| To complete the navigation task,Tour-guide robot must first solve the problem of accurate navigation positioning.In practical applications,traditional laser radar positioning and magnetic strip navigation positioning have difficulties such as high equipment requirements,great influence by obstacles,and inconvenient laying of magnetic strips,it is urgent to study a precise positioning scheme to meet the navigation and positioning needs of the Tour-guide robot.The positioning error based on the Inertial Measurement Unit(IMU)accumulates rapidly over time,and it needs to be combined with other positioning technologies for position correction;the positioning technology based on Ultra-Wide Band(UWB)has high positioning accuracy,but is susceptible to non-line-of-sight(NLOS)interference.This thesis studies a positioning scheme based on the fusion of IMU and UWB,and carried out relatively systematic research and experiments.It mainly focuses on the characteristics of IMU and UWB when they are positioned alone,the strategy of IMU/UWB fusion,the construction of the experimental platform of the guide robot,and the experiments of the line-of-sight(LOS)and NLOS environments.The main contents of this thesis are as follows:(1)In view of the influence of error accumulation on the positioning accuracy during IMU calculation,analyze the cause of the accumulated error.Through user-defined gyro bias value,perform inertial navigation simulation analysis on the gyro bias with the largest influence factor of the accumulated error in Matlab,and conduct the experiment after error correction according to the analysis results.In the attitude solution experiment,after IMU performs pitch,roll and yaw actions,the attitude error obtained by the solution is less than 1°;In the position solution experiment,the error obtained after the IMU moves for 5.1s is less than 5cm.The experiment shows that the corrected single IMU can maintain high positioning accuracy in a short time.(2)Aiming at the problem that UWB ranging and positioning accuracy is affected by working characteristics and NLOS,the two-way bilateral ranging method is applied to UWB ranging to reduce the ranging error caused by clock offset;Then,the UWB ranging value is corrected by linear fitting to reduce the error caused by the operating characteristics of the UWB;Finally,the traditional least squares algorithm is improved and used for positioning solution according to the weight distribution of distance and ranging quality.The experimental results show that the proposed method can effectively improve the positioning accuracy of UWB in the NLOS environment.(3)In order to further reduce the positioning error caused by NLOS effects on UWB,and study the impact of fusion scheme on positioning performance,IMU/UWB loose and tight combined positioning models are established respectively,and Kalman Filter(KF)is used for data fusion of loose combined model;Aiming at the problem of low fusion positioning accuracy caused by the loose combination positioning model when the UWB is in NLOS measurement,a tight combination scheme based on Extended Kalman Filter(EKF)fusion is proposed,which combines the robot’s uniform motion model to discriminate the NLOS errors.The effectiveness of the algorithm is verified by the Acenna inertial navigation simulation framework.(4)In order to verify the effect of the fusion scheme in this thesis in the indoor environment,a Tour-guide robot experimental hardware platform was built and upper computer software platform for data acquisition and monitoring was developed,and two environments,LOS and NLOS,were set up to carry out experiments.The experimental results show that the two fusion schemes have a good inhibitory effect on the NLOS error.The tight combination scheme combined with the robot motion model has higher accuracy than the loose combination scheme in the NLOS environment,and the positioning root mean square error and the maximum position errors are all less than 25cm,which meets the indoor positioning accuracy requirements of the Tour-guide robot.The research content of this thesis has certain reference significance for the research of indoor mobile robot positioning methods and the improvement of positioning accuracy. |
