Celestial navigation is an important type of autonomous navigation technology which could be used as an alternative to Global Navigation Satellite Systems (GNSS) when a vessel is at sea. After several centuries of development, a variety of astronomical vessel position (AVP) determination methods have been invented, but the basic concepts of these methods are all based on angular observations with a device such as a sextant, which has disadvantages including low accuracy, manual operation, and a limited period of observation. This paper proposes a new method that utilises a fisheye camera to image the celestial bodies and horizon simultaneously. Then, we calculate the obliquity of the fisheye camera's principal optical axis according to the image coordinates of the horizon. Next, we calculate the altitude of the celestial bodies according to the image coordinates of the celestial bodies and the obliquity. Finally, the AVP is determined by the altitudes according to the robust estimation method. Experimental results indicate that this method not only could realize automation and miniaturization of the AVP determination system, but could also greatly improve the efficiency of celestial navigation.

Astronomical positioning and orientation are important components of astronomical geodesy tasks, and they could also be used for real-time celestial navigation. In traditional methods, astronomical positioning and astronomical orientation can only be implemented separately, and all the observations with different levels of precision are given the same weight. To change this, a unified astronomical positioning and orientation model based on robust estimation was developed. First, the problem that the coordinate rotation parameters are difficult to calculate directly was solved by using the Rodrigues matrix. Consequently, the astronomical longitude, latitude, and azimuth can be calculated at the same time. Second, as a result of the introduction of a robust estimation method, specific weight is matched to observations with corresponding precision when calculating the parameters, and thus more accurate astronomical longitude, latitude, and azimuth valuation can be calculated simultaneously. Experiments indicate that the unified model was able not only to achieve astronomical positioning and orientation simultaneously but also improve the accuracy of final results.

Celestial navigation is an important means of maritime navigation; it can automatically achieve inertially referenced positioning and orientation after a long period of development. However, the impact of different accuracy of observations and the influence of nonstationary states, such as ship speed change and steering, are not taken into account in existing algorithms. To solve this problem, this paper proposes an adaptively robust maritime celestial navigation algorithm, in which each observation value is given an equivalent weight according to the robust estimation theory, and the dynamic balance between astronomical observation and prediction values of vessel motion is adjusted by applying the adaptive factor. With this system, compared with the frequently used least square method and extended Kalman filter algorithm, not only are the real-time and high-precision navigation parameters, such as position, course, and speed for the vessel, calculated simultaneously, but also the influence of abnormal observation and vessel motion status change could be well suppressed.

Heading angle is a vital parameter in maintaining a vessel's track along a planned course and should be guaranteed in a stable and reliable way. An innovative method of heading determination based on a fisheye camera, which is almost totally unaffected by electromagnetism and geomagnetism, is proposed in this paper. In addition, unlike traditional astronomical methods, it also has a certain degree of adaptability to cloudy weather. Utilising the super wide Field Of View (FOV) of the camera, it is able to simultaneously image the Moon and the horizon. The Moon is treated as the observed celestial body and the horizon works as the horizontal datum. Two experiments were conducted at sea, successfully proving the feasibility of this method. The proposed heading determination system has the merits of automation, resistance to interference and could be miniaturised, making application viable.

The fish-eye star sensor with a field of view (FOV) of 180° is an important piece of equipment for attitude determination, which improves the visibility of stars significantly. However, it also brings the star identification (star-ID) difficulties because of imprecise calibrations. Thus, a fish-eye star-ID algorithm supported by the integration of the precise point positioning/inertial navigation system (PPP/INS) is proposed. At first, a reference star map is generated in combination with the distortion model of the fish-eye camera based on the position and attitude information from the PPP/INS. Then the star points are extracted in a specific neighbourhood of the reference star points. Subsequently, the extracted star points are individually tested and identified according to angular distance error. Finally, the real-time precise attitude is determined based on the star-ID results. Experimental results show that, 270–310 stars can be identified in a fish-eye star map with an average time of 0.03 s if the initial attitude error is smaller than 1.5° and an attitude determination accuracy better than 10″ can be achieved by support from PPP/INS.