基于改进AEO算法的多机器人主动嗅觉室内味源定位研究  

An Indoor Odor Source Locating Method for Multi-robot Active Olfaction Based on Improved AEO

作　　者：傅均[1] 沈路遥 刘锐蕊 FU Jun;SHEN Luyao;LIU Ruirui(School of Computer Science&Information Engineering,Zhejiang Gongshang University,Hangzhou Zhejiang 310018,China)

机构地区：[1]浙江工商大学计算机与信息工程学院,浙江杭州310018

出　　处：《传感技术学报》2021年第10期1406-1411,共6页Chinese Journal of Sensors and Actuators

基　　金：国家自然科学基金项目(61305030)。

摘　　要：用机器人主动嗅觉来自动寻找和定位毒气泄漏、火灾等气味源在防灾、反恐、探险等场景中具有重要意义。针对室内环境中机器人气味源定位方法对于策略切换阈值敏感问题,本文提出了一种用于多机器人气味源搜索的风向人工生态系统优化算法WAEO。该方法将机器人看作生产者、消费者或者分解者,通过三者的能量传递机制进行算法的优化,使算法在劣势初始位置时依旧保持稳定的性能。同时,考虑到气味源扩散的特点设计了离散风向系统,在生态系统中增加了追风者角色,从全局和个体维度引入风向信息和层次化利用风向信息。通过与两种常见粒子群算法的对比实验,发现WAEO算法可以减少味源定位的搜索时间,提高搜索成功率,特别是在机器人数量较少时优势更为明显。Using robot active olfaction to automatically search and locate toxic gas leaks,fire spots and other odor sources is of great significance in some scenes such as disaster prevention,counter-terrorism and exploration.Aiming at the problem that indoor odor source localization is sensitive to the threshold of strategy switching,this paper proposes a wind artificial ecosystem-based optimization algorithm named WAEO for multi-robot odor source search.In this method,some robot is regarded as a producer,a consumer or a decomposer,and the algorithm is optimized through the energy transfer mechanism of the three,so that the algorithm still maintains a stable performance when the initial position is inferior.Moreover,considering the characteristics of odor diffusion,a discrete wind system is designed,and a role of wind chaser is added in the ecosystem,where wind information is introduced from global and individual levels,and is utilized hierarchically.The comparative experiments with two common partical swarm optimization algorithms show that the WAEO algorithm can reduce the search time of the source localization and improve the search success rate,especially when the number of robots is small.

关 键 词：主动嗅觉 人工生态系统优化算法 多机器人 味源定位 风向信息 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]