摘要
为了提高无人机完成任务效率,在执行攻击任务前必需规划设计出高效的无人机飞行航路。提出了一种Q-学习的自适应蚁群算法的无人机航路规划方法,建立了基于真实地形数据和火力威胁区的威胁模型;针对传统蚁群算法在搜索过程中出现停滞现象,提出的Q-学习的自适应蚁群算法有效地解决了这一缺陷。并使用该算法对无人机的攻击任务航路进行了仿真计算,仿真结果表明该方法是一种有效的航路规划方法。
To improve the efficiency of UAVs in combat mission, a highly effective flight path must be worked out ahead of taking the mission. A path planning scheme for UAVs based on Ant - Q System is studied in the paper, which seems quite promising for the path planning problem. We present a threat model on the basis of real terrain data and fire threats. Since stagnation may appear during searching in use of traditional ant colonies algorithm, we put forward an adaptive Ant-Q system algorithm for solving the problem. We give a simulation example in Fig. 3 and Fig.4 to show that the method has some good characteristics and is effective in the path planning.
出处
《电光与控制》
北大核心
2007年第6期36-39,共4页
Electronics Optics & Control
基金
优秀博士学位论文创新基金项目(BC06003)
军队重点科研基金资助项目(HX05205)
关键词
无人机
航路规划
蚁群算法
数学形态学
威胁空间建模
UAV
modeling path planning
ant colonies algorithm
mathematical morphology
threat space