基于BF-PSO优化的未知环境下移动机器人导航与环境建模

发布时间：2018-05-12 14:48

本文选题：移动机器人 + 导航　；参考：《广西大学》2017年硕士论文

【摘要】：随着科学技术的不断发展,人类社会的智能化程度与自动化程度在不断提高,智能化的移动机器人变得越来越贴近人类的生活,像扫地机器人,服务机器人,救灾机器人等与人们的生活息息相关。对移动机器人的导航和环境建模的研究一直以来都是国内外学术研究工作者们研究的重点。如何才能使得移动机器人在未知环境中导航时,能像人类一样做出合理的决定,一直是所有机器人技术研究者们期望实现的目标。而如何使得移动机器人能在移动过程中完成对未知环境的精确建模也是移动机器人研究的一个重要方向。本论文针对采用传统人工势场方法实现移动机器人导航存在的缺点和不足,提出了一种改进的人工势场方法。该方法通过改进斥力函数和添加旋转力对人工势场法进行改进,最小化斥力势场的扭曲程度,实现在目标点取到势场的全局最小值。仿真表明采用这种方法进行的人工势场法的改进,可以提升移动机器人运动轨迹的平滑度,减少障碍物周围的不规则抖动,实现目标点的可达性。针对移动机器人的势场函数参数,步长等参数,提出了基于粒子群优化细菌觅食算法(BF-PSO)算法的参数优化。为研究BF-PSO算法优化所获得参数对移动机器人导航的优化效果,分别设计了基于优化参数设置和基于经验参数设置的改进人工势场法移动机器人导航实验。仿真实验验证了BF-PSO优化参数对移动机器人导航优化效果提升的可行性,和获得更短路径的有效性。分别研究了基于扩展卡尔曼滤波算法(EKF)算法与无迹卡尔曼滤波器算法(UKF)算法在未知环境下的移动机器人环境建模方法。两种环境建模算法采用同样的仿真环境进行了仿真实验。两种方法的实验均获得了一条估计到的机器人运动路径和观测到的路标,输出了估计路径与真实路径的误差,观测路标位置与真实路标位置的误差。实验证明了,基于UKF算法的移动机器人环境建模方法获得机器人的估计路径与观测路标具有更好的精确度。
[Abstract]:With the continuous development of science and technology, the degree of intelligence and automation of human society is constantly improved, intelligent mobile robots become more and more close to human life, such as floor sweeping robots, service robots, Disaster relief robots and so on are closely related to people's lives. The research on navigation and environmental modeling of mobile robots has been the focus of academic researchers at home and abroad. How to make the mobile robot navigate in the unknown environment and make reasonable decisions like human beings is always the goal that all the robotics researchers expect to achieve. How to make the mobile robot complete the accurate modeling of the unknown environment in the mobile process is also an important research direction of the mobile robot. In this paper, an improved artificial potential field method is proposed to solve the shortcomings and shortcomings of the traditional artificial potential field method for mobile robot navigation. This method improves the artificial potential field by improving the repulsion function and adding the rotation force to minimize the distortion of the repulsion potential field and achieve the global minimum value of the potential field at the target point. Simulation results show that the improved artificial potential field method can improve the smoothness of mobile robot trajectory reduce the irregular jitter around obstacles and achieve the reachability of target points. Aiming at the parameters of potential field function and step size of mobile robot, the parameter optimization of BF-PSO-based bacterial foraging algorithm based on particle swarm optimization (PSO) is proposed. In order to study the optimization effect of the parameters obtained by BF-PSO algorithm on mobile robot navigation, an improved artificial potential field method for mobile robot navigation experiments based on optimized parameters setting and empirical parameter setting was designed respectively. The simulation results show that the optimization parameters of BF-PSO are feasible to improve the navigation efficiency of mobile robot, and the effectiveness of obtaining shorter path is verified. The modeling methods of mobile robot environment based on extended Kalman filter (EKF) algorithm and unscented Kalman filter algorithm (UKF) in unknown environment are studied respectively. Two environmental modeling algorithms are simulated in the same simulation environment. In the experiments of both methods, an estimated robot motion path and observed road sign are obtained, and the errors between the estimated path and the real path are outputted, and the errors between the position of the road sign and the position of the real road sign are obtained. The experimental results show that the environment modeling method of mobile robot based on UKF algorithm has better accuracy in obtaining the estimated path and observation signpost of the robot.
【学位授予单位】：广西大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP242

【参考文献】