基于烟花算法的移动机器人气味源定位方法研究

发布时间：2018-06-24 00:20

本文选题：移动机器人 + 气味源定位　；参考：《中国矿业大学》2017年硕士论文

【摘要】：随着机器人应用技术的发展,智能算法也逐渐开始在被待优化领域广泛应用,面对日渐增多的有害化学品意外泄露和生化恐怖袭击事件,急需我们研究出更加高效、更加鲁棒的气味源主动监测/定位方法,维护社会的稳定和安全。本文旨在利用携带气味传感器的机器人,研究不同环境下主动嗅觉的实现方法。已有的研究成果都不能兼备气味信息发现、气味跟踪和气味源位置确认的功能,需要机器人在主动嗅觉的三个阶段切换算法来完成完整的任务,这大大降低了机器人搜索的效率。针对移动机器人气味源定位方法面临的困境,烟花算法的出现正好可以弥补已有研究成果存在的缺陷,因此本文将烟花算法应用到机器人主动嗅觉领域。首先提出一种基于FWA的单机器人气味源定位方法,利用Gambit、Fluent以及Tecplot软件搭建不同的实验环境,并利用烟花爆炸的原理来指导机器人进行仿真实验,实验结果证明了该方法的可行性。其次,考虑到实际环境中的气味源在扩散过程中会形成旋涡,为了提高机器人在陷入局部搜索无效区域时的搜索效率,我们搭建了30*30m2的不同障碍物的仿真试验环境,其中风向和风速是连续变化的,并对FWA进行优化,提出一种基于OSL-FWA的单机器人气味源定位方法,大量的仿真实验结果证明这个方法大大地提高了机器人搜索的效率。最后,考虑到单机器人气味源定位系统在实际环境应用中,其鲁棒性较差,我们利用多机器人系统进行气味源定位,进一步地对OSL-FWA算法优化,提出一种基于AR-FWA的多机器人气味源定位方法,并在50*50m2的仿真环境中,增加障碍物的数量,改变风速等环境参数,通过大量的仿真实验证明了该方法的可行性和良好的鲁棒性能。
[Abstract]:With the development of robot application technology, intelligent algorithms are gradually being widely used in the field to be optimized. In the face of the increasing number of accidental leakage of harmful chemicals and biochemical terrorist attacks, we need to find out more efficient. More robust odor source active monitoring / positioning method to maintain social stability and security. The purpose of this paper is to study the realization of active olfaction in different environments by using a robot carrying odour sensor. The existing research results can not be combined with the functions of odor information discovery, odor tracking and odor source location confirmation, so it is necessary for the robot to complete the task by switching the three stages of active olfaction. This greatly reduces the efficiency of robot search. Aiming at the dilemma of the mobile robot odor source location method, the appearance of the fireworks algorithm can make up for the shortcomings of the existing research results. Therefore, this paper applies the fireworks algorithm to the robot active olfactory field. Firstly, a single robot odor source location method based on FWA is proposed. Gambit-fluent and Tecplot software are used to build different experimental environments, and the principle of fireworks explosion is used to guide the robot to carry out simulation experiments. The experimental results show that the method is feasible. Secondly, considering that the odour source in the actual environment will form vortex in the process of diffusion, in order to improve the search efficiency of the robot in the local search invalid area, we set up the simulation test environment of different obstacles in 30*30m2. The wind direction and wind speed are continuously changing, and FWA is optimized. A single robot odor source location method based on OSL-FWA is proposed. A large number of simulation results show that this method greatly improves the efficiency of robot search. Finally, considering the poor robustness of the single-robot odor source location system in the actual environment, we use multi-robot system to locate the odour source, and further optimize the OSL-FWA algorithm. A multi-robot odor source location method based on AR-FWA is proposed. In the simulation environment of 50*50m2, increasing the number of obstacles and changing the environmental parameters, such as wind speed, etc., the feasibility and robustness of the method are proved by a large number of simulation experiments.
【学位授予单位】：中国矿业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【参考文献】