Research Of Open-space Camera Positioning Based On UAV Ranging

With the rapid development of the Internet of Things,more and more sensing devices are being installed and deployed in real-world environments.As the most commonly used sensing device,the need for accurate positioning of cameras in the real world is becoming more and more urgent.When cost and accuracy are not considered,GNSS positioning systems,base station positioning,or other solutions can easily achieve open-space camera positioning.However,in open space camera positioning,a large number of cameras to be positioned makes manually assisted positioning infeasible,and the high demand for positioning accuracy makes ground-based base station positioning infeasible.Open-space camera positioning requires low power consumption and high positioning stability,while economic and practical considerations are needed,and current ranging schemes can hardly meet the above requirements.To solve the above problems,this thesis proposes a camera positioning system based on UAV LORA ranging,with the help of UAV position and the distance between the UAV and the target device to be positioned as measured by the UAV at multiple positions,the open space camera position estimation is obtained with the help of ranging data with the help of positioning algorithm.Since the localization error is greatly affected when the basic localization algorithm is directly adopted,this thesis proposes a localization algorithm combining non-visual measurement point identification and PSO＿LS(Particle swarm optimization robust Least Square Estimation),which reduces the localization error while using fewer ranging groups.The localization error is reduced by using fewer ranging groups,making open space camera localization budget-friendly.To verify the practicality of the above positioning system,the distance between the UAV equipped with LORA ranging device and the target device to be positioned is measured at several known locations,and all the ranging data are stored in the ranging device on the UAV,and after the data collection is completed,the ranging data are read using the serial port and the above positioning algorithm is applied for positioning estimation.The experimental results show that the non-line-of-sight measurement point identification algorithm can filter the nonline-of-sight measurement points of the UAV more effectively under the premise of using only100 sets of repeated measurement data,and the PSO＿LS algorithm can filter part of the nonline-of-sight measurement data to obtain more accurate positioning results without using a large amount of repeated measurement data.In addition,comparing with other algorithms,it can be seen that the positioning system in this paper is relatively good in the actual measurement environment where the non-line-of-sight error is more serious.The experiment shows that the positioning error is reduced to 1.81 m without occlusion and 2.11 m with occlusion,which reduces the positioning cost while the accuracy meets the current positioning requirements.