A Study On SLAM Technology Of Mobile Robot In Room

With the rapid development of society and science and technology,mobile robots are widely used in people's daily life,which greatly meet people's living needs,and can prevent and reduce the occurrence of all kinds of safety accidents.Especially in the prevention of Newcastle disease,mobile robots can assist to reduce the risk of infection.Among the key technologies needed in the process of completing these auxiliary work is synchronous positioning and map construction(Simultaneous Location and Mapping,SLAM).However,there are still some problems in indoor SLAM technology,such as low positioning accuracy and large drawing error.Therefore,this thesis will focus on the filtering problem of indoor SLAM algorithm of mobile robot.This has great theoretical value to improve the accuracy of mobile robot positioning and drawing.Firstly,each module of the SLAM system of mobile robot is modeled.The SLAM probability model is established in order to process and optimize the calculation in mathematical form,the Cartesian coordinate system is used to establish the coordinate system model for indoor SLAM research of mobile robot,and the simplified model of mobile robot is built on this coordinate system.Secondly,in view of the problem that the particle filter algorithm in the FastSLAM leads to the loss of diversity with the increasing number of resampling,the deviation correction exponential weighted average algorithm is introduced to optimize the combination of the weight of the particles.The improved resampling method can ensure the diversity and stability of particles.Then,in view of the problem of sample particle set degradation in robot pose estimation,the FastSLAM algorithm is integrated into intelligent group algorithm to make the particle set more approximate to the real state,and the previously improved resampling method is added.A FastSLAM algorithm based on improved butterfly algorithm is proposed.The improved butterfly algorithm mainly integrates the observation and state information of the latest time of the robot into the fragrance formula of the butterfly algorithm,and adds the adaptive fragrance radius and the adaptive adjustment step factor to the process of updating the butterfly position to improve the operation efficiency and prediction accuracy of the algorithm.Finally,through the simulation experiments of MATLAB and ROS platform,it is concluded that the improved algorithm proposed in this thesis is superior to FastSLAM,in estimating accuracy and stability,and can meet the practical application requirements of mobile robot positioning and mapping at the same time.And lay a solid theoretical foundation for the subsequent autonomous navigation research of mobile robot.