| With the continuous application and development of the marine technology, the accuracy requirements of the navigation information of the ship also get higher and higher. The attitude determination technology of the ship has become a key technology in the navigation field.Various attitude sensors are able to complete the job. Among these sensors, the inertial navigation system is widely used because of its advantages in precision and stability. The star sensor has also attracted more and more attention as a kind of attitude sensor with high precision. It can provide the attitude information with high precision about the carrier with the angle of second level. With the continuous developments and applications of the various integrated navigation systems, based on the two kinds of sensors above, the combination of INS/star sensor attitude determination system is getting more attention and extensive research due to its high accuracy, stability and autonomy. Based on the basic theory of the inertial navigation and the star sensor, this paper discusses the method of the data fusion and the error compensation scheme of the gyroscope based on the online calibration.Considering the actual working environment of the star sensor, when it can not effectively acquire and identify the star map due to the weather conditions, it will fail to work in a short period of time. In this case, the method based on the fuzzy reasoning model is adopted. The scheme is to train the ANFIS model when the system is in the normal working condition, then change to the alternate navigation alternate when the star sensor fail to work,so as to estimate the error of the navigation parameters and guarantee the navigation process.The error compensation is an effective mean to improve the precision of the integrated navigation system, and the error calibration is the precondition of error compensation.Different from the past way which making the data fusion directly on PC, this research is based on the actual situation when the star sensor is fixedly connected to the gyroscope directly, during this time we not only need to consider the internal error sources inside the inertial navigation system, the position errors newly introduced between the inertial system and the star sensor are also need to be concerned about, set a new expanded-dimension error model of the gyroscope, the error is real-time estimated by the method of state estimation and compensated then, the precision of the navigation system will be greatly improved.Finally, based on the actual dynamic experiment, I design the interactive pages, the online calibration and the data fusion method based on ANFIS is applied to the system, the effectiveness of the two algorithms in improving the precision of the system is verified through the analysis of actual data.Through the research of data fusion and the error compensation technique based on the inertial navigation system and the star sensor, further discussion is maked based on the basic combination,it has certain theory value and a guiding role to the practical application. |
PDF Full Text Request