Optical navigation system for spacecraft

Abstract

Systems and methods related to optical navigation systems for spacecrafts are disclosed herein. According to the optical navigation system, several types of sensors are combined into a compact optical sensor system, and the function of each type of sensor is kept. An optical navigation system uses optical sensors, artificial intelligence, automatic dimming, and other functions to perform navigation functions, such as sun sensing, horizon sensing, and star tracking. Artificial intelligence may include deep neural networks that can provide complex machine vision applications after being trained. The optical navigation system may be organized into modular slices to increase customizability. The types of slices may include optical slices, network switched slices, visual navigation slices, and inertial navigation system switched slices. Software packages for the navigation system are also customizable, reducing user cost and complexity to the extent required only by a given application.

Description

Cross Reference to Related Applications

[0001] This application claims priority to U.S. Patent Application No. 18 / 735,146, filed June 5, 2024, which claims the benefit of U.S. Provisional Patent Application No. 63 / 471,497, filed June 6, 2023, both of which are incorporated by reference herein in their entirety for all purposes. BACKGROUND

[0002] Space exploration heralds great advances in human knowledge. For example, space exploration provides a unique perspective for studying Earth and the solar system, produces innovative technology, increases understanding of natural phenomena, and enhances our ability to prevent potential threats or disasters. Many tools, such as a sun sensor, a star tracker, a vision navigation processor, a navigation camera, an inertial measurement unit (IMU), a precision clock, and a global navigation satellite system, can work individually or together to provide valuable data and meaningful functionality for spacecraft navigation and positioning. For example, a sun sensor detects the position of the sun, a star tracker measures the position of stars, and an IMU measures and reports specific accelerations, angular velocities, and orientations of a related spacecraft. With each additional sensor or tool, a spacecraft has more capabilities, but is also heavier and more expensive. In addition to the cost of the additional physical features themselves, the increased weight of each physical feature increases the cost of launching the spacecraft into space. More sensors and tools enable a spacecraft to have more functionality, but are also more expensive. SUMMARY

[0003] The present invention relates to a spacecraft navigation system. Many tools, such as an optical image sensor, a sun sensor, a star tracker, a vision navigation processor, a navigation camera, an inertial measurement unit (IMU), a precision clock, and a global navigation satellite system, can be combined together to create an optical navigation system that is smaller than the sum of its parts, while still maintaining the functionality of each tool. Launching a heavier spacecraft into space is more expensive, so a method is needed to reduce the weight of a spacecraft without sacrificing functionality.

[0004] By combining multiple sensors with artificial intelligence, the optical navigation system integrates the sensors while maintaining the functionality of each sensor. For example, depending on customer needs, the navigation system can provide sun sensing, horizon sensing, celestial body tracking, target centroid location, target tracking, cis-lunar optical navigation, terrain relative navigation, hazard detection, customer-defined machine vision, and general computing applications with minimal additional hardware. The artificial intelligence can include deep neural networks that are trained to, for example, provide terrain relative navigation (TRN) and hazard avoidance (HAZ) navigation data, as well as other complex machine vision applications. The spacecraft opto-electronic navigation system can be organized into modular slices to increase customizability. The types of slices include optical slices described herein, network switch slices, vision navigation slices, and inertial navigation system switch slices. The software packages for the navigation system are also customizable, reducing user cost and complexity to only what is needed for a given application. Furthermore, the navigation system can be small and light enough to be placed on very small spacecraft, booms, masts, solar arrays, antennas, robotic arms, crewed spacesuits, rovers, airlocks, mobile devices, and science payloads.

[0005] In contrast to traditional spacecraft, particular embodiments of the invention described herein reduce the cost, size, mass, power, and interface complexity of a spacecraft while increasing artificial intelligence capabilities and customizability. The compact general-purpose system can be upgraded in flight by uploading new application software.

[0006] In particular embodiments of the invention described herein, an optical navigation system for a spacecraft is provided. The optical navigation system includes: an array of optical image sensors to capture visible light data; and an artificial intelligence machine vision system on a chip communicatively coupled to the array of optical image sensors and storing at least one machine learning model to use the optical image sensor data for optical navigation, the optical navigation including sun sensing, horizon sensing, relative position, and celestial body tracking. In some examples, the optical navigation system provides 4p coverage.

[0007] In particular embodiments of the invention described herein, an optical navigation system for a spacecraft is provided. The optical navigation system includes: a stack of modular housings, where each housing constitutes a slice of the stack; and an optical head having an integrated array of image sensors and forming a cap of the stack.

[0008] In a specific embodiment of the invention described herein, a method for configuring an optical navigation system for a spacecraft is provided. The method includes: selecting at least one modular housing from a set of modular housings; placing the at least one modular housing in a stack, wherein each housing constitutes a slice of the stack; and forming a cover for the stack using an optical head, wherein an image sensor is part of the optical head. Attached Figure Description

[0009] The accompanying drawings illustrate various embodiments of the system and method, as well as embodiments of various other aspects of the invention. Those skilled in the art will understand that the element boundaries (e.g., boxes, groups of boxes, or other shapes) shown in the figures represent one example of a boundary. In some examples, one element may be designed as multiple elements, or multiple elements may be designed as one element. In some examples, an element shown as an inner component of one element may be implemented as an outer component in another element, and vice versa. Furthermore, elements may not be drawn to scale. A non-limiting and non-exhaustive description is described with reference to the following drawings. The components in the figures are not necessarily drawn to scale, but are emphasized to illustrate principles.

[0010] Figure 1A and Figure 1B A system according to a specific embodiment of the invention disclosed herein is provided.

[0011] Figure 2 A system according to a specific embodiment of the invention disclosed herein is provided.

[0012] Figure 3 Block diagrams are provided for specific embodiments of the invention disclosed herein.

[0013] Figure 4 A communication system according to specific embodiments of the invention disclosed herein is provided.

[0014] Figure 5 A flowchart of a specific embodiment of the invention disclosed herein is provided. Detailed Implementation

[0015] Reference will now be made in detail to various aspects and variations of the systems and methods described herein. Although several exemplary variations of the systems and methods are described herein, other variations of the systems and methods may include aspects of the systems and methods described herein, combined in any suitable manner, having a combination of all or some of the described aspects.

[0016] In accordance with the foregoing overview, the present invention details different systems and methods for optical navigation systems for spacecraft. The methods and systems disclosed in this section are non-limiting embodiments of the present invention, for explanatory purposes only, and should not be used to limit the full scope of the present invention. It should be understood that the disclosed embodiments can or can not overlap each other. Thus, a portion of one embodiment or a particular embodiment thereof can or can not fall within the scope of another or a particular embodiment thereof, and vice versa. Different embodiments from different aspects can be combined or implemented individually. Many different combinations and sub-combinations of the representative embodiments illustrated in the broad framework of the present invention will be apparent to those skilled in the art, but are not explicitly shown or described, and should not be construed as being excluded.

[0017] Figure 1A and Figure 1B A system 100 of a particular embodiment of the present invention disclosed herein is provided. Figure 1A A three-quarter view 101 of the system 100 is provided. Figure 1B A top view 102 of the system 100 is provided. The system 100 can be suitable for space applications and can be incorporated into a spacecraft. For example, the system 100 can be used as a primary flight computer or a network-connected general-purpose computer for a spacecraft. The system 100 includes a stack of tiles, such as optical tile 103-a (e.g., an upper optical tile) and optical tile 103-b (e.g., a lower optical tile), vision system tile 104-a (e.g., an upper vision system tile) and vision system tile 104-b (e.g., a lower vision system tile), patch antenna 105, external connector 106, optical sensor 107, and tile connector 108 (e.g., an example of a tile connector). The optical tile 103-a can act as an optical head of the stack as it forms the top of the stack and covers the tiles underneath it. As shown, the optical tile 103-a includes an integrated array of image sensors and forms a cover of the stack.

[0018] Optical slice 103-a includes optical sensor 107-a and is connected to vision system slice 104-a. Optical slice 103-b includes optical sensor 107-b and is connected to vision system slice 104-b. Optical slice 103-b can include four optical sensors 107-b, similar to optical slice 103-a, but only two optical sensors 107-b are shown. Both optical slices 103-a, 103-b can be associated with respective patch antennas, although only patch antenna 105 associated with optical slice 103-a is shown. One or more faces of optical slice 103 can be circular, and optical sensor 107 can be on the circular portion of the face. For example, the top of optical slice 103-a can have a circular dome-like portion, and optical sensor 107-a can be attached to the dome-like portion. In particular embodiments, optical sensor 107 faces outward from system 100 at an angle. Optical slices such as optical slice 103-a and optical slice 103-b can also be referred to as optical heads.

[0019] Slice connectors 108 can be repeated around system 100, for example, in the four corners of optical slice 103 and vision system slice 104. Slice connectors 108 can connect slices in a stack. The subsystem of optical slice 103-a and vision system slice 104-a is shown with the respective slices connected together by slice connectors 108. Slice connectors can be screws inserted through threaded holes in the slices. The subsystem of optical slice 103-b and vision system slice 104-b is shown with the respective slices connected together by slice connectors 108. Although shown separately, vision system slice 104-a and vision system slice 104-b can be further connected together by slice connectors 108.

[0020] System 100 can be connected in a variety of ways. Vision system slice 104-a can include external connector 106-a, and vision system slice 104-b can include external connector 106-b. External connectors 106 can allow system 100 to be connected to external hardware or peripherals (not shown). For example, external connectors 106 can include RS-422 interfaces, USB interfaces, Ethernet interfaces (e.g., Gigabit Ethernet), Wi-Fi access points, etc.

[0021] The system 100 can incorporate an optical navigation system that includes a computer vision system for space applications. The optical slices 103-a and 103-b can each capture 2p steradian coverage of a field of view via optical sensors 107-a and 107-b, respectively. Together, the optical slices 103-a and 103-b can capture 4p steradian coverage. The optical sensors 107 can be evenly spaced across their respective optical slices 103. The optical sensors 107 can capture visible light, infrared light, or both. The system 100 can be integrated with a machine vision computing navigation platform. The system 100 can replace traditional sun sensors, star trackers, optical navigation cameras, vision navigation processors, inertial measurement units (IMUs), external precision clocks, global navigation satellite systems (GNSS), and their associated power, data, and thermal control management in a single compact package. In particular embodiments, the single compact package can be less than 1 rack unit (U) and weigh less than 0.8 kilograms (kg).

[0022] The optical sensors 107 can provide 4p steradian all-sky coverage and can enable spherical machine vision coverage of all visible objects. In particular embodiments, such coverage can be obtained without changing spacecraft rotation or attitude control modes. The 4p steradian coverage can be used for any class of mission operations, including dynamic events such as space rendezvous, proximity operations, robotic manipulation, and planetary landing. Overlapping fields of view can be stitched into seamless Virtual Reality (VR) video and still images that can be transmitted as compressed or uncompressed data sets for immersive VR experiences by Earth-based users. Sub-frames of the 4p steradian field can allow machine vision algorithms to focus on one or more specific fields of view of interest (e.g., the location of the sun, a star field, elements of the spacecraft itself within the field of view, resident space objects (RSOs) including satellites and on-orbit debris, nearby planets, the moon, a disc of an asteroid or comet, a limb or horizon of a planet, etc.). In particular embodiments, multiple units of the system 100 (each system 100 including one or more optical sensors) can allow for higher resolution coverage and / or redundancy through field of view overlap in the 4p steradian field.

[0023] The system 100 can include artificial intelligence (AI). The system 100 can use an AI-compatible system on a chip (SoC) on a system on a module (SoM). The SoC can be radiation tolerant (e.g., selected for its radiation tolerance characteristics). The AI can be capable of using visible light data, infrared light data, or both. The visible light data, infrared light data, or both can be collected by the optical sensor 107. The AI can be a machine vision system on a chip that is communicatively coupled to the optical sensor 107 and can store at least one machine learning model to use visible light data from the optical sensor 107 for optical navigation, including sun sensing, horizon sensing, relative position, and star tracking. The AI can include a star tracking system, where the star tracking system determines a relative positioning of the optical navigation system based on image data from the optical image sensor array. The star tracking system can determine the relative positioning of the system 100 (e.g., the optical navigation system of the system 100) without performing spectral analysis and without locking onto a guide star.

[0024] The AI can include an encoding of at least one ephemeris. The AI can perform optical navigation (with a known ephemeris) relative to large and small planetary objects (stars, planets, dwarf planets, moons, satellites, comets, asteroids, etc.). For example, the optical navigation can include terrain relative navigation, feature tracking, hazard detection, and simultaneous localization and mapping (SLAM) using visible light data, infrared data, or both. The machine intelligence model can include an encoding of an ephemeris because it has been trained using images or other astronomical features (as objects of the ephemeris) of direct or simulated observations of the ephemeris. The training set can include labeled inputs, where the input is a view of an astronomical feature and the label is a local position relative to the astronomical feature from which the view was obtained. The simulated data can be generated from an existing ephemeris or using actual images taken of the astronomical feature while recording the position of the camera at the time the image was taken. In this way, the machine intelligence model can encode the ephemeris because it has been trained using data that is typically applied to ephemeris and data that is typically collected as a response from the ephemeris.

[0025] The system 100 can include an automatic dimming function, such as a dimming mechanism that reduces the amount of light entering one or more optical sensors 107. The reduction of the amount of light entering the optical sensor 107 can be based on the amount of light reaching the optical sensor 107 meeting a threshold (e.g., the amount of light reaching a minimum amount, such as the amount of light associated with the sun within the field of view of the optical sensor 107). The system 100 can include a 4p steradian optical field of view, and the sun is always visible unless in an eclipse or shadow. For some applications, a fixed inertial attitude can cause the sun to remain in the field of view of the sensor for a long period of time (e.g., months or years), with the potential for sensor damage or sun image aging. To avoid this, the automatic dimming function can be automatically engaged (e.g., when enabled) when the sun is in the field of view. Thus, a single optical sensor can be used for sun tracking or celestial body tracking, and replace a dedicated shielded optical sensor for sun tracking and an optical sensor for celestial body tracking that is positioned in such a way that when positioned, the incoming light is magnified before being applied to the sensor.

[0026] The automatic dimming function can be implemented through electronic dimming circuitry, photochromic lenses, or mechanical filters. For example, the automatic dimming function can include an automatic iris, shutter, or a solid state solution. The solid state solution can avoid moving parts, and thus can increase reliability. The automatic dimming technology of the system 100 can be similar to the automatic dimming used in welding helmets, or can use photochromic lenses. For example, the optical sensor 107 can include lenses that change tint in response to sunlight. Radiation Resistant Optical Glasses (RROG) can be integrated into the lenses of the system 100. The RROG can electronically control the automatic dimming function of the lenses. In particular embodiments, 85% of the sunlight can be blocked from entering the lenses due to the automatic dimming.

[0027] The system 100 can be used for space situational awareness applications. In particular embodiments, the system 100 can include built-in high performance video, still image, and image processing capabilities, as well as large capacity on-board data storage capabilities. For example, the system 100 can have 60 frames per second video, high resolution panchromatic still images, and on-board H.265 MPEG and JPG compression in hardware.

[0028] The system 100 can allow for flight testing of various machine vision functions. For example, the system 100 can allow for flight testing of sun sensing (e.g., <0.1° RMS), horizon sensing (e.g., <0.1° RMS), star trackers (e.g., <10 arcsec), target centroid location (e.g., azimuth <0.1 RMS, scale error to known target distance <0.5%), and lunar optical navigation (e.g., position and velocity in lunar space based on centroid and onboard ephemeris, error is a function of integration time of Kalman filter).

[0029] Software for implementing the artificial intelligence machine vision functions of the system 100 can be tightly integrated and optimized for multi-core advanced reduced instruction set computer machine (ARM) microprocessors, digital signal processors (DSPs), and graphics processing units (GPUs). The ability to integrate advanced AI software functions is a benefit of the system 100. The ability to specify and procure space-qualified hardware and integrate current and future software functions into a turnkey hardware / software solution will result in a cost-effective product that can support the functions of a sun sensor, star tracker, vision navigation processor, and navigation camera. Thus, the system 100 can replace many spacecraft- dependent sun sensors and star cameras with net size, weight, power, and cost (SWAP-C) savings. The system 100 can also replace separate conventional vision navigation computers and / or spacecraft main flight computers and associated navigation cameras, situational awareness cameras, and even some science cameras.

[0030] The system 100 reduces cost, mass, power, and interface complexity while increasing the autonomous artificial intelligence functions of vision-based spacecraft and robotic systems. The system 100 can be incorporated into large and small spacecraft. The system 100 provides the possibility of high performance for small, low-cost systems and extends the functionality, performance, and productivity of spacecraft.

[0031] Figure 2A system 200 according to embodiments of the application disclosed herein is provided. The system 200 can incorporate elements of the system 100. The system 200 includes a stack 201 of slices. The slices can include an optical head 202 (also referred to as an optical slice), a network switch slice 203, an inertial navigation slice 204, and a vision system slice 205. The system 200 can include different combinations of slices. The system 200 can include multiple types of slices, or can omit slice types. For example, the system 200 can include only the optical head 202, the vision system slice 205, and the inertial navigation slice 204. In another example, the system 200 can include only the optical head 202, the vision system slice 205, and the network switch slice 203. In another example, the system 200 can include two optical heads 202, the network switch slice 203, the inertial navigation slice 204, and the vision system slice 205. Each slice can be a modular enclosure. The optical head 202 can form a cover of the stack 201, and can include an array of optical sensors 207 (e.g., optical image sensors). The optical sensors 207 can be organized into an array or integrated array. The stack 201 can include an AI machine vision system contained on a chip in one or more slices. The slices (e.g., enclosures) can include a hermetic seal around a center of the stack 201. The slices can communicate with each other (e.g., internal communication) via connections through the slices in the center of the stack 201, and can communicate with other systems (e.g., external communication) via external connections 206 through the enclosures of the slices. Different types of slices can have different types of external connections 206, and can have different numbers of external connections 206.

[0032] The slices in the stack 201 of the system 200 (such as the optical head 202) are easily reconfigurable, modular, and replaceable. The organization of the system 200 can allow for standard and custom optical configurations. The optical head 202 can include any number (e.g., four) of optical sensors 207 (imagers, sensors, readout electronics, and optics) that can be configured with any type of combination of optical sensors 207. For example, the optical sensors 207 can be visible range monochrome or color (e.g., Bayer filter), near infrared (NWIR), short wave infrared (SWIR) with integrated thermoelectric cooler, long wave infrared (LWIR) (e.g., microbolometer array), time-of-flight (TOF), or neuromorphic sensors. The optical sensors 207 can be standard, commercial M12 or C-mount metal lenses that are ruggedized for flight, or custom optics for the sensor and application, from fisheye to narrow field of view (appropriate for imaging and / or machine vision applications).

[0033] Many configurations of optical heads 202 are possible. For example, two identical back-to-back wide field of view (fisheye) sensors can be incorporated into system 200 for basic 4p spherical degree coverage. As another example, each optical head 202 has four optical sensors 207 (e.g., evenly spaced) and a standard configuration of two optical heads 202 can be incorporated into system 200. This example can result in the highest practical resolution 4p spherical degree field of view (FOV) coverage. In this and other cases, system 200 can be mounted (e.g., on a mast or boom) to improve coverage. Optical navigation can be performed based on images of optical sensors 207 (e.g., 4p spherical degree images). In particular embodiments, optical navigation can be performed without performing attitude changes or maneuvers to capture information beyond that provided by the 4p spherical degree images.

[0034] System 200 can include a light adjustment mechanism to reduce the amount of light reaching optical sensors 207. For example, an automatic light adjustment function can be enabled once the amount of light entering or reaching the optical lenses of optical sensors 207 meets a threshold. Situations that meet the threshold can include the amount of light reaching a minimum threshold associated with the amount of light output by the sun. In other words, once the sun is in the field of view of optical sensors 207, the automatic light adjustment function of optical sensors 207 can be enabled. The automatic light adjustment function can include electronic light adjustment circuitry, photochromic lenses, mechanical light filters, or a combination thereof.

[0035] Vision system slice 205 can include a central processing unit (CPU) (e.g., an 8-core CPU), a GPU, a digital signal processor (DSP), a 1000T Ethernet interface, a USB3 interface, a Wi-Fi interface, an RS-422 interface, a 5-36V input power supply with internal conversion / distribution, and a built-in closed loop heater. Vision system slice 205 can also be referred to as an optical processing slice, a vision navigation system slice, or a vision processing slice.

[0036] The stack 201 can include a network switch slice 203 that can provide high speed connectivity to other spacecraft systems, subsystems, and components. The network switch slice 203 can include any number (e.g., 12) of ports. The network switch slice 203 can include an integrated communication controller that supports standard managed Ethernet as well as rate-limited and time-triggered Ethernet traffic. The controller can be a radiation hardened application specific integrated circuit (ASIC) on a custom carrier card that includes a physical layer (PHY) magnetic transceiver, which can allow hot mate / de-mate and physical in-flight separation without concern for dead facing. The network switch slice 203 can allow the stack 201 to be a high speed data hub for vision-based high performance computing applications, where the primary flight software can be implemented on the system 200. Peripherals, components, subsystems, and other systems can be connected at speeds up to gigabit with very low latency. The network switch slice 203 can include 6 lanes of 1000T and 6 lanes of 100T. The lanes can be configured to be time-triggered, rate-limited, or best effort by traffic class. The network switch slice 203 can also include 5-36V power switch ports (e.g., 6 ports) and a built-in closed loop heater.

[0037] The inertial navigation slice 204 can include an IMU (e.g., dual range, triple redundant IMU, micro-electromechanical system (MEM)), a chip scale atomic clock (CSAC), a GNSS receiver, a 5-36V input power supply with internal conversion, a 1000T Ethernet interface, a USB2 interface, an RS-422 interface, and a built-in closed loop heater. The inertial navigation slice 204 can be referred to as an inertial navigation system (INS) switch slice or INS slice. The slices can be hermetically sealed together. The slices can include a set of hermetic seals around the center of the stack 201. The stack can also include venting features that can be built into any of the slices to allow trapped air to vent into space in a manner that maintains a Faraday cage with respect to potential radio frequency emissions or sensitivities.

[0038] The system 200 can include additional hardware components that can increase the system’s capabilities and functionality. Some of the additional hardware components can be connected to the system 200 via external connections 206, and some of the additional hardware components can be built into the slices. The hardware components can include a new or evolved system on chip (SoC), a system on module (SoM) board (e.g., custom, specified by space-ng), or a new or evolved custom carrier board or electronic slice that includes key new chip-level functionality. The custom carrier board can include an internal high integrity IMU, a high precision clock (e.g., a chip scale atomic clock (CSAC)), GNSS receivers and antenna ports, and an auto dimming capability (e.g., for long time sun gazing). Different SOCs can be used based on other features of the system 200 (e.g., the intended capabilities of the system 200). For example, a SoC similar to the SoC for a level 3 autonomous vehicle can be incorporated into the system 200. The SoC can include features such as boot image encryption, hardware encryption engines, hardware performance and interface monitors, built-in self-test, data integrity encoding, radiation mitigation using excess software performance (e.g., sufficient for memory scrubbing of large storage volumes), and a real time operating system (RTOS) certified for safety-critical and safety-critical applications.

[0039] The electronic architecture of the SoC can be contained in multiple slices or boards. One of the slices can include a system on module (SoM) board. Another slice can include a SoM navigation board. The navigation slice can include one or more GNSS, IMU, microcontroller, internal closed loop heater control, watchdog circuit, power conversion and distribution, over current / voltage protection, and a clock (e.g., a CSAC).

[0040] The particular embodiments described herein can be configured to be cost effective, suitable for most aerospace systems. In particular embodiments, high-end applications and missions that require high performance can be supplemented with additional external hardware. The external hardware and data paths can be connected to the system 200 through wired (e.g., through RS-422, Ethernet, and / or USB connections) or through wireless networks. Hardware components and associated capabilities can be added while maintaining the low mass, low cost, and low volume of the integrated system.

[0041] In particular embodiments, the system 200 includes modular software modules that can be installed as software packages according to customer needs, including adding software-defined functionality developed long after the launch. The software packages can also include ground-based software, such as system application development, machine vision simulation, neural network training, and system testing (calibration testing, functional and performance testing, real-time hardware-in-the-loop testing, and faster testing than real-time Monte Carlo methods).

[0042] Software packages can be configured according to the needs of different users. Some users can only need attitude quaternions from sun sensing and star tracking. Other users can need horizon detection. Still others can need target tracking, marker recognition, QR code scanning, or terrain-relative navigation and hazard identification for lunar landing. Modular software packages can customize the system 200 for users with modular slices, reducing user cost by providing only the functionality needed by each customer.

[0043] An integrated AI (e.g., AI-augmented) vision system within the system 200 can replace traditional attitude knowledge hardware (sun sensors, star trackers, horizon sensors, situational awareness cameras) with lower size, weight, power, cost, integration, and operational complexity. The AI vision system can be incorporated into a system-on-a-chip, can be contained in one or more slices of the stack 201, and can include at least one machine learning model. The AI vision system can include a star tracking system that determines the optical navigation system’s relative orientation of the system 200 based on image data from an array of optical sensors 207 (e.g., optical image sensors). The star tracking system can determine the optical navigation system’s relative orientation without performing spectral analysis or locking onto navigation stars. The AI vision system, e.g., machine learning model, can include an encoding of at least one ephemeris and can use visible light data from the optical sensors 207 for optical navigation relative to large and small planetary objects with known ephemerides, including terrain-relative navigation, feature tracking, hazard detection, and simultaneous localization and mapping (SLAM).

[0044] Figure 3 A block diagram of a system 300 according to particular embodiments of the application disclosed herein is provided. The system 300 can incorporate aspects of the system 100, the system 200, or combinations thereof. The system 300 can include image sensors and readout integrated circuits (ROICs). The system 300 can use widely adopted interface standards and can allow for a wide selection of sensors and sensor types. The sensors can have capabilities including color (e.g., through a Bayer filter or the like), monochrome (e.g., black and white), short wave infrared (SWIR), rolling shutter, global shutter, time-of-flight (ToF), neuromorphic (e.g., active readout), video, and still imaging, as well as resolutions up to very high levels of resolution (e.g., >30 megapixels) for each sensor.

[0045] System 300 can use a system-on-module (SoM) on system-on-chip (SoC) (e.g., a SoM designated by space-ng). The SoC can be radiation tolerant (e.g., selected for its radiation tolerance characteristics). The SoC can include at least one multi-core CPU, a programmable digital signal processor (DSP), a programmable neural net processing (NNP) engine, a programmable graphics processing unit (GPU) engine, a video compression engine (e.g., a H.265 video compression engine), and a peripheral component interconnect express (PCIe) bus. The SoC can also include support for high-speed wired (e.g., GigE) and wireless (e.g., 802.11ad Wi-Fi) network interfaces, as well as support for utility serial interfaces including serial peripheral interface (SPI), inter-integrated circuit (I2C), and RS-422. System 300 can include physical layer (PHY) implementations of wired and wireless external data interfaces, including transceivers for Gigabit Ethernet, Wi-Fi, USB, and RS-422.

[0046] The SoM of system 300 can provide direct support for spaceflight-specific functions. For example, system 300 can include a hardware-controlled, software-settable, distributed, internal closed-loop heater circuit controlled by an internal closed-loop heater controller. The closed-loop heater circuit can protect temperature-sensitive (e.g., cold-sensitive) components at the board level. The closed-loop heater circuit can be a small array of heaters mounted directly on the circuit card and can be adjacent to low-temperature sensitive electronic elements including image sensors.

[0047] In addition to fault protection functions on the SoC, a watchdog circuit system (e.g., a watchdog circuit) can be implemented in system 300 (e.g., on the SoM). The watchdog circuit system can monitor, reset, and power cycle the SoC, as well as field programmable gate arrays (FPGAs) and microcontrollers on the SoM.

[0048] System 300 can include power conversion, distribution, and control circuitry. The power conversion, distribution, and control circuitry can convert spacecraft power (e.g., 5±, 12±, or 28± V DC nominal) to secondary voltages. The SoC and other components can use the secondary voltages. The power conversion, distribution, and control circuitry can switch power to various internal elements of system 300, including the SoC, FPGAs, microcontrollers, and Wi-Fi radios.

[0049] The system 300 can include a sleep mode. In the sleep mode, functions (e.g., all functions except for closed loop heater functions) can be turned off by an external command. These functions can be turned back on by another command (e.g., an RS-422 command or by a power cycle of the system 300).

[0050] A mobile industry processor interface (MIPI) camera high-speed interface can support rolling shutter, global shutter, short-wave infrared (SWIR), time-of-flight (ToF), or neuromorphic sensors as options, enabling other classes of imaging and machine vision applications. A MIPI-to-Gigabit Multimedia Serial Link (GMSL) converter can provide extended cable length for remotely mounted imagers (e.g., sensor arrays, ROICs, and optics). The system 300 can provide full-sky, 4p steradian coverage. In other words, imaging and machine vision functions can be performed without spacecraft maneuvering.

[0051] In particular embodiments, the system 300 can be scaled by N+1 at the unit level. In other words, additional units of the system 300 that are cooperatively networked can incrementally provide redundancy, increased fault tolerance, higher resolution imaging capabilities, higher accuracy navigation estimation capabilities, greater computing capabilities, greater data processing capabilities, greater data storage resource capabilities, and capabilities of overlapping coverage from different perspectives, among other benefits. In particular embodiments, multiple units of the system 300 (each system 300 including multiple optical sensors or cameras) can allow for higher resolution coverage and / or redundancy through field of view overlap in a 4p steradian field of view.

[0052] In particular embodiments, two or more systems 300 can be logically connected. Logically connected systems 300 can share data and compute resources. For example, systems 300 can share data and compute resources over a gigabit Ethernet. The gigabit Ethernet can be connected through an external network switch, which can be limited from a data perspective only by the number of available network switch ports. As another example, systems 300 can share data and compute resources over Wi-Fi. The Wi-Fi can be connected through an antenna or a coaxial cable to an external passive attenuator unit (e.g., a radio frequency attenuator unit) or a coupler unit. The Wi-Fi connection can be limited by the number of available physical coaxial cable connections. The Wi-Fi systems on systems 300 can be configured as clients, or access points that serve other Wi-Fi clients that can include other systems 300, or any other Wi-Fi compatible device. These systems can include network slices to facilitate logical connections between two or more of these systems.

[0053] Systems 300 can perform machine vision tasks, as well as conventional video and still imaging tasks, as standalone units or as part of a “hive mind.” Hive mind functionality includes sharing data, compute, and storage resources, allowing a range of configurations and behaviors. For example, as more systems 300 communicate with each other and individual fields of view proportionally narrow, full-sky 4p steradian coverage can be achieved with increasingly higher resolution. Individual optical sensors (e.g., imagers) or individual systems 300 of a hive mind can assume specific navigation, machine vision, or surveillance tasks (e.g., tracking of dim or distant objects). Multiple systems 300 can enable higher levels of fault tolerance than a single system 300. For example, redundant or overlapping machine vision or imaging coverage, block-redundant hardware units, N+1 redundant hardware units, and functional redundancy with other spacecraft computing, imaging, or data storage subsystems can increase fault tolerance.

[0054] In particular embodiments, systems 300 can include one or more built-in chip-level tactical-grade IMUs, internal high-precision clocks (e.g., CSACs), and GNSS receivers as navigation aids. These features can replace other avionics boxes or circuit cards and further reduce the requirements for spacecraft navigation hardware and interfaces with low incremental size, weight, power, and cost.

[0055] Systems 300 can include Figure 3additional features not shown. For example, the system 300 can include one or more USB 3.0 interfaces for interfacing with external devices (microphones, speakers, displays, cameras, solid state storage drives, media converters, co-processor, controllers, etc.), or as another spacecraft command and data interface. The system 300 can include one or more power output interfaces for low power accessories (e.g., LED lights, motor drivers, actuator signals, or other external devices). The system 300 can include four or more cameras, the number of cameras corresponding to SoC functionality.

[0056] System 300 can include an IMU, such as a high-integrity IMU, a triple-redundant IMU, and / or a self-checking IMU. The IMU associated with system 300 can include a three-axis gyroscope and a three-axis accelerometer. The gyroscope can have performance that meets or exceeds tactical level requirements (e.g., a minimum range of 400 + / - 1 degree / second, a maximum bias stability in run of 1.5 degrees / hour, a maximum scale factor accuracy of 0.02%, a minimum bandwidth of 50 hertz, a maximum bias over temperature of 0.3 degrees / second, and a maximum angular random walk of 0.2 degrees / sqrt(h)). The accelerometer associated with system 300 can have performance that is consistent with or better than tactical level requirements (a minimum range of 8 g and a minimum bandwidth of 50 Hz). A high-integrity IMU can be an IMU that has a high degree of confidence that it will never transmit false information and will fail silently if it does fail. This can be achieved by using multiple MEMS chip-level IMUs in parallel and embedding liveliness checks, best value selection logic, filtering, integrity checks, soundness limits, voting, and error correction coding in radiation-hardened firmware. The IMU can not need long-term precision performance since the optical portion of system 300 can provide high-precision attitude quaternions and angular rates at 10 Hz or higher. However, the IMU can cover instantaneous mission events and potential safety mode entries in which the optical navigation is compromised or not functioning. The selection and integration of the high-integrity IMU of system 300 can involve a trade study (SWAP-C) that considers the individual MEMS IMU size, weight, power, and cost; the MEMS IMU radiation characteristics and shielding; the circuit design and embedded firmware design for high-integrity output; the internal electrical interface, electromagnetic compatibility (EMC), electromagnetic interference (EMI), grounding; the thermal interface and internal control, mechanical packaging, and chassis integration; and the individual MEMS IMU performance. The individual MEMS IMU performance can include range (and the ability to change range in flight), resolution, bandwidth and sample output rate, noise, noise density, random walk, bias, bias stability, repeatability, temperature sensitivity, scale factor, scale factor error, nonlinearity, orthogonality, and cross-axis sensitivity, as well as g and g2sensitivity. The IMU can be located on a dedicated inertial navigation slice or integrated into any other slice.

[0057] System 300 can include a clock (e.g., a high-precision clock or a chip-level atomic clock). The clock can be customized according to customer requirements (e.g., less than 120 mW of power consumption, 10 MHz output, 1 PPS output and 1 PPS input synchronization, 3.0 x 10 -10). When used in conjunction with radio navigation technology and / or high-precision onboard ephemeris, high-precision onboard clocks can be used to supplement optical navigation. The clock can be, for example, a chip-scale ultra stable oscillator (USO) and / or a space-rated CSAC. System 300 can train the onboard clock and timer to be a high-quality source of absolute time for extended periods of time, even when communication with Earth ground stations is not possible, impractical, or denied. Training the onboard clock and timer provides the user with an important measurement reference for a variety of tactical and scientific applications. The clock of system 300 can be a user-friendly high-precision onboard clock. System 300 can use a backup battery to ensure the clock continues to run even if system 300 is temporarily unpowered or cycled. The backup battery can be a coin cell size. The clock can be on the inertial navigation slice or the GNSS slice.

[0058] System 300 can include a GNSS receiver. The GNSS receiver can be compatible with GPS, QZSS, Galileo, GLONASS, and Beidou satellite systems. The built-in GNSS navigation functionality and the precise time provided by GNSS can be particularly useful for low Earth orbit (LEO) applications (e.g., the GNSS receiver can have a minimum 4 g acceleration limit, a minimum 20,000 km altitude, a minimum 10 km / s orbital velocity, a maximum 0.05 m / s velocity accuracy, and a maximum 0.3 degree dynamic heading accuracy). System 300 can include a space-enabled GNSS solution that eliminates some or all of the commercial performance limitations. The GNSS system can be suitable for cases with orbital altitudes and velocities and can be associated with low enough cost, mass, power, and volume to integrate into system 300. An antenna port (e.g., a subminiature version A (SMA) antenna port) can be included on the enclosure of system 300 such that an external antenna can be provided (e.g., separately). The GNSS receiver can be on the inertial navigation slice or the GNSS slice.

[0059] System 300 may include one or more USB interfaces (e.g., USB 3.0), internal MIPI-compatible machine vision sensors, external machine vision sensors, internal closed-loop heater control, watchdog circuitry, power conversion and distribution, overcurrent / voltage protection, JTAG port, multi-user multiple-input multiple-output (MU-MIMO), Wi-Fi port, non-volatile memory (NVM) solid-state drive (SSD), Serial Peripheral Interface (SPI) port, RS422 PHY, Gigabit Ethernet PHY, 4–36V input, D-sub connector, SMA coaxial cable, PCIe backplane, test block, and SoC. The SoC of System 300 may include one or more CPUs (e.g., an 8-core CPU), an image signal processor (ISP) (e.g., a 10-bit HDR, 30fps zero shutter delay Rec 2020 color gamut dual 13 MPix imager), a digital signal processor (DSP) and a neural processing engine (NPU), a graphics processing unit (GPU), and random-access memory (RAM).

[0060] System 300 may include an automatic dimming function. Automatic dimming can be implemented for prolonged sun staring and can be implemented on an individual basis for each imager, optics, lens, or sensor. The automatic dimming function can dim at least 85%. Automatic dimming can be implemented via passive (e.g., progressive), solid-state, or mechanical means.

[0061] The System 300, with its internal IMU, precision clock, GNSS receiver, auto-dimming, and additional SoC capabilities and interfaces, allows for the replacement of conventional aerospace hardware such as solar sensors, star trackers, visual navigation processors, navigation cameras, IMU precision clocks, and GNSS receivers.

[0062] Figure 4A communication system 400 is provided according to a specific embodiment of the invention disclosed herein. The communication system 400 includes a spacecraft 401 in space and an antenna system 402 on Earth 403. The spacecraft 401 operates using a system 405, which may incorporate aspects of systems 100, 200, 300, or combinations thereof. The antenna system 402 transmits communication information 404 to the spacecraft 401. The communication information 404 may be, for example, a software update received by the antenna of system 401. The software update may update the functionality of an optical navigation system. For example, the software update may update the AI ​​machine vision functionality on the chip system of spacecraft 401. System 405 may include onboard navigation, guidance, control, and flight monitoring software products, as well as related ground system simulation and testing software for remote operation on Earth 403. System 405 may also additionally include third-party open architectures or open-source applications.

[0063] System 405 of spacecraft 401 may include a networked circuitry system (e.g., via Ethernet or WiFi) that can receive updates via communication information 404. System 405 can be securely updated in flight (e.g., in space, in the atmosphere) via secure communication information 404, featuring advanced, software-defined machine vision and high-performance processing capabilities to enhance and extend mission capabilities. System 405 may be a software-defined system updated in flight using machine vision software modules that include solar sensing, horizon sensing, celestial tracking, target centroid localization (e.g., azimuth and distance), target tracking relative navigation (e.g., rendezvous and approach operations), lunar optical navigation (e.g., Earth- or Moon-centered inertial position and velocity), terrain-relative navigation (e.g., precision landing), hazard detection (and avoidance), and customer-defined machine vision or general computing applications.

[0064] System 405 can operate on long-duration spaceflights beyond geosynchronous orbit (GEO). System 405 can be configured to operate anywhere in the solar system and can provide absolute and relative navigation using only optical technologies. For example, the sun, horizon, and celestial bodies can be used for attitude knowledge. The sun, planets (near and far), the moon, and celestial bodies can be used to determine position, velocity, and trajectory. System 405 can use illuminated objects for relative navigation, positioning, and orientation. System 405 can use illuminated surfaces for terrain-relative navigation and hazard detection.

[0065] Software applications that can be incorporated into System 405 include surface relative positioning (e.g., terrain relative navigation); hazard detection and avoidance; panoramic imaging of the landing site; digital terrain model (DTM) of the landing site; flight video and still image acquisition, compression, and data / image storage and management.

[0066] System 405 can be upgraded in flight with new features, algorithms, cooperative behaviors, or enhanced performance. Deep neural network systems can be trained in flight using in-flight data. System 405 can create maps of the environment (e.g., map generators) while tracking the position of the flight system, for example, by using Simultaneous Localization and Mapping (SLAM).

[0067] System 405 can integrate spacecraft optical navigation, video and imaging functions, and can replace traditional optoelectronic hardware consisting of a solar sensor, a star tracker, a visual navigation processor and a navigation camera.

[0068] Figure 5 A flowchart 500 is provided according to a specific embodiment of the invention disclosed herein. At 501, at least one modular housing can be selected from a set of modular housings. This set of modular housings may include optical processing slices, network slices, and inertial navigation slices. At least one selected modular housing may be an optical processing slice. The optical processing slice may be similar to vision system slice 205.

[0069] At 502, at least one modular housing from 501 can be placed in a stack, where each housing constitutes a slice of the stack. This stack can constitute any number of optical processing slices, network slices, and inertial navigation slices. The at least one modular housing can be placed anywhere in the stack. The at least one modular housing can be a first, a second, etc., or may be the only slice in the stack. The housings in the modular housing stack can form a set of hermetically sealed around the center of the stack. The slices of the stack can communicate internally via connectors passing through the slices within the center of the stack, and externally via connectors passing through the housings.

[0070] At 503, an optical head can be used to form a stacked cover, wherein an image sensor is part of the optical head. The optical head can be similar to optical slice 103 or optical head 202. The image sensor can be similar to optical sensor 107 or optical sensor 207. The optical head may include one or more image sensors, and each of the multiple image sensors may be evenly spaced.

[0071] The system implementing Flowchart 500 can be based on commercially available, spaceflight-compatible electronic components with customized mechanical, optical, imaging, operating system, and software configurations. When selecting components, hardware components can be added to the system based on the performance of the most cost-effective, space-compatible available components. The system can provide tightly integrated hardware capabilities, adding functionality and performance representing “best value” for most potential customers. For example, commercial electronic components can be tested, screened, and hardened for spaceflight. The integrated hardware may be mature, have flight heritage, require minimal development, be easy to integrate and test, have low swap-C, and have performance that meets most potential commercial, civilian space, and defense customers. The core electronics system is capable of hosting machine vision software, interconnecting with two, three, four, or more high-resolution imagers, and conforming to both traditional and emerging space propulsion and data interfaces. The core imaging and computational electronics in a compact, space-compatible form factor allow for the integration of AI using deep neural networks trained to provide terrain-relative navigation (TRN) and hazard avoidance (HAZ) navigation data (e.g., for autonomous planetary landers).

[0072] Compared to conventional methods and systems, the specific embodiments described herein significantly reduce overall size, weight, power, cost, and interface complexity while achieving equivalent or higher functionality and performance. The compact, general-purpose vision system described in these specific embodiments can be fundamentally upgraded in flight by uploading new application software with entirely new functionalities (e.g., threat detection for spatial situational awareness, cooperative navigation for rendezvous and approach operations, autonomy for robotic manipulators, etc.). The specific embodiments described herein can be combined with high-performance AI-based flight software for complex machine vision applications. The specific embodiments described herein can be combined with highly integrated and simplified ground-based analysis, training, simulation, testing, and validation software, providing a complete end-to-end development and validation environment for the flight system.

[0073] Although the specification has been described in detail with reference to specific embodiments of the invention, it should be understood that modifications, variations, and equivalents of these embodiments will be readily conceived by those skilled in the art upon understanding the foregoing. Any method steps discussed above can be performed by a processor operating in conjunction with a computer-readable, non-transitory medium storing instructions for those method steps. The computer-readable medium may be memory within a personal user device or network-accessible memory. While the examples in this disclosure are generally directed to spacecraft, the same methods can be used in other applications employing various sensors. These and other modifications and variations to the invention can be implemented by those skilled in the art without departing from the scope of the invention, the scope of which is more specifically set forth in the appended claims.

Claims

1. An optical navigation system (100) for a spacecraft (401), comprising: an array of optical image sensors (207) that capture visible light data; and an on-chip artificial intelligence machine vision system (300) that is communicatively coupled to the array of optical image sensors (207) and that stores at least one machine learning model to use the visible light data for optical navigation, the optical navigation including sun sensing, horizon sensing, and star tracking.

2. The optical navigation system (100) of claim 1, further comprising: a dimming mechanism that reduces an amount of light that enters an optical image sensor (207) of the array of optical image sensors (207) based at least in part on a threshold amount of light reaching the optical image sensor (207).

3. The optical navigation system (100) of claim 2, wherein: the dimming mechanism is electronic dimming circuitry.

4. The optical navigation system (100) of claim 2, wherein the on-chip artificial intelligence machine vision system (300) comprises: a sun tracking system, wherein the sun tracking system determines a relative positioning of the optical navigation system (100) based on image data from the array of optical image sensors (207).

5. The optical navigation system (100) of claim 4, comprising: a hermetically sealed stack (201) of slices; wherein the on-chip artificial intelligence machine vision system (300) is on a first slice of the stack (201) and the array of optical image sensors (207) is on a second slice of the stack (201).

6. The optical navigation system (100) of claim 1, wherein the on-chip artificial intelligence machine vision system (300) comprises: a star tracking system, wherein the star tracking system determines a relative orientation of the optical navigation system (100) based on image data from the array of optical image sensors (207).

7. The optical navigation system (100) of claim 6, wherein: the star tracking system determines the relative orientation of the optical navigation system (100) without performing spectral analysis.

8. The optical navigation system (100) of claim 6, wherein: the star tracking system determines the relative orientation of the optical navigation system (100) without locking onto a guide star.

9. The optical navigation system (100) of claim 1, wherein, the at least one machine learning model includes: an encoding of an ephemeris.

10. The optical navigation system (100) of claim 9, wherein, the at least one machine learning model utilizes a known ephemeris for optical navigation relative to large and small planetary objects, the optical navigation including topographical relative navigation, feature tracking, hazard detection, and simultaneous localization and mapping (SLAM) using the visible light data.

11. The optical navigation system (100) of claim 1, further comprising: a stack (201) of modular housings, wherein each housing constitutes a slice of the stack (201); and the at least one machine learning model includes: an encoding of an ephemeris. the at least one machine learning model utilizes a known ephemeris for optical navigation relative to large and small planetary objects, the optical navigation including topographical relative navigation, feature tracking, hazard detection, and simultaneous localization and mapping (SLAM) using the visible light data. an optical head (202) forming a cover of the stack (201), wherein the array of optical image sensors (207) is part of the optical head (202); wherein the artificial intelligence machine vision system (300) on a chip is contained within one of the slices of the stack (201).

12. The optical navigation system (100) of claim 1, further comprising: two or more optical heads (202) positioned to obtain 4π steradian images from the spacecraft (401); wherein the array of optical image sensors (207) is part of one of the two or more optical heads (202).

13. The optical navigation system (100) of claim 12, wherein: the optical navigation system (100) performs optical navigation based on the 4π steradian images.

14. The optical navigation system (100) of claim 13, wherein: the optical navigation system (100) performs optical navigation without performing attitude changes or attitude maneuvers to capture information other than that provided by the 4π steradian images.

15. The optical navigation system (100) of claim 1, further comprising: a closed loop heater associated with a temperature sensitive component of the optical navigation system (100).

16. The optical navigation system (100) of claim 1, further comprising: network circuitry; wherein the optical navigation system (100) uses the network circuitry to receive WiFi or Ethernet updates to functions of the artificial intelligence machine vision system (300) on a chip.

17. An optical navigation system (100) for a spacecraft (401), comprising: a stack (201) of modular enclosures, wherein each enclosure constitutes a slice of the stack (201); and an optical head (202) having an integrated array of image sensors and forming a cover of the stack (201).

18. The optical navigation system (100) for a spacecraft (401) of claim 17, wherein: the slices of the stack (201) include an optical processing slice and at least one of a network slice and an inertial navigation slice (204).

19. The optical navigation system (100) of claim 17, further comprising: at least one first optical image sensor (207) positioned to obtain a first 2π steradian image from the spacecraft (401); wherein the at least one first optical image sensor (207) is part of the optical head (202).

20. The optical navigation system (100) of claim 19, further comprising: at least one second optical image sensor (207) positioned to obtain a second 2π steradian image from the spacecraft (401); wherein combining the first 2π steradian image and the second 2π steradian image creates a 4π steradian image from the spacecraft (401).

21. The optical navigation system (100) of claim 17, wherein: The housings in the stack (201) of modular housings form a set of hermetic seals around the center of the stack (201); and The slices of the stack (201) communicate internally via connections through the slices in the center of the stack (201) and externally via external connections (206) through the housings.

22. The optical navigation system (100) of claim 17, the integrated array of image sensors comprising: four image sensors evenly spaced on the optical head (202).

23. The optical navigation system (100) of claim 17, further comprising: a light adjustment mechanism to reduce the amount of light entering the image sensors of the integrated array of image sensors based at least in part on a threshold amount of light reaching the image sensors.

24. The optical navigation system (100) of claim 17, further comprising: an artificial intelligence system comprising: a star tracking system, wherein the star tracking system determines a relative position of the spacecraft (401) based on image data from the integrated array of image sensors without performing spectral analysis or locking onto a guide star.

25. The optical navigation system (100) of claim 24, wherein the artificial intelligence system further comprises: at least one machine learning model comprising an encoding of an ephemeris.

26. The optical navigation system (100) of claim 25, wherein, The at least one machine learning model utilizes a known ephemeris for optical navigation relative to large and small planetary objects, including topographical relative navigation using visible light data, feature tracking, hazard detection, and simultaneous localization and mapping (SLAM).

27. The optical navigation system (100) of claim 17, further comprising: a closed loop heater associated with temperature sensitive components of the optical navigation system (100).

28. The optical navigation system (100) of claim 17, further comprising: a network interface, wherein the network interface is a WiFi or Ethernet interface and receives network updates to functionality of the optical navigation system (100).

29. A method (500) for configuring an optical navigation system (100) for a spacecraft (401), comprising: selecting (501) at least one modular housing from a set of modular housings; placing (502) the at least one modular housing in a stack (201), wherein each housing constitutes a slice of the stack (201); and forming (503) a cap of the stack (201) using an optical head (202), wherein image sensors are part of the optical head (202).

30. The optical navigation system (100) for a spacecraft (401) of claim 29, wherein: the at least one modular housing comprises an optical processing slice; and the set of modular housings comprises: a network slice and an inertial navigation slice (204).

Images