| With the continuous development of internet of things technology,artificial intelligence has come into people's daily life.At the same time,with the acceleration of the pace of life and the increasing problem of aging,home service robot come into being.Home service robot is a kind of robot,which provides services according to people's instructions.Due to the complexity and particularity of indoor environment,the most important link for mobile robot to complete the relevant instructions is to obtain the information of robot's and target's position,and complete the corresponding path planning.Therefore,based on studying the relevant theoretical background,this thesis studies the bluetooth-based man-machine locationing and tracking algorithm in indoor environment.In this thesis,it has investigated the existing technologies,which used in indoor environment.Combining with the particularity and complexity of indoor environment,the RSSI semaphore localization characteristics of bluetooth are studied,and an indoor localization method based on heterogeneous plane anchor,node mapping and grid quantization is proposed.The method improves the indoor location model by comparing the relationship between RSSI signal and distance received under different anchor node layouts.At the same time,it adoptes the method of iterative quantization correction and virtual anchor node calculation to continuously correct and adjust the position information calculated by the positioning algorithm,so that to achieve the purpose of improving the positioning accuracy.Secondly,the path planning in indoor environment is studied.There are two main research directions: one is the combination of reactive behavior and thinking layer behavior planning;the other is to combine global planning with local planning.Therefore,this thesis respectively from the robot behavior mode and algorithm point of view for path planning proposed based on grid method and behavior method combined algorithm and based on PSO-ACO fusion algorithm for path planning simulation.In the simulation of path planning,the indoor environment is rasterized firstly,and transformes the traditional indoor model into a quantized two-dimensional array form.Moreover,it uses the global path-planning algorithm for path design,and divides the path planning into different local paths according to the location of dynamic grid information in the environment,to achieve the effect of path optimization.Localization and tracking in indoor environment is one of the key problems of smart home robot.Based on bluetooth signal,this thesis proposes a localization method and path planning method for indoor environment.The verification results show that the proposed localization algorithm and path-planning scheme have better performance than the existing scheme,and have certain reference value for indoor mobile robot localization and path planning. |
