计算机工程

• 人工智能及识别技术 • 上一篇    下一篇

动态航迹规划关键技术研究

杨 俊1,朱 凡1,张 健1,郝 震2   

  1. (1. 空军工程大学工程学院,西安 710038;2. 中国人民解放军94270部队,济南 250117)
  • 收稿日期:2012-03-20 出版日期:2013-09-15 发布日期:2013-09-13
  • 作者简介:杨 俊(1987-),男,硕士研究生,主研方向:智能控制;朱 凡、张 健,副教授;郝 震,助理工程师

Research on Key Technology of Dynamic Flight Path Planning

YANG Jun 1, ZHU Fan 1, ZHANG Jian 1, HAO Zhen 2   

  1. (1. Engineering Institute, Air Force Engineering University, Xi’an 710038, China; 2. The 94270 Unit of PLA, Jinan 250117, China)
  • Received:2012-03-20 Online:2013-09-15 Published:2013-09-13

摘要: 针对实际作战环境中常遇到的突发威胁、运动威胁、任务改变等情形,提出一种基于战场态势预测的滚动粒子群优化多步预测规划算法。采用卡尔曼滤波和随机理论分别完成对威胁源指定时刻的状态预测和任务目标的概率对准,改进高效粒子群优化算法,并将其运用到局部滚动优化过程中,以多步规划、单步执行的形式实时调整航迹。仿真结果表明,该算法对战场威胁态势具有预测作用,对任务的改变反应灵敏,能较好地满足动态环境下的实时性要求。

关键词: 态势预测, 运动威胁, 滚动粒子群优化, 卡尔曼滤波, 动态规划

Abstract: For some cases of new or moving threats and task changing in the practical dynamic circumstance, a new roll Particle Swarm Optimization(PSO) multistep forecasting planning algorithm based on battlefield situation prediction is proposed in this paper. By using the Kalman filtering and probability theory, it completes the situation prediction for threat source, and the probability of the target task. Combined the upgrade PSO arithmetic with the roll optimization, it can adjust the path in real time by multi-step planning and one-step executing. Simulation result indicates that this algorithm plays well in dynamic circumstance especially in aspect of prediction the threat state and responding to the task changing.

Key words: situation prediction, mobile threat, roll Particle Swarm Optimization(PSO), Kalman filtering, dynamic planning

中图分类号: