Research On Integrated Location System Based On Lidar And Fiber Optical Gyroscope

With the constant upgrading of industrial structure of our country,the intelligent logistics industry has ushered in a major historical opportunity for development.The intelligent logistics system is also the cornerstone of building a future intelligent factory.As an advanced automatic transport equipment,Auto Guided Vehicle(AGV)plays an important role in the future intelligent logistics system.AGV integrates multiple disciplines,mechanical,microelectronics,computer,control,network communication,artificial intelligence,etc.With its high degree of intelligence,flexible networking capability,good safety and reliability and occuping less space,it is widely used in enterprise production equipment and warehouse automation system upgradeFirstly,the basic principle of the dead reckoning method was introduced.The omnidirectional mobile AGV location algorithm was deduced,and the installation error of orthogonal omnidirectional wheel was analyzed.The calibration algorithm of the installation error was deduced,then the installation error was compensated.The static error model of fiber optic gyroscope was established,calibration algorithm of fiber optic gyroscope was deducted.The fiber optic gyroscope was calibrated on the high precision turntable.The velocity error of incremental photoelectric encoder was analyzed,the speed measurement algorithm was improved.The accuracy of velocity measurement was improved.The position error of the dead reckoning system was accumulating over time,so other position method was needed to assist location.Because of the angular drift error of fiber optic gyroscope and the displacement accumulation error existed in the encoder,so the position error of the track reckoning system was accumulating,and other position methods were needed for auxiliary calibration.Then,the lidar SLAM algorithm was studied.The probability updating process of grid map was analyzed.The HECTOR_SLAM algorithm was studied emphatically,it mainly included three aspects: Map Access,Scan Matching,Multi-Resolution Map Representation.Since the position error of lidar did not diverge with time,the position error of the track reckoning system was corrected.Then,the basic principle of complementary filter was introduced.The track reckoning system had high position accuracy in short time and the dynamic performance of the gyroscope was good,however,the position error continued to accumulate.The lidar SLAM position error was relatively stable,its error was not divergent,but the position accuracy was relatively low and the real-time performance of lidar was poor.The yaw angle complementary filter and position complementary filter were designed according to the respective frequency characteristics of the track reckoning system and the lidar system.The filter can effectively suppress the accumulated error in the track reckoning system and improve the positioning accuracy of the systemFinally,the hardware and software platforms of position system were built.The track reckoning algorithm and HECTOR_SLAM algorithm were vertified by experiments,and the corresponding error were analyzed.The complementary filter were designed according to the yaw angle and the position information of track reckoning and lidar,the effectiveness of the complementary filter were proved.