基于改进RRT算法的室内火灾动态逃生路径规划  

Dynamic escape path planning of indoor fire based on improved RRT algorithm

作　　者：张学锋 刘明 ZHANG Xuefeng;LIU Ming(School of Computer Science&Technology,Anhui University of Technology,Ma'anshan 243032,China)

机构地区：[1]安徽工业大学计算机科学与技术学院,安徽马鞍山243032

出　　处：《苏州科技大学学报（自然科学版）》2022年第3期74-80,共7页Journal of Suzhou University of Science and Technology（Natural Science Edition）

基　　金：安徽省教育厅重大课题基金资助项目(KJ2017ZD05,TZJQR002-202)。

摘　　要：针对发生火灾时的智慧楼宇动态逃生路线问题,提出一种基于改进RRT算法的室内逃生路径动态规划算法。通过模拟浓烟和温度的改变,进行分区间量化,实现RRT算法中随机树扩展时更新最大步长,使随机树在扩展时小于不可通过区域的半径长度,从而减少碰撞检测次数,提高检测效率。同时,基于实际场景运用MATLAB建立栅格地图,在火灾发生的不同阶段进行更新栅格地图和动态步长,从而模拟出实际火灾下不同时间段的逃生情况,实现逃生路径动态规划。结果表明,该算法在火灾发生的不同阶段能够快速规划出逃生路线,为被困人员争取宝贵的逃生机会,具有实际应用价值。Absrtact:Aiming at the dynamic escape route of intelligent buildings during fire,a dynamic planning algorithm based on improved RRT algorithm was proposed.By simulating the smoke and temperature changes,the interval quantization was performed to update the maximum step length when the random tree extended in the RRT algo-rithm so that the random tree was less than the radius length of the non-passable region,the number of collision detection reduced and the detection efficiency improved.Based on the actual scene,a grid map was established by using MATLAB.In different stages of the fire,the grid map and dynamic steps were updated so as to simulate the escape situation of different time periods under the actual fire and realize the dynamic planning of the escape path.The results show that the algorithm can quickly plan the escape route at different stages of fire and win over valuable escape opportunities for trapped people,which takes great practical value.

关 键 词：改进RRT算法 路径规划 火灾疏散 栅格法 

分 类 号：TU993.1[建筑科学—市政工程] TU17