Research On Localization Algorithm Of Automatic Guided Vehicle

Medical and health services guarantee the safety of people’s lives.Major hospitals produce a large amount of medical waste every day.Their hazards far exceed those of ordinary household waste.If misoperation occurs during the treatment process,it will cause serious harm.This paper studies and designs an Automated Guided Vehicle(AGV)for medical waste transportation,which can replace medical staff to transport medical waste generated during treatment to a special storage place,avoid direct contact with medical waste,and protect medical care.The safety of personnel also effectively reduces the degree of harm of medical waste.When the medical waste transportation AGV moves,the power comes from a rechargeable battery,the main control is an embedded industrial control host,and it is equipped with a large number of sensors,such as encoders,inertial measurement units(inertial measurement units,IMU),lidars and cameras,etc.The sensor can detect the working hours of the medical waste transportation AGV and the external working environment,so that it can perform tasks.In the work of medical waste treatment,the hospital’s indoor space is narrow,dense and variable lighting and other environmental conditions,it is necessary for the medical waste transportation AGV to have an accurate positioning function to meet the requirements of accurate navigation and transportation of medical waste.However,the amount of information in a single sensor is small and the measurement noise is large,which causes the gradual error of the medical waste transportation AGV to become larger and larger during the operation process,the estimated pose is inaccurate,and the camera is easily affected by the lighting conditions and cannot be robustly estimated.Posture.A method of pose tracking is used here to study and analyze the robust extended Kalman filter(Extended Kalman Filter,EKF)multi-sensor fusion algorithm,and use the medical waste transportation AGV as the platform to solve the problem of inaccurate positioning.The improvement is to fuse the sensor data of the encoder,IMU and lidar to estimate the accurate pose.Collected in the hospital environment(external corridor of the laboratory),the results show that the motion trajectory of the medical waste transportation AGV after processing by the EKF algorithm is closer to the true value than the calculation result of a single sensor.Evaluation Criteria 1,the root mean square error of the abscissa of the EKF trajectory is 0.0682 m,the root mean square error of the ordinate is 0.0869 m,and the root mean square error of the attitude angle is0.0174rad;evaluation criterion 2 shows that the transverse mean square of the EKF trajectory The root error is 0.0728 m,the vertical root mean square error is 0.0831 m,the Euclidean distance root mean square error is 0.1104 m,and the poses in the EKF trajectory are all accurate and better poses.The EKF algorithm can accurately estimate the pose of the medical waste transportation AGV,which has greatly improved in terms of accurate positioning,so that it can be competent for the work of medical waste transportation.