空安全运行成为近几年备受关注的研究热点。本文主要针对重大自然灾害下的通用航空器低空应急救援飞行航迹规划问题,将低空救援航空器航迹规划问题划分为预战术和战术两个层面进行研究,最终要实现航空器实时自主地规划航迹。预战术层面通过结合实际的低空救援飞行环境,先针对单个航空器航迹规划问题,在三维空域网格划分的基础上,基于改进的A?算法得到最优的初始飞行计划航迹。在单个航空器初始飞行航迹基础上,结合时间窗原理,提出了两种方法解决了多航空器无冲突飞行计划航迹规划问题,并通过计算机仿真对两种方法进行了对比分析。战术层面则在预战术航迹规划的基础上,结合已划设的空域网格,提出了基于航空器飞行状态模式估计和飞行意图推理的航迹预测算法。该算法结合了交互式多模型算法和改进的意图推理算法的优点,摒弃了两种算法单独使用时的不足之处,进而能够确保预测得到的航迹实时准确。基于航迹预测的冲突解脱算法在探测到飞行冲突时,为了实现航空器自主避让飞行冲突,将航空器当作独立自主的智能体Agent,依据救援目视飞行规则,在Airspace模型上,构建Agent模型。基于优先级准则,进行非协作式航向调整冲突解脱策略,并提出了协作式航向调整冲突解脱策略思想,确保航空器能够得到最优无冲突飞行航迹。最后,仿真验证了该算法的实时性和有效性。
【Abstract】 China is the country which suffered most serious natural disasters and various types of accidents around the world. Air rescue, as the significant part of three-dimensional rescue organic whole, has been remarkably concerned by all walks of society. Because it shows many advantages, such as faster, high efficiency, and less restricted. Apart from these advantages, as time goes by since Provisions for the use of low altitude airspace management has issue. The low-altitude safe operation of aircraft has become a hot research focus with the maturity of open policy on low-altitude flying. This paper is mainly about route planning of aviation aircraft in the low altitude emergency rescue. This research is divided into pre-tactical and tactical level on aircraft trajectory planning in low-altitude rescue. The ultimate purpose is to ensure that aircraft can independently search trajectories promptly and accurately.On the pre-tactical level, by combining the actual flight conditions, the optimal original trajectory is obtained. It is based on the single aircraft trajectory planning, three-dimensional mesh division of airspace and the improved A-star algorithm. Then with the outcome of single aircraft trajectory, combined with Time Window Theory, the thesis propose two methods about how to plan multi-aircraft trajectories without conflicts, and also to analyze the advantages and disadvantage by simulation experiments. On the tactical level, based on strategic trajectory and previous airspace grids, the paper puts forward SIBTP algorithm. This algorithm combines the advantages of both IMM algorithm and IIIA, and abandons the disadvantages when they are used separately, in this way, the new algorithm can ensure that the predicted trajectories are promptly and accurately. Based on the conflict resolution algorithm for trajectory prediction, with an effort to realize that aircrafts can avoid conflicts independently when there would be flight conflicts, the aircraft is treated as an independent intellectual agent and it is built on the airspace model according to visual flight collision avoidance rules. The optimal conflict-free trajectory has been worked out on the basis of priority rules and solutions to non-collaborate and collaborate trajectory adjustment. In the end, the thesis takes simulation tests on whether the SIBTP algorithm is actual in real time as well as be efficient.
【Key words】 Low-altitude rescue; trajectory planning; estimating state and flight mode of the aircraft; intent inference; Agent model;