Research On The Method Of Improving Accuracy Of Photoelectric Tracking System Based On Multi-source Data Fusion

Photoelectric tracking system is the key equipment of tracking measurements, which is used for tracking and measuring of shooting range, landing guidance of carrier-based aircraft, assisting the laser weapon’s attack, spacecraft launching test and other tasks. The rapid development of military equipment and aerospace engineering puts forward higher requirements for the capturing and tracing ability of photoelectric tracking and measuring system. Research and improvement the accuracy of photoelectric tracking system has very important significance for meeting the target acquisition of high speed and maneuverability, ensuring the tracing and measuring. It is also the promise ensure to correction accuracy and the inevitable requirement of high-quality imaging measurements.In this paper, the study object is a single measurement of optical tracking system with three kinds of information sources which are visible light, infrared and guidance information. And to improve the tracking accuracy of servo control system, the error correction algorithm for information fusion system, tracking filtering algorithm and state estimation fusion algorithm are deeply discussed, which bases on multi-data source information fusion theory and coaxial tracking technology to improve. The concrete research content is as follows:First, the function and structure of two-axis photoelectric measurements system and tracing mode of servo control system are analyzed. By summarizing the principle, realization form and its role in improving tracking accuracy of composite control, the control method of multiple data source information fusion theory with coaxial tracking technology is proposed, and several key technology research points are obtained that coaxial tracking information fusion system must to complete.Second, several key issues need to be solved about system error correction process are in-depth study. On the basis of the existing error correction model, for systematic errors in the nonlinear factors, the system error correction algorithm based on extreme learning machine is proposed and verified by using simulation experiments. The results show that the proposed algorithm can effectively suppress interference of nonlinear factors and improve the correction.Third, the causes and effects of miss distance lag are analyzed. For the issues of non-integer frame about miss distance lag, the two steps stochastic delay measurement tracking filter algorithm is proposed. By comparing the simulation results of the new raised algorithm and existing algorithms, the consequence shows that the algorithm proposed in this paper can be effectively improve the estimation accuracy of the target state than the existed algorithms. At the same time, for the model inaccuracies of optical measurement system in actual work, two-step stochastic delay measurement strong tracking filtering algorithm is proposed and two-step stochastic delay measurement strong tracking Gaussian filter is deduced which is realized by Kalman filter.The simulation result shows that the proposed algorithm can further improve tracking accuracy and has more robust.Forth, coaxial tracking information fusion system is studied. In this section, federal filtering algorithm are derived and analyzed. Also coaxial track composition and implementation of information fusion system are detailedly analysis. For the problem of estimate accurately reduced caused by angle and distance measuring value Delay characteristic inconsistency, two-step random delay measurement strong tracking sequential filter algorithm is proposed. The asynchronous incoordinate interval information fusion algorithms proposed in this paper can effectively deal with the problem of long integration period caused by the difference of multi-source data sampling period and the problem of fusion system accuracy decrease. The simulation results show the co-axis tracking information fusion system can achieve accurate estimation of the target angular position and velocity. And compared with the equivalent compound control system, it can further improve the accuracy of the tracking servo control system.Lastly, coaxial tracking information fusion system is fully applied in the servo control system and the experiments of optical dynamic target and outfield tracking are respectively conducted on the experimental platform constructed by existing facilities. Also, the average time delay of the photoelectric detector is tested. Experimental results show that the coaxial tracking information fusion system studied in this paper is effective and improve the tracking accuracy of photoelectric tracking system.