| With the vigorous development of deep space exploration,in order to improve the information capacity,resolution,transmission distance and speed,the construction needs of large structures in space are increasingly prominent.Traditional large space structures need to be built on the ground,and then launched into orbit by launch rockets.Due to the influence of bearing weight and volume factors,and the single point failure of thousands of deployment mechanisms,it is very easy to lead to mission failure.Therefore,high-power,high-resolution and large-scale antennas in space will be deployed in the way of in orbit construction in the future.The shape accuracy of large antenna directly determines its performance,and the adjustment and calibration of its shape depend on high-precision geometric measurement.By analyzing the principle,system composition,technical characteristics,measurement accuracy and scope of application of the geometric measurement method of large antennas on the ground,the project plans to use laser radar for global measurement,and the laser radar will be placed in the center of large structures;At the same time,local measurement is carried out by combining photogrammetric technology,and the photogrammetric system traverses the entire large structure with the construction robot.Before collecting multimodal measurement data,in order to improve the measurement accuracy and avoid data redundancy,it is necessary to lay certain control points on the paraboloid,use the conversion accuracy of data registration to screen the layout scheme,and consider the influence of the number of control points,the interval of control points,and the error of added image points on the actual measurement points,so as to finally obtain the optimal layout scheme.The application of multi-sensor measurement devices in large-scale structure measurement is more and more extensive,but most of them are only integrated under the same standard,without systematic research on the basic theory of multi-sensor measurement method planning,data matching and fusion,which leads to poor cooperation and complementarity between sensors,and the advantages of multi-sensor measurement technology are not reflected.The measurement data on the paraboloid is usually approximate to Gaussian distribution or Gaussian distribution,so the Kalman filtering algorithm and the extended Kalman filtering algorithm are selected to estimate the measurement point positions of the photogrammetric system and the lidar measurement system respectively.According to the sampling characteristics,the two kinds of data are received alternately and fused,so that the fused data is closer to the actual position,and the measurement efficiency is also improved. |
