Research On UAV Indoor Positioning Technology Based On Fusion Of Optical Flow And UWB

At present,the outdoor navigation flight of UAVs based on GNSS/SINS integrated navigation has become mature and has been widely used.However,it is still difficult to accurately locate UAVs in indoor environments with weak satellite navigation signals.In response to this problem,the UAV indoor positioning method based on the fusion of ultra-wideband and optical flow is studied.This paper studies the basic theory of UAV,optical flow positioning algorithm and its improvement,UWB positioning technology,UWB and optical flow fusion method,etc.,and designs the UAV indoor positioning software and hardware system,and conducts many experimental verification.The UAV uses the quaternion method to calculate the attitude,and uses the first-order Runge-Kutta to update the quaternion.Use display complementary filtering to fuse accelerometer and gyroscope data to suppress attitude angle drift.The cascade PID algorithm is used to control the attitude and position.The block matching optical flow method is introduced and optimized,which improves the matching effect when the yaw of the UAV changes.In addition,in view of the multi-channel control of UAVs,two remote control signals,PPM and SBUS,are introduced.Among them,the SBUS signal adopts all-digital coding and conforms to the RS232 protocol.UWB positioning system based on TOA is used to provide indoor positioning for UAVs.In order to solve the problem of clock synchronization in TOF ranging,an asymmetric bilateral two-way ranging method is adopted between the base station and the tag.In view of the impact of non-line-of-sight ranging on UWB positioning accuracy,the cross-area method is used to identify non-line-of-sight base stations to eliminate non-line-of-sight base stations and provide a basis for data fusion.In order to eliminate the UWB's high-frequency noise and the cumulative error of optical flow,UWB and optical flow positioning data are fused.Firstly,the UWB raw data is used to eliminate outliers by the threshold method,and the moving average filter is used for smoothing.The IMU is used to compensate the optical flow raw data to suppress the optical flow measurement error when the UAV is shaking in situ.Then,the weighted average filtering is used to fuse the two kinds of data,and the weights are changed based on the number of line-of-sight base stations,and the weighted average of UWB and optical flow positions is obtained.A Kalman filter is also designed to fuse two kinds of data,using UWB data as the observational measurement,optical flow data as the control variable,setting the target position and the optical flow error as the state variable,iterative calculation is performed,and the position data in the state variable is calculated.Output.Designed the software and hardware system of flight control and UWB base station,and carried out experiments based on this system.In the experiment of circular block matching optical flow method,when the picture rotates more than 1.5°,the 93.3%matching result is better than the traditional square block matching.In the UWB and optical flow data fusion experiment,the position X and Y coordinate jitter after the weighted mean filtering method fusion is within 6.5cm,which effectively suppresses the cumulative error of the optical flow.The position jitter after Kalman filter fusion is within4 cm,which almost eliminates the cumulative error of optical flow.Based on the position coordinates obtained by the fusion,the UAV indoor navigation experiment was carried out.The standard deviation of the X and Y coordinates of the UAV flight trajectory fused by the weighting method is within 7cm,and the standard deviation of the trajectory based on the Kalman filter fusion is within 3cm.The aircraft have completed the task of flying along the preset trajectory,and when the number of UWB base stations is insufficient to provide positioning,they can still rely on optical flow positioning to continue navigation and flight.It achieved the intended effect.