A method of GPS-denied celestial navigation and system implementing the same

Abstract

A celestial navigation system (100) for determining position and heading direction of a vehicle comprising an imaging submodule (101), a processing means (400) and an onboard timepiece is disclosed. Imaging submodule (101) comprises at least one image acquisition means (102 on a pivoted support means (103) movable on at least two orthogonal axes and comprising rotary encoders for each axis, said processing means (400) comprises an ephemeris storage medium (401), and configured to measure parallactic angle and angular speed of a celestial object (500) based on data provided by image acquisition means (101), and determine a current heading and a current position of a moving vehicle.

Description

[0001] A METHOD OF GPS-DENIED CELESTIAL NAVIGATION AND SYSTEM IMPLEMENTING THE SAME

[0002] Technical Field of the Present Invention

[0003] The present invention relates generally to methods of celestial navigation used for determining the position of various vehicles on the globe, and more particularly to systems and methods of navigation using celestial objects in GPS-or-GNSS-denied or degraded environments utilizing unique observational data from said objects.

[0004] Background of the Present Invention

[0005] Precise location of celestial objects at any given moment and from anywhere on Earth can be determined with remarkable accuracy. These celestial bodies may include prominent stars, planets, the Sun, and the Moon, all of which follow predictable paths mapped with advanced astronomical data. Achieving such accuracy in determining their positions relies primarily on two key factors: geographical coordinates of the observer, such as latitude and longitude, which need to be known within a margin of 1 to 2 kilometers, and the accurate, exact date and time of observation. This precise alignment of spatial and temporal data is known to reliably enable effective celestial navigation, astronomical studies, and various applications in science and technology.

[0006] An inertial measurement unit (IMU) is an advanced device designed to measure and report the specific force, angular velocity and orientation of a body in space. This functionality is achieved through the integration of three primary components: accelerometers, which measure linear acceleration; gyroscopes, which detect rotational motion; and magnetometers, which serve as a reference for heading by measuring magnetic field direction. IMUs are indispensable in various applications such as navigation and maneuvering ofaircraft, such as those equipped with an attitude and heading reference system, as well as unmanned aerial vehicles (UAVs), spacecraft, satellites, and landers. In recent years, advancing technology has enabled the development of GPS-enabled or GPS-assisted autopilot enhancing the versatility of navigation systems. By integrating an IMU, a GPS receiver can be furthered into attitude and heading reference systems (AHRS) which can maintain functionality even in environments where GPS signals are unavailable or unreliable, such as tunnels, building interiors, or in areas affected by electronic interference. AHRSs are crucial improvements over pure GPS systems in that GPS in that for an airborne vehicle, GPS may provide location information, but cannot provide information as to whether the vehicle is level, nose-up, or tilted to the side. AHRS provides this orientation (attitude) data, which allows for determining attitude and heading of the vehicle as it moves. In navigation systems the type used on sea vessels, airplanes or other vehicles, combining GPS data integrated within an AHRS allows for full awareness of both position and orientation, which is especially pertinent when precise navigation is required for tasks like autonomous flight, precise landings, or tracking movements through GPS coordinates and orientation in 3D space.

[0007] An IMU operates by detecting linear acceleration along one or more axes with accelerometers and measuring rotational rates along the same axes with the aid of gyroscopes. Many IMUs also incorporate magnetometers to improve orientation accuracy, and thus leverage benefits of AHRS systems as well. A typical configuration includes one accelerometer, one gyroscope, and one magnetometer for each of the three principal axes —pitch, roll, and yawproviding a comprehensive understanding of the movement and orientation of a body in three-dimensional space. As such, IMUs are critical for navigation, control, and stabilization systems.

[0008] Inertial Measurement Units (IMUs) are frequently integrated into InertialNavigation Systems (INSs), which leverage raw inertial measurement data to calculate critical navigational parameters such as attitude, angular velocity, linear velocity, and position relative to a global reference frame. IMU or generally INS serves as the foundation for the navigation and control systems of a wide range of commercial and military vehicles, including crewed aircraft, missiles, ships, submarines, and satellites. INSs also play a vital role in the guidance and operation of unmanned systems, such as unmanned aerial vehicles (UAVs), unmanned ground vehicles (UGVs), and unmanned underwater vehicles (UUVs), where precision and reliability are essential. AHRS can utilize data from IMUs to compute a vehicle's attitude and heading relative to magnetic north. The sensors within an IMU provide continuous measurements of acceleration and angular velocity, enabling a computer to determine the craft's position and movement through a process known as dead reckoning. By tracking changes in velocity and orientation over time, dead reckoning allows for position estimation even in environments where external signals, such as GPS, are unavailable or unreliable. This capability makes INSs indispensable tools for autonomous navigation, enhancing the functionality of vehicles operating in diverse and challenging conditions, including underwater, underground, and in space. IMUs can be integrated into a GPS-based autonomous navigation system or vehicle tracking system, giving the system a dead reckoning capability and the ability to gather additional data about the momentary speed, turn rate, heading, inclination and acceleration of the vehicle.

[0009] Another solution known to be utilized in the art is a digital / virtual sextant. A sextant is a precision navigation instrument using double reflection whereby the angular distance between two visible objects is measured. Its primary purpose is to measure the angle, or altitude, between a celestial object (e.g. the Sun, Moon, or stars) and the horizon. This process, commonly referred toas sighting the object, is fundamental to celestial navigation. By combining the measured angle with the exact time of the observation, navigators can calculate a position line on a nautical or aeronautical chart, aiding in determining their precise location. Sextant is most commonly used to observe the Sun or other celestial bodies to determine latitude, a critical component of navigation prior to the advent of modern GPS technology, in addition to measuring a lunar distance between the Moon and another celestial body, such as a star or planet. This measurement allows navigators to calculate Greenwich Mean Time (GMT), which is essential for determining longitude.

[0010] A major disadvantage of using inertial navigation systems is that they typically suffer from accumulated error. Since the guidance system is continually integrating acceleration with respect to time to calculate velocity and position, any measurement error, however small, will be accumulated over time notwithstanding the constant systematic errors known as sensor bias. Furthermore, GNSS information is also error-prone and easily spoofable. In the cases where celestial navigation is calculated by hand, the best measurements still have a sensitivity between 1 and 2 nautical miles, a process which is open to human error and fundamentally insufficient for extremely location-sensitive tasks and applications such as guidance and operation of unmanned systems, e.g. unmanned aerial vehicles (UAVs), unmanned ground vehicles (UGVs), and unmanned underwater vehicles (UUVs).

[0011] While it is possible to extract and utilize virtual horizon information using inclinometers, IMUs or INSs, this comes at a cost of limited accuracy. For higher accuracy, systems need to integrate higher-end and expensive technologies such as fiber optic gyroscopes, in turn decreasing the economic feasibility. The most cost-effective and high-accuracy approach is generally considered to be an integrated, zoom-capable image acquisition means, which can directly extract the horizon at sea level. Star-trackers used in satellite technologies arealso costly and bulky devices, which calculate celestial object layouts within observable conditions using plate solving. This approach is also deemed suitable for high-altitude crafts wherein high visibility of celestial objects such as stars is also required. Nighttime conditions allowing for stars to be identified and observed accurately also has the additional challenge of limited amounts of light reaching the observer with sometimes quite narrow angular quantities.

[0012] Solutions in the art utilizing digital systems or approaches require very high-precision angular encoders, while also using advanced next to accurate determination of the horizon, such as a true horizon observable by an aircraft. For certain applications, measurements need to be collected from at least 5, or in some cases more, individual celestial objects. As a general rule of thumb, the more the number of observed objects, the greater the accuracy in determining the position. However, as in the case of the synthetic digital sextant, in cases where a cloud cover is present or in adverse weather conditions, horizon cannot be directly calculated, forcing reliance on an artificial horizon which directly affects the precision of the system. Concurrently, there exists a need for reliable and accurate methods of navigation while remaining structurally and economically feasible, all the while retaining a manageable error margin.

[0013] US20170131096A1 discloses a technology for determining a position of a platform. According to this teaching, a location of a horizon line can be determined using a sensor onboard the platform. One or more celestial objects in the sky can be detected using the sensor onboard the platform. Differential angular measurements between the horizon line and at least one of the celestial objects in the sky can be determined over a duration of time. The position of the platform can be determined based on the differential angular measurements between the horizon line and the celestial objects.US2024230341A1 discloses a dead reckoning based celestial navigation system (CNS) designed for determining position of a vehicle in GPS denied or degraded environment by imaging celestial objects and measuring vehicle ground speed, attitude, and time. Vehicle position is calculated by a processor using dead reconning navigation algorithm and heading measurements from celestial sensor, ground speed measurements from ground speed sensor, pitch and roll measurements from IMU, and time from the onboard clock.

[0014] US2015042793A1 teaches celestial compass with a sky polarization feature including an inclinometer, a camera system for imaging at least one celestial object and a processor programmed with a celestial catalog providing known positions at specific times of at least one celestial object and algorithms for automatically calculating target direction information based on the inclination of the system as measured by the inclinometer and the known positions of at least one celestial object as provided by the celestial catalog and as imaged by the camera. Preferred embodiments include backup components to determine direction based on the polarization of the sky when celestial objects are not visible.

[0015] Objects of the Present Invention

[0016] Primary object of the present invention is to provide a robust system and method of celestial navigation in GNSS-denied settings and environments.

[0017] Another object of the present disclosure is to provide a system and method of celestial navigation highlighted in that it is cost-effective and offers a modest payload.

[0018] Another object of the present disclosure is to provide a system and method of celestial navigation based on obtaining parallactic angle and vertical-horizontal angular speed components of at least one celestial object based on capturedimages.

[0019] Another object of the present disclosure is to provide a system and method of celestial navigation optionally based on obtaining at least an altitude and a set of parallactic speed components based on captured images.

[0020] Another object of the present disclosure is to provide a system and method of celestial navigation optionally based on obtaining limb brightness and position angle of a celestial object based on captured images.

[0021] Another object of the present disclosure is to provide a system and method based on obtaining polarization information of a celestial object for celestial navigation.

[0022] Another object of the present disclosure is to provide a system and method based on obtaining sunspots on the Sun surface to determine parallactic angle for celestial navigation.

[0023] Brief Description of the Present Invention

[0024] The disclosed invention primarily aims to introduce a robust system and method for celestial navigation in environments where Global Navigation Satellite System (GNSS) signals are unavailable or unreliable. It is designed to address the challenges posed by compounding errors inherent in traditional approaches using celestial navigation systems while utilizing enhanced technologies and legacy methodologies in the sense of celestial navigation. The invention includes an image acquisition system, featuring at least one camera, coupled with a processing unit capable of analyzing celestial object data from captured images. This system is cost-effective and provides denser information compared to the complexity required to implement similar solutions, making it suitable for deployment in a variety of applications where payload constraintsand operational efficiency are critical.

[0025] Method and system utilizing the method as disclosed according to the present teaching offers seamless functionality under suitable visibility conditions, even in scenarios where consistent and uninterrupted GNSS data cannot be relied upon. To achieve this, disclosed invention leverages the use of well-understood and easily identifiable celestial objects, such as the Sun, the Moon, and a selection of stars and planets. During daytime, the Sun and Moon are particularly advantageous due to their high visibility and distinctive features, enabling reliable celestial navigation under typical conditions. Said distinctive features are utilized to obtain unique information pertaining to the celestial body in question.

[0026] The innovative method of the disclosure further incorporates unique characteristics of these celestial objects to enhance navigation precision. Properties such as parallactic angles, angular velocities, and light polarization data can be integrated into the navigation and localization process, providing an advanced framework for positional accuracy. Furthermore, in several embodiments, the system and method can also use properties such as bright limb angle, altitude, parallactic speed components. The system and method bring about a significant reduction of operational and observational requirements typical of a system, by often limiting them to a few celestial objects -or, in some cases, just one. As such, several constraints for the size and complexity of such a celestial navigation system can be further streamlined and deployed on autonomous and manned vehicles alike, with greater convenience.

[0027] The system's considerations ensure operability in a diverse set of conditions, offering an effective alternative to purely or heavily GNSS-dependent technologies. It holds promise for a wide range of applications, includingmaritime, aerial, and space navigation, particularly in remote or contested environments where traditional navigation systems are compromised or unavailable. By combining legacy celestial navigation principles with modern image processing capabilities, the disclosed invention offers a strong and robust outcome in achieving precise, reliable, and efficient navigation solutions.

[0028] Brief Description of the Figures of the Present Invention Accompanying drawings are given solely for the purpose of exemplifying a method of GNSS or GPS-denied celestial navigation method and a system using said method, whose advantages over prior art were outlined above and will be explained in brief hereinafter.

[0029] The drawings are not meant to delimit the scope of protection as identified in the claims nor should they be referred to alone in an effort to interpret the scope identified in said claims without recourse to the technical disclosure in the description of the present invention.

[0030] Figure 1 demonstrates a system architectural diagram of the celestial navigation system according to an embodiment of the present invention.

[0031] Figure 2 demonstrates a representational flow diagram of the celestial navigation method according to an embodiment of the present invention. Detailed Description of the Present Invention

[0032] 100) celestial navigation system

[0033] 101) imaging submodule

[0034] 102) image acquisition means

[0035] 103) pivoted support means

[0036] 104) sun polarization detection means

[0037] 200) inertial measurement unit300) inertial navigation system

[0038] 400) processing means

[0039] 401) storage medium

[0040] 500) celestial body / object

[0041] The disclosed invention sets forth a reliable method and system for celestial navigation, specifically for scenarios where widely utilized Global Navigation Satellite System (GNSS) signals are either unavailable or unreliable. The system set forth according to this teaching effectively addresses the limitations such as compounding errors commonly associated with traditional celestial navigation methods and technologies, while leveraging established technologies and a streamlined celestial observation approach. By doing so, it ensures enhanced precision and dependability in diverse operational environments.

[0042] At the core of this invention is a celestial navigation system (100) comprising an imaging submodule (101) that utilizes at least one image acquisition means (102), such as a camera to capture detailed visual representations of celestial objects. This imaging submodule (101) is paired with a specifically configured processing means (400) capable of analyzing the acquired data to extract critical navigational information by applying image processing algorithms to the visual information capturable by said imaging submodule (101), and comparing obtained information with catalogued data pertaining to various celestial bodies (500). Synergy between these components allows the system to accurately determine position and heading using celestial references. Additionally, the invention also offers the advantage of practical redundancy while remaining comparatively quite cost-effective and lightweight. These features make it an ideal solution for a wide range of applications where payload restrictions and operational efficiency are key considerations.Furthermore, disclosed method and system implementing said method offers a modest payload and adaptability, making it suitable for deployment across various platforms and vehicles, including unmanned vehicles, spacecraft, and maritime vessels. Its reliance on unique and well-documented properties of celestial bodies and navigation using said properties not only offers a robust alternative to GNSS but also ensures resilience in scenarios such as electronic warfare, dense urban environments, or remote regions where GNSS signals may be obstructed or deliberately jammed. By combining imaging and digital signal processing technologies with celestial navigation approaches limited to a narrow and reliable set of objects, contemporary navigation challenges are addressed, delivering a versatile, reliable, and efficient solution for modern navigational needs.

[0043] Techniques and approaches known in the art require obtaining measurements and data from at least five, but generally more, celestial objects, which are in many applications selected from a group of stars. In most conditions, light reflected from the surface of the Moon proves to be brighter than that emitted by most stars, which detrimentally affects the visibility of stars suitable for celestial measurement for navigational purposes. In limited settings, celestial observations can be restricted to altitude and azimuth measurement based relative position vectors of the Moon and the Sun only, which greatly limit the accuracy.

[0044] Disclosed invention takes advantage of the utility of unique celestial observations for a variety of conditions concerning prominent celestial objects (500). Unique observations based on a given set of preliminary coordinate and time information are considered to enable location determining with great accuracy, said observations include object altitude, angular speed components at time delta-t, polarization information, parallactic angle and parallactic speed. At least a group of unique observations selectable from the aforementionedgroup comes to the forefront as the strength of the disclosed invention's method.

[0045] In certain conditions, calculating a distance to the horizon can be challenging. An example is when there exists an increased rate of evaporation or prevalence of aerosols at sea, or failures of aerial vehicles of extracting a horizon in overcast weather conditions, even when at an altitude past the meteorological boundary layer. Disclosed invention uses celestial object (500) altitude, angular speed components such as vertical and horizontal speed at time delta t (At), polarization information, unique object trajectories, parallactic speed and angle to compensate for the failure of establishing a horizon. For this purpose, in some embodiments, Kalman filter can be configured such that the method compensates for an optimum, predetermined threshold for accuracy.

[0046] A celestial navigation system (100) according to the disclosed invention, in at least one embodiment, comprises a processing means (400) in the form of a processing unit that is configured to utilize Ephemeris data retrievably stored in a storage medium (401) and image processing. Said system further comprises an onboard timepiece and an imaging submodule. Imaging submodule (102) comprised by the system may, according to various embodiments, further comprise at least one image acquisition means for obtaining images of celestial objects, which can be structurally associated with a pivoted support means (103) configured to support motion on at least two orthogonal axes, optionally on three orthogonal axes: One such pivoted support means (103) can be realized in the form of a three-axis gimbal specifically built to stabilize and allow movement in three independent, mutually perpendicular directions: pitch (tilt), roll, and yaw (pan). Each of the axes of a three-axis gimbal can be equipped with its own motor and sensors to counteract unwanted movements and vibrations, providing smooth and stable operation as well as rotary encoders configured to operate to apredetermined angular resolution, preferably on the order of 1 arcsec or less. Orthogonal axes of the pivoted support means (103) may, according to certain embodiments of interest, associated with multiple dedicated image acquisition means (102), such as an embodiment where one image acquisition means (102) is provided per axis of independent motion. Image acquisition means (102) according to certain embodiments can be selected from a group of cameras including, a CMOS camera, a visible-band camera, a polarization image sensor, a shortwave infrared camera (SWIR), a fisheye lens camera, a wide field-of-view camera, a narrow field-of-view camera. Furthermore, said image acquisition means (102) can also be selected from a group of imaging means comprising thermal, daytime and / or event-based image sensors.

[0047] Disclosed invention's celestial navigation system (100) can be arranged to obtain a horizon and / or a local vertical information from subunits comprised thereby, such as a star tracker or an IMU (200). Furthermore, in certain embodiments, said system (100) can further comprise a horizon calculation means which is characterized by a horizon finding camera based on the information supplied by which a horizon can be calculated. Said horizon finding camera, according to a particular embodiment, is specifically arranged to acquire the image of the horizon, whereby orientation and a local vertical can be determined. A local vertical and a horizon can be determined in several ways which will be explained below with reference to several embodiments.

[0048] According to one embodiment, system (100) comprises a horizon calculation means comprising a horizon finding camera, such as narrow field-of-view camera such as a zoom-capable camera or a binocular vision camera, and a pivoted support means such as a gimbal comprising at least an encoder for movement along at least one axis, preferably a vertical axis. Corresponding with each image obtained by the image acquisition means (102), said horizon calculation means comprising a horizon finding camera is configured to trackthe horizon in a synchronous manner. Based on images captured by a horizon calculation means and an image acquisition means (102) as well as an encoder information, said system (100) is configured to determine the attitude of the vehicle.

[0049] According to one embodiment, system (100) comprises a horizon calculation means comprising one wide field-of-view camera. Horizon calculation means arranged in this manner can obtain information as a horizon sensor or a local vertical information sensing means, whereby said stationary horizon finding cameras are spatially configured to cover a full 360 degrees, whereby a difference between radiation properties of light coming from different portions of sky, such as the horizon, the space, the Earth. Said cameras can be IR cameras.

[0050] According to one embodiment, system (100) comprises a horizon calculation means comprising a wide field-of-view camera, whereby visual information representing at least multiple celestial bodies can be obtained and, based on obtained information, positions and relationships between at least multiple celestial bodies can be determined, based whereon a local vertical and / or a horizon is determined.

[0051] According to one embodiment, system (100) comprises a horizon calculation means can be arranged to determine a local vertical and / or a horizon based on information supplied by an IMU (200).

[0052] Said celestial navigation system (100) may also further comprise an inertial navigation system (INS, 300) which comprises at least an inertial measurement unit (IMU, 200). Said inertial measurement unit (200) may at least comprise individual gyroscopes, accelerometers and magnetometers. These subsystems can be utilized to determine the horizon with an acceptable level of accuracy. Information obtainable from the INS (300) can be utilized by the processingmeans (400) such that the information acquired by the imaging submodule (101) can be incorporated into a loss function whereby the accuracy of heading and position of the vehicle can be improved.

[0053] Disclosed invention's celestial navigation system (100), based on the images obtainable by the imaging submodule (101), determines the parallactic motion aspects of celestial objects (500) such as the Moon and the Sun, as well as unique properties such as light polarization information obtainable by a sun polarization detection means (104) comprised by said submodule (101) according to at least one embodiment. Said sun polarization detection means (104) may be configured as a polarization image sensor, such as a polarization camera, and may also further comprise a sun filter. Various other unique properties used in tandem with a given set of unique properties can also be used to accurately determine the position of a vessel on the globe. Information obtained by these unique properties and measurements based on images captured by the imaging submodule (101) are comparatively solved for integrating Ephemeris data based on the incorporation of methods and algorithms processing means (400) is configured to implement, and used to determine current heading and position. In some embodiments, an inertial navigation system (300) is comprised as part of the disclosed system, which can further aid in determining the current heading and position, to which the result of the determined current heading and position is supplied.

[0054] According to an embodiment of the present disclosure, said storage medium (401) comprised by said processing means (400) comprises a database arranged to store Ephemeris data. Said data includes a star catalogue including information about a plurality of celestial objects (500), such as some or all of the celestial objects (500) in the form of a multi-dimensional lookup table (LUT), whereto information obtained using the imaging submodule (101) can be referred. As an example, images obtained by the imaging submodule (101)can include still images of the Moon and Sun, which can be subsequently analyzed to determine their apparent angular diameter and centroid in the focal plane of said image acquisition means (101). This raw data can then be converted into range and bearing angle measurements using planetary data and precise star tracker inertial attitude, which are included in said multidimensional LUT. In an exemplary embodiment, range and bearing can be determined with data obtainable by IMU (200) attitude; using self-attitude determination from star centroiding. In these exemplary scenarios, calculations and measurements can be utilized by said processing unit (400) configured to use a Kalman filter to update an onboard state vector. Calibration and distortion corrections can be applied to images based on the known geometries of the Moon, the Sun, or star centroids.

[0055] Based on the images captured by the imaging submodule (101), the system processes them to identify and classify celestial objects (500) by crossreferencing them against said Ephemeris data which may further comprise data in the form of a preloaded celestial catalog. This catalog is arranged as a vast database of known celestial bodies (500), including stars (e.g. the Sun, Sirius, Arcturus etc.), constellations, and various other celestial objects such as planets, each with precisely recorded positional data. Using this information, the system is configured to identify known stars and celestial bodies (500) in the captured images. Subsequent to the identification of the celestial body (500) observed using the imaging submodule (101), the system proceeds to calculate the precise positions of the observed celestial objects. For certain embodiments, said processing means (400) is configured to implement photometric processing such that raw images obtained by the imaging submodule (101) are preprocessed to using different algorithms and approaches which will be detailed below. Said processing means (400) can also be configured to implement angular measurements to determine the relativepositions of stars and other celestial bodies (500) of an image / images captured by the imaging submodule, triangulation wherein the computation of distances based on the known positions of stars and other celestial objects (500) in the background is performed.

[0056] According to one example, based on the information obtained based on the imaging submodule (101) data comprising visual information pertaining to at least one celestial body (500), comparing the observed positions with known reference data. Subsequent to this, navigational algorithms are utilized to calculate the vehicle's precise position and orientation in space. This step may involve combining multiple reference points, including, according to various embodiments, information pertaining to at least one other celestial object (500), whereby a three-dimensional navigational solution accounting for the position and attitude of the vehicle is calculated. Per certain embodiments, this navigation solution can be a cost function implementable by the processing means (400).

[0057] The processing means (400) capable of image processing according to various embodiments is configured to perform image processing algorithms usable for, based on any given image or series of images part of a video feed or an acquired video obtainable by the imaging submodule (101), celestial object altitude, angular speed components at time delta-t (At), polarization information, parallactic angle and parallactic speed. Processing means (400) can be configured to perform several image processing algorithms based on the images / video acquired by an image acquisition means (101), selectable from a group including the following: Resizing, whereby the dimensions of an acquired image are adjusted to a specific size to ensure uniformity in image dimensions. Normalization whereby pixel values are scaled to a standard range, typically [0, 1] or [-1, 1]. Grayscale conversion, whereby an acquired color image is converted to grayscale for reducing computational complexity andfocuses on intensity information in the acquired image. Cropping whereby unwanted outer areas of an image are removed and discarded such that focus is landed on the region of interest and irrelevant parts are removed. Blurring, whereby a filter is applied to the acquired image to smooth the image such that noise and detail are balanced, which can be useful for certain applications like edge detection. Histogram equalization, whereby the contrast of an acquired image is adjusted by modifying its histogram, thus the contrast is enhanced, making features more distinguishable. Rotation whereby an acquired image is rotated by a certain angle such that the information is enhanced by providing different orientations of the same image. Flipping, whereby the acquired image is flipped horizontally or vertically such that the strict orientation of the visual information present in the acquired image is modified. Color jittering whereby the brightness, contrast, saturation, and hue of the acquired image is changed based on predetermined considerations such that ability to generalize by accounting for different lighting conditions is enabled. Feature detection, whereby key points can be identified using algorithms such as Harris Corner Detection, Shi-Tomasi, or SIFT (Scale-Invariant Feature Transform) to detect salient points in a given frame; feature selection whereby features that are distinct and can be reliably tracked are highlighted; feature description whereby descriptors like SIFT, SURF (Speeded-Up Robust Features), or ORB (Oriented FAST and Rotated BRIEF) are used to describe at least a set of detected features which help in matching features across different frames of acquired images; feature matching using algorithms like the Kanade-Lucas-Tomasi (KLT) tracker or optical flow methods to match features across different frames of acquired images; motion estimation whereby the motion of the features between frames are calculated using techniques such as optical flow, which estimates the motion vector of each pixel; filtering and smoothing whereby filters like the Kalman filter are applied to smooth the motion estimates and reduce noise; feature update wherebyfeatures are continuously updated with respect to positions and descriptors as the different frames of acquired images consecutively appear as a part of acquired video, ensuring that the tracking remains accurate even if the appearance of the features changes.

[0058] Processing means (400), according to at least one embodiment of the present disclosure, may also be configured to perform celestial navigation algorithms incorporating at least data obtained by said imaging submodule (101), said algorithms comprising: Kalman filtering, based on data supplied by an INS (300), or a similar recursive filter estimation based on data acquired by said imaging submodule. Celestial fix-based spherical trigonometry, whereby a position on the globe is solved for celestial altitude h, declination 5, and Greenwich Hour Angle (GHA) based on data acquired by imaging submodule. Least squares using information of a given celestial body based on data acquired by imaging submodule, for integrating multiple celestial measurements, in the order of predicting altitude, calculating residuals for each observation, and minimization of the residuals based on an objective function. Polarization pattern analysis, wherein a polarization angle belonging to a celestial body is observed, polarization pattern is matched to a predetermined or catalogued pattern based on the object's and Sun's known and estimated position(s), orientation is adjusted ref. IMU data and polarization pattern alignment, and a position estimate is refined using at least one other celestial observation and / or output of a dead reckoning computation. Dead reckoning, where a known position, velocity and initial orientation determining starting heading, pitch and roll is acquired based on data provided by the INS and / or AHRS, position is updated using linear acceleration and angular velocity data provided by the IMU, celestial observations are compared with AHRS or INS-derived estimates to determine an error vector, velocity and orientation are recalibrated. Iterative lunar distance method.A method of celestial navigation is disclosed according to the present invention. Said method can be implemented by a celestial navigation system (100) as explained hitherto, comprising an imaging submodule (101) and a processing means (400) as well as an onboard timepiece. According to this method, a horizon is established using a horizon finding camera, an inclinometer, an inertial measurement unit (IMU) and / or inertial navigation system (INS) and / or an attitude and heading reference systems (AHRS). Subsequently, at least one celestial body is observed using said imaging submodule (101) and tracking said celestial body by obtaining images at at least two different times. Based on the obtained images, at least one of the following steps are implemented: Angular difference calculation of observed celestial body with respect to the established horizon; determining parallactic property (angle and speed) of the celestial body; calculating bright limb angle of the celestial body; determining polarization angle of light coming from / reflected off the surface of the celestial body; calculating and determining lateral and vertical angular velocity of the celestial bodies. In conjunction with any combination of said steps, and time information supplied by an onboard timepiece, data obtained are integrated into a cost function and the position and heading of a vehicle on Earth is determined. Subsequent to this, position and heading information is delivered to an inertial navigation system (300).

[0059] A method of celestial navigation according to the present invention is disclosed for determining the position information (e.g. earth coordinates) of a vehicle. According to said method, following information is integrated for determining the position information of a vehicle: A horizon image and / or an orientation / altitude information obtained by a horizon calculation means, a visual feature detected or captured by an imaging submodule and a polarized camera, based on which a parallactic angle of the Sun and / or the Moon is determined, angular velocities of the Sun and / or the Moon, polarizationinformation of the Sun, height of a celestial object from the horizon, a visual information detected by an imaging submodule based on which information about the bright limb of the Moon is determined, celestial object information provided by a celestial object catalog, and information provided by an onboard timepiece. Based on integrated information, position information is determined and a substitute latitude and longitude estimation is calculated.

[0060] In an embodiment, a celestial navigation system (100) for determining position and heading direction of a vehicle comprising an imaging submodule (101), a processing means (400) and an onboard timepiece is disclosed. In one aspect of the disclosed invention, said imaging submodule (101) comprises at least one image acquisition means (102) selectable from a group including thermal, daytime and / or event-based image sensors, provided on a pivoted support means (103) allowing motion on at least two orthogonal axes. In one aspect, said pivoted support means (103) comprising rotary encoders for each of said orthogonal axes. In another aspect, said processing means (400) comprises a storage medium (401) for retrievably storing Ephemeris data. In another aspect, said processing means (400) is configured to at least measure parallactic angle and angular speed of at least one celestial object (500) based on images captured by said at least one image acquisition means (101), and determine, by using said measurements, at least a current heading and a current position of a moving vehicle.

[0061] According to another aspect of the invention, said processing means (400) is further configured to measure polarization, limb and altitude information of said celestial object (500) based on the images captured by said imaging submodule (101).

[0062] According to another aspect of the invention, said processing means (400) is further configured to extract a textural information of said at least one celestialobject (500) based on the images captured by said imaging submodule (101).

[0063] According to another aspect of the invention, said system (100) further comprises a speed sensor configured to provide real-time estimates of vehicle ground speed.

[0064] According to another aspect of the invention, said imaging submodule (101) further comprises a sun polarization detection means (104).

[0065] According to another aspect of the invention, said system (100) further comprises an inertial navigation means (300) further characterized by an inertial measurement unit (IMU, 200) comprising at least a gyroscope, an accelerometer and a magnetometer.

[0066] According to another aspect of the invention, said celestial object (500) is selectable from a group including the Sun, the Moon, at least one star.

[0067] According to another aspect of the invention, said system (100) further comprises a horizon calculation submodule configured to find true horizon and provide sextant readings to the processor.

[0068] According to another aspect of the invention, a method of celestial navigation for determining position and heading of a vehicle implementable by a system comprising at least an imaging submodule and a processing means is disclosed. It comprises the steps of; obtaining, by said imaging submodule, an image of a celestial body selectable from a group including the Sun, the Moon and at least a star; computing, based on the obtained image of said celestial body, a parallactic angle value; computing, based on the obtained image of said celestial body, a polarization angle of the light emanating from said celestial body; determining, based on the obtained image of said celestial body, horizontal and vertical angular speed components of said celestial bodydetermining a position and heading of the vehicle, by comparing computed and determined values of a parallactic angle value, polarization angle, horizontal angular speed and vertical angular speed of said celestial body with Ephemeris data.

[0069] In another aspect of the invention, said method further comprises a preliminary step of horizon finding, whereby a horizon for the vehicle is calculated.

[0070] In another aspect of the invention, said method further comprises a step of computing altitude, whereby an angle between the horizon and celestial body is determined.

Claims

CLAIMS1) A celestial navigation system (100) for determining position and heading direction of a vehicle comprising an imaging submodule (101), a processing means (400) and an onboard timepiece, characterized in thatsaid imaging submodule (101) comprises at least one image acquisition means (102) selectable from a group including thermal, daytime and / or eventbased image sensors, provided on a pivoted support means (103) allowing motion on at least two orthogonal axes,said pivoted support means (103) comprising rotary encoders for each of said orthogonal axessaid processing means (400) comprises a storage medium (401) for retrievably storing Ephemeris data, andsaid processing means (400) is configured to at least;measure parallactic angle and angular speed of at least one celestial object (500) based on images captured by said at least one image acquisition means (101), anddetermine, by using said measurements, at least a current heading and a current position of a moving vehicle.2) A celestial navigation system (100) for determining position and heading direction of a vehicle as set forth in Claim 1 characterized in that said processing means (400) is further configured to measure polarization, limb and altitude information of said celestial object (500) based on the images captured by said imaging submodule (101).3) A celestial navigation system (100) for determining position and heading direction of a vehicle as set forth in Claims 1 and 2 characterized in that said processing means (400) is further configured to extract a textural information of said at least one celestial object (500) based on the images captured by said imaging submodule (101).4) A celestial navigation system (100) for determining position and heading direction of a vehicle as set forth in any preceding Claim characterized in that said system (100) further comprises a speed sensor configured to provide real-time estimates of vehicle ground speed.5) A celestial navigation system (100) for determining position and heading direction of a vehicle as set forth in any preceding Claim characterized in that said imaging submodule (101) further comprises a sun polarization detection means (104).6) A celestial navigation system (100) for determining position and heading direction of a vehicle as set forth in any preceding Claim characterized in that said system (100) further comprises an inertial navigation means (300) further characterized by an inertial measurement unit (IMU, 200) comprising at least a gyroscope, an accelerometer and a magnetometer.7) A celestial navigation system (100) for determining position and heading direction of a vehicle as set forth in any preceding Claim characterized in that said celestial object (500) is selectable from a group including the Sun, the Moon, at least one star.8) A celestial navigation system (100) for determining position and heading direction of a vehicle as set forth in any preceding Claim characterized in that said system (100) further comprises a horizon calculation submodule configured to find true horizon and provide sextant readings to the processor.9) A method of celestial navigation for determining position and heading of a vehicle implementable by a system comprising at least an imaging submodule and a processing means comprising steps of:obtaining, by said imaging submodule, an image of a celestial body selectable from a group including the Sun, the Moon and at least a star,computing, based on the obtained image of said celestial body, a parallactic angle value,computing, based on the obtained image of said celestial body, a polarization angle of the light emanating from said celestial body,determining, based on the obtained image of said celestial body, horizontal and vertical angular speed components of said celestial body,determining a position and heading of the vehicle, by comparing computed and determined values of a parallactic angle value, polarization angle, horizontal angular speed and vertical angular speed of said celestial body with Ephemeris data.10) A method of celestial navigation for determining position and heading of a moving vehicle as set forth according to Claim 9 characterized in that said method further comprises a preliminary step of horizon finding, whereby a horizon for the vehicle is calculated.11) A method of celestial navigation for determining position and heading of a moving vehicle as set forth according to Claims 9 and 10 characterized in that said method further comprises a step of computing altitude, whereby an angle between the horizon and celestial body is determined.

Images