Research On Indoor Contactless Distribution Technology Based On WiFi,Bluetooth And SLAM Composite Positioning

With the arrival of covid-19 pneumonia,the delivery of express terminals has been greatly affected.Contactless distribution has emerged as a new form of disease,which has made great contributions to the prevention and control of epidemic diseases.Meanwhile,the logistics industry has been greatly affected by the economic downturn.However,at this stage,contactless distribution mostly stops outdoors,and contactless distribution has not been widely used indoors.At present,there are still some problems in indoor contactless distribution,such as high labor cost,low distribution efficiency,poor positioning accuracy,and path planning can not adapt to the indoor complex environment.To solve these problems,this paper studies and designs an indoor contactless distribution scheme.The specific research work is as follows:1.Aiming at the problem of poor positioning accuracy of indoor contactless distribution,firstly,a WiFi fingerprint positioning scheme is studied and designed.For some unsatisfactory APS in WiFi access points,this paper proposes an AP selection strategy.In the offline stage,the AP with stable signal source is selected by calculating the loss rate,and then the standard deviation of each AP in each position is calculated to screen the AP with small fluctuation of received signal strength value.In the online stage,KNN and WKNN are matched and positioned for the RSSI collected by the test points,and the Manhattan distance is introduced to replace the Euclidean distance.Finally,the positioning results of the test points are obtained to solve the problem of poor positioning accuracy of indoor contactless distribution.2.Aiming at the problems of easy interference and poor robustness of WiFi fingerprint location,this paper proposes a composite location method,which combines visual inertia slam technology and Bluetooth signal to provide additional auxiliary information for the location algorithm and further improve the location accuracy of indoor contactless distribution.At the same time,the indoor environment is complex and the contactless distribution system is easy to be disturbed.When the visual odometer is not sure whether the image is matched,the multi-source information is used to verify whether the location is matched,and finally a more accurate positioning result is obtained,which further improves the positioning accuracy of the indoor contactless distribution system.3.In this paper,a method of optimizing the path of the robot in the complex environment based on the experience of the tree is proposed,which can enhance the convergence of the algorithm in the unknown environment,The optimal path is obtained through continuous trial and error learning,which solves the problem of unreasonable path planning in indoor contactless distribution.