High-Precision Synchronization Method For Multi-Sensor And Its Applications In Mobile Mapping System

Multi-sensor integration and synchronization control is one of the key technologies of Mobile Mapping Systems(MMS), which controls various sensors to acquire data synchronously in accordance with a certain "beat" to guarantee all the kinds of data are aligned with time. The high-precision synchronizer is able to reduce the data registration error and improve the surveying accuracy.In recent years, although great progress has been made in the field of multi-sensor integration and synchronization control of MMS, there are still some serious inadequacies: first, there are few accurate time-transfer delay-models, through which we are able to analysis time synchronization accuracy, and estimate the performance of the device; second, how to precisely control multiple sensor to acquire data synchronously is also needed to deeply research; third, there is insufficient analysis of delay of the sensor itself, which is an important factor that affects the accuracy of data registration. In this thesis, an intact set of precious multiple-sensor integration and control methods and theories are proposed to solve these problems, including the establishment of precious time reference, time transfer and its delay model, the multi-sensor synchronization method, and the compensation method for sensors delay. The laser dynamic continuous deflection measurement vehicle is taken as the specific example to verify the presented methods and theories.The innovations of this thesis include:(1) A system which is able to establish and maintain precise time reference is designed. Precise time reference is the basis of multi-sensor synchronization control system. In this thesis, we first analyze the effect of time-misalignment on data registration, and then introduce various methods to establish the precise time reference. Subsequently, we design a high-precision synchronizing system which uses the GPS PPS pulse to rectify the high stability quartz crystal. The system adopts CPLD incorporating with the direct digital frequency synthesis(DDS) to establish the high-precision time reference:A single chip microcomputer is used to correct the DDS phase control-word based on the phase difference outputted by the CPLD, and the edge of the clock pulse outputted by the DDS is aligned with the one of the PPS pulse outputted from GPS. This system takes foll advantage of the high long-term stability of the GPS PPS pulse and the high short time stability of the quartz crystal. As a result, the system is able to maintain the high-precision time reference for the whole surveying process. (2) A multi-sensor synchronization-control method applied in the MMS is proposed. According to the drive source, the synchronization-control modes can be classified into time-driven trigger and distance-driven trigger. And according to the characteristics of sensors, the synchronization-control modes can be classified into active synchronization, passive synchronization and timing synchronization. Based on the detailed analysis of design principles of each mode, this thesis proposes an integrated synchronization-control mode which is applied in the MMS effectively. Furthermore, through analyzing the influences on the time transfer accuracy in the process of synchronization, a mathematical model to estimate the delay-error of time-transfer is established. This model is also able to assist the design of synchronization controller.(3) A method to fast calibrate the delay parameters of the Fiber Optic Gyroscope (FOG) is proposed. In this method, the rigid beam of the laser dynamic continuous deflection measurement vehicle is used as the calibration platform, and a laser doppler velocimetry is used as the test equipment. The kinematics equation of the FOG is established and discretized first, and then the time-delay parameters are solved by the improved least square arithmetic. This method is also able to calibrate the scale factor and the bias of the FOG without additional devices, so it is a simple, fast and high-precision method. Furthermore, two ways to eliminate data registration-error caused by the sensor-delay are proposed, that is, the software delay-offset and the hardware delay-trigger, and their own advantages and disadvantages are analyzed as well. The analysis and experimental results indicate that hardware delay-trigger method is able to eliminate the data registration-error more effectively and holds better generality and higher synchronization accuracy.(4) A multi-sensor synchronization controller for the continuous laser deflection measurement vehicle is designed and implemented. The continuous laser deflection measurement vehicle represents the most advanced technique in the field of the fast deflection measurement. This thesis presents the detailed architecture, principles and ideas of design of the multi-sensor synchronization controller for the laser dynamic deflection measurement vehicle. The multi-sensor synchronization controller is composed of a backboard and a sub-board which are installed in a19inch case, and connected with each other through a CPCI connector. This architecture provides great convenience for disassembly and maintenance. The thesis also discusses how to select device types and design principles for all the circuit modules. Especially, we emphatically analyze the design ideas and design parameters of the main control board and other circuit boards of the synchronization controller, multi-sensor synchronous data acquisition and ego-diagnosis technologies of system faults. Finally, experiments are carried out to test and verify the proposed multi-sensor synchronization controller for the laser dynamic continuous deflection measurement, including evaluating the time reference accuracy, estimating the delay parameters of FOG and time-delay interval of trigger pulse, and measuring the actual deflection.Experimental results show that the precision and accuracy of proposed synchronization-control methods meet the requirements of mobile mapping systems. Additionally, the calibration method for time-delay parameters of the FOG is fast and accurate, and the multi-sensor synchronization controller is able to effectively control multiple sensors to acquire data synchronously. To conclude, the proposed methods, models and designs by this thesis meet the demands of Mobile Mapping Systems for the multi-sensor integration and synchronization control.