Precision navigation instrument and system

Abstract

Navigation instrument (10), of the type for applications in IoT devices, in the industrial sector and in the transport sector, the navigation instrument having: - at least one digital gyroscope (30), configured to detect and / or measure roll, pitch and yaw angles, at least one accelerometer (40), configured to detect and / or measure accelerations along three main axes, calculating the detected force with respect to the mass of the object, - a miniaturized processor (50), in which - the navigation instrument is free of analog gyroscopes; and - the processor (50) is supported by artificial intelligence algorithms and neural networks, configured to enable the acceleration of the data acquisition and processing processes from the digital gyroscope (30) and the accelerometer (40).

Description

[0001] PRECISION NAVIGATION INSTRUMENT AND SYSTEM

[0002] D ESCRI PTION

[0003] Technical Field of the Invention

[0004] The present invention relates to a precision navigation instrument, also definable as Motion Reference Unit (MRU), used to determine the direction, attitude and wave motion using the gyroscopic effect and advanced sensor fusion technologies.

[0005] Background art

[0006] The technology focuses on precision navigation, with primary applications in the military, maritime, aerospace, and advanced agriculture sectors, as well as in renewable energy and offshore extraction. These fields require precision elements to ensure high accuracy and reliability in determining routes and navigation conditions.

[0007] For example, navigation systems based on signals from sensors operating in GPS (Global Positioning System) mode are well-known. These systems, developed using satellite technology, however, do not allow precise positioning in areas where the GPS signal is weak or completely absent (GNSS-denied environments). Furthermore, in marine applications, the single-point GPS signal cannot effectively compensate for the effects of sea waves and requires significant electrical power.

[0008] Gyroscopic sensors and / or triaxial accelerometers are also well-known. However, the high-performance devices currently available have high power consumption and are bulky, preventing their use in miniaturizedsystems (drones, loT), and often require frequent calibration. In this specific field, analog gyroscopes offer precise measurements but require expensive converters and suffer from low sampling rates, limiting the effectiveness of real-time algorithms such as Kalman filters.

[0009] However, these systems also present numerous challenges:

[0010] - high power consumption, especially for high-performance devices, - poor integration with modern digital systems,

[0011] - no miniaturization of components is possible for advanced systems, - they cannot guarantee precise orientation and are unreliable in extreme conditions, such as extreme weather,

[0012] - poor compatibility with new navigation systems,

[0013] - the need for annual calibration, especially for devices with a mechanical configuration.

[0014] Gyroscopic instruments also present specific challenges:

[0015] - known gyroscopes are typically analog: they offer precise angular velocity measurements but require conversion to digital signals,

[0016] - they are expensive, bulky, and consume significant power, - they typically feature an analog-to-digital converter (ADC), which has high resolution but a slow sampling rate. This limits rapid motion detection, reducing responsiveness. Furthermore, the use of real-time algorithms, such as Kalman filters, is difficult because the slow flow of data delays the convergence of these algorithms,

[0017] - insufficient computational power and the resulting latency in data processing compromise accuracy.

[0018] There is therefore a need to identify a solution that can resolve or atleast mitigate the aforementioned drawbacks.

[0019] Summary of the Invention

[0020] To address the technical challenges outlined above, the proposed solution is a precision navigation instrument / Motion Reference Unit (MRU) belonging to the loT class of devices, capable of integrating high-end MEMS sensors, an inertial measurement unit (IMU), and an attitude and heading reference system (AHRS) for dynamic position and wave motion measurements, with predictive algorithms to anticipate future scenarios.

[0021] The invention comprises a family of products that includes variants with integrated GNSS receivers or auxiliary interfaces, as well as OEM versions for integration into third-party systems. The core of the system is a software architecture based on "Sensor Fusion" that combines neural networks for dynamic drift compensation and bias adaptation.

[0022] Therefore, according to the present invention, a precision navigation instrument is provided having the features set forth in the independent claim, appended to this specification.

[0023] Further preferred and / or particularly advantageous embodiments of the invention are described according to the features set forth in the appended dependent claims.

[0024] Brief Description of the Drawings

[0025] The invention will now be described with reference to the accompanying drawings, which illustrate some non-limiting embodiments, in which:

[0026] - Figure 1 is an isometric view of a navigation instrument according to a preferred embodiment of the present invention;- Figure 2 shows an exploded view of the navigation instrument of Figure 1;

[0027] - Figure 3 shows a first printed circuit board of the navigation instrument of Figure 1;

[0028] - Figure 4 shows a second printed circuit board of the navigation instrument of Figure 1; and

[0029] - Figure 5 is a schematic representation of a precision navigation system, including the navigation instrument of Figure 1.

[0030] It is understood that the dimensional variants and OEM configurations described below are derivable from the same circuit architectures illustrated below.

[0031] Detailed Description

[0032] By way of example and without limitation, the present invention will now be described with reference to the aforementioned figures.

[0033] With reference to Figure 1, the precision navigation instrument 10 according to a preferred embodiment of the present invention is presented in compact and diverse embodiments to adapt to specific operational needs.

[0034] The invention provides for several embodiments or variants:

[0035] - standard configuration, equipped with an auxiliary RS-232 port for connection to an external GNSS;

[0036] - GNSS configuration, including a GNSS receiver integrated directly into the device's electronics;

[0037] - OEM configuration: embodiment without a casing, consisting of the "Core" PCB module, intended for direct integration into third-party boards or systems.With reference to Figures 2 through 4, the navigation instrument 10 comprises:

[0038] - a first printed circuit board 11 (Figures 2 and 3), used to integrate a plurality of sensors, such as gyroscopes, accelerometers, auxiliary sensors, and to stabilize the power supply for these sensors. The first printed circuit board 11 therefore comprises at least one gyroscope 30, at least one accelerometer 40, a power regulation circuit equipped with power supply chips 16, as well as capacitors, resistors, and connectors. The communication interfaces are updated to support industrial and maritime standards: in addition to wireless connectivity, the instrument includes wired RS-232 / RS-422 interfaces capable of speeds up to 0.5 Mbaud and a 10 / 100 Base-T Ethernet interface with DHCP-UDP and TCP protocol support;

[0039] - a second printed circuit board 12 (Figures 2 and 4), which serves data processing, wireless communication, and system-level power regulation. The second printed circuit board 12 therefore includes a miniaturized Micro Controller Unit (MCU) processor 50, wireless connection modules 18, a power management circuit equipped with an inductor 19, an RF antenna, as well as capacitors, resistors, and connectors;

[0040] - a housing 13 that provides mechanical support and ensures that the navigation instrument module 10 can be securely attached to a fixed surface or a mobile platform. The housing 13 is designed for high environmental robustness. The device offers an IP68 protection rating (or IP66 for "lightweight" variants), is shock-resistant up to 100 g, and operates within a temperature range of -25°C to +70°C. These featuresmake the instrument suitable for marine applications compliant with IEC 60945 and IMO Res.A.424(XI);

[0041] - a cover 14; and

[0042] - a connector 15 for connecting external wiring for data transmission and power supply.

[0043] Also referring to Figure 5, the main functional components of the navigation instrument 10 are therefore:

[0044] - at least one digital gyroscope 30, for example three, to detect and / or measure roll, pitch, and yaw angles. These digital gyroscopes 30 may be made using different technologies, such as fiber optic gyroscopes (FOG), ring laser gyroscopes (RLG), or dynamic rotation gyroscopes (DTG), - at least one accelerometer 40, for example three, to detect and / or measure acceleration along three main axes, calculating the force detected relative to the object's mass, and

[0045] - the miniaturized Micro Controller Unit (MCU) processor 50, supported by Artificial Intelligence (Al) algorithms, particularly Al algorithms that enable acceleration of data acquisition and processing, and also supported by neural networks.

[0046] Still referring to Figure 5, schematically, a navigation system 100 comprising the navigation instrument 10 may also comprise:

[0047] - at least one magnet 20, three in the example in Figure 5, to ensure the correct orientation of other components, for example, toward geographic North,

[0048] - at least one device 60, which operates using GPS (Global Positioning System) satellite technology, and- a terminal 70 for monitoring the acquired data, which in turn is equipped with a control panel and associated software. Specifically, the control software manages the inertial measurement unit (IMU) using a serial and / or Ethernet port for testing, configuring, and recording the IMU, and allows for real-time 2D and 3D data viewing.

[0049] Navigation instrument 10 is capable of ultra-low-latency measurements and estimates thanks to the integration of the Al-supported processor 50, which ensures minimal or almost no delay between the measured event and the transmission of the information.

[0050] Direction determination is performed using the gyroscopic effect. The operating principle is based on one or more high-speed gyroscopes, which maintain their rotation axis oriented toward true North through conservation of angular momentum.

[0051] The heart of the navigation instrument 10 is its miniaturized processor 50. Thanks to the processor's integration of Al processing architecture, dynamic compensation, and yaw estimation, Navigation Instrument 10 offers advantages that address the major challenges encountered in the current state of the art.

[0052] First, its processor 50 is at least an order of magnitude faster than known products. This is because artificial intelligence optimizes data processing, improving responsiveness and precision. As a result, this speed increase allows for extremely reliable monitoring of extreme navigation conditions, such as wave motion.

[0053] In the GNSS-denied option (no GPS signal), the processor 50 implements an advanced "Sensor Fusion" software architecture supportedby artificial intelligence, configured to maximize accuracy in both dynamic and static conditions, operating with a sampling rate of 5 kHz, a cycle time of < 1 ms, and 20% CPU utilization.

[0054] To ensure the system's dynamic robustness and mitigate intrinsic drift, the processor uses a neural network specifically designed to compensate for the numerical drift error typical of digital systems. This error is compensated periodically, preferably every 100 or so calculation steps. The neural network is trained to approximate the numerical integration error as a function of the number of integration steps and the system's inputs and outputs over a short time window (approximately 100 steps). This estimated error represents the approximation error of the mathematical integration model used. The residual error after this compensation therefore represents the sole contribution of sensor measurement errors, which, over a limited range of 100 steps, contribute extremely little and with an average value tending to zero. This mechanism significantly mitigates instrument drift during motion, maintaining high accuracy even in GNSS-denied conditions.

[0055] In parallel with the dynamic compensation described above, the system includes a "Bias Adaptation" module for managing steady-state conditions and yaw correction. This module uses artificial intelligence algorithms to recognize "static windows" (time intervals in which the instrument is stationary or motion is negligible). During these windows, the algorithm learns and dynamically updates the gyroscope bias, refining the Kalman filter parameters in real time.

[0056] The synergy between numerical integration error compensation(dynamically operating every 100 steps) and bias adaptation (operating on static windows) allows the system to provide extremely stable "corrected yaw" output, with roll and pitch accuracy in the order of 0.01° RMS and near-zero drift, creating a robust and autonomous navigation system.

[0057] As a further consequence of increased efficiency and simplified hardware, the navigation instrument 10 can be made smaller, for example, at least three times smaller than known gyroscopic instruments.

[0058] Furthermore, the smaller size and increased speed allow for lower costs: a reduced need for materials and simpler construction make the navigation instrument more economical.

[0059] Finally, thanks to the control panel software of the terminal 70, remote calibration of the navigation instrument is possible without the need to send the device to the manufacturer annually or biennially for on-site recalibration.

[0060] The processor 50 is equipped with algorithms based on the Kalman filter and adaptive filters supported by artificial intelligence. The Kalman filter is a recursive algorithm for state estimation based on matrix inversion. It is used in navigation systems and gyroscopic sensor fusion but presents computational difficulties in real-time applications.

[0061] An adaptive filter, on the other hand, allows for dynamic parameter adjustments based on predefined inputs, resulting in faster processing for rapid adaptation to varying frequencies. It is conveniently applied when noise cancellation / attenuation, signal prediction, and real-time calculations are required.

[0062] The processor 50, supported by Al acceleration algorithms, alsofeatures integrated support for vector instructions that enable improved signal processing and neural networks. Its outputs are optimized through access to Al libraries that process digital signals and neural networks.

[0063] The choice of Al-powered processors for filters allows for a significant acceleration of computational tasks, thanks to vector instructions that improve efficiency.

[0064] Additionally, access to Al libraries that process digital signals simplifies and accelerates operations such as matrix inversion and signal processing. This improves the efficiency of the Kalman filter with fast matrix operations and real-time noise reduction, thereby accelerating signal processing and filtering for better performance.

[0065] Access to Al-based neural networks supports adaptive filters with dynamic learning for more precise noise cancellation. It also enables faster data processing for rapid adaptation to varying frequencies.

[0066] Finally, Al acceleration algorithms reduce computational latency, enabling higher sampling rates and improved responsiveness. Furthermore, they allow for compensation for low-resolution sensors while maintaining high precision and reliability.

[0067] All this makes the miniaturized 50 processor faster and more efficient than general-purpose processors.

[0068] In essence, the innovation of the navigation instrument 10 according to the present invention is based on the replacement of traditional analog gyroscopes with digital gyroscopes 30 and on the use of a miniaturized processor 50 (MCU) equipped with artificial intelligence algorithms and support for neural networks for high-speed processing.Compared to traditional products, the navigation instrument 10 features a simplified electronics design, eliminating the use of analog gyroscopes and analog-to-digital converters. It also offers costeffectiveness by reducing the overall number of components. Specifically:

[0069] - the transition from analog to digital gyroscopes eliminates the need for expensive components such as analog-to-digital converters and high-precision analog gyroscopes,

[0070] - concentrating all processing into a single, miniaturized processor with Al enhancement reduces overall hardware costs by up to four times, and

[0071] - eliminating analog components also eliminates the need for periodic factory calibrations, further reducing maintenance and after-sales costs.

[0072] The navigation instrument 10 also features smaller dimensions and lower power consumption thanks to a compact design with fewer peripherals. Fewer components and simplified circuitry result in a smaller printed circuit board (PCB) footprint. The approximately fourfold reduction in size compared to existing systems makes this navigation instrument more suitable for space-constrained applications such as underwater drones, aerial drones, and loT devices.

[0073] The use of artificial intelligence and neural networks compensates for the lower resolution of digital gyroscopes and enables real-time noise reduction and accurate state estimation. It also improves sensor performance in terms of:

[0074] - precision and speed: maintaining high sensor performance withrapid dynamic response,

[0075] - robust noise reduction thanks to Al-accelerated processing.

[0076] Experimental tests show significant improvements in the performance of the navigation instrument 10 compared to existing systems:

[0077] - 90% reduction in calculation times for Kalman filters.

[0078] - 90% improvement in real-time response for adaptive filters, - optimized energy efficiency for loT and drone use: power consumption is less than 1 W for the standard configuration and less than 0.8 W for the GNSS configuration, with a supported supply voltage range of 6.5 - 36 V.

[0079] Furthermore, the navigation instrument 10 enables:

[0080] - precise measurement of wave motion with an accuracy of 5 cm or 5%, essential for marine applications,

[0081] - measurement of Euler angles, i.e., roll, pitch, and yaw, the three angles that describe the orientation of a rigid body in space with respect to a fixed coordinate system,

[0082] - measurement of linear velocities and linear accelerations, - measurement of wave motion for marine applications,

[0083] - resistance to magnetic disturbances,

[0084] - precise and consistent course management,

[0085] - extremely precise azimuth determination,

[0086] - integration with navigation systems,

[0087] - integration with stabilization systems,

[0088] - prolonged resistance to adverse weather conditions.This innovation allows expensive analog gyroscopes to be replaced with a digital solution, in which artificial intelligence compensates for the lower resolution of MEMS sensors by drastically reducing calculation times (up to 90% for Kalman filters) while maintaining high performance. The availability of the Core PCB module (OEM) and compliance with maritime standards extend its applicability from underwater drones to commercial shipping.

[0089] In addition to the embodiment of the invention, as described above, it should be understood that numerous other variations exist. It should also be understood that such embodiments are only exemplary and do not limit either the scope of the invention, nor its applications, nor its possible configurations. On the contrary, although the description above allows the person skilled in the art to implement the present invention at least according to one of its exemplary embodiments, it should be understood that many variations of the described components are possible, without thereby departing from the scope of the invention, as defined in the appended claims, which are interpreted literally and / or according to their legal equivalents.

Claims

CLAI M S1. Navigation instrument (10), of the type for applications in loT devices, in the industrial sector and in the transport sector, the navigation instrument comprising:- at least one digital gyroscope (30), configured to detect and / or measure roll, pitch and yaw angles,- at least one accelerometer (40), configured to detect and / or measure accelerations along three main axes, calculating the detected force with respect to the mass of the object,- a miniaturized processor (50),the navigation instrument (10) being characterized in that- it is free of analog gyroscopes; and- the processor (50) is supported by artificial intelligence algorithms and neural networks, configured to enable the acceleration of the data acquisition and processing processes from the digital gyroscope (30) and the accelerometer (40).

2. Navigation instrument (10) according to claim 1, further comprising:- a first printed circuit board (11) integrating the at least one digital gyroscope (30) and the at least one accelerometer (40),- a second printed circuit board (12) comprising the processor (50), - a casing (13),- a cover (14), and- a connector (15) for connection to external wiring for data transmission and power supply.

3. Navigation instrument (10) according to claim 1 or 2, wherein the artificial intelligence algorithms are configured to execute a sensor fusion algorithm.

4. Navigation instrument (10) according to claim 3, wherein the sensor fusion algorithm comprises a neural network-based numerical drift compensation module, the neural network being trained to estimate an approximation error of the numerical integration and wherein the processor (50) is configured to subtract the estimated approximation error from the orientation calculation with cyclic periodicity.

5. Navigation instrument (10) according to claim 4, wherein the cyclic compensation is performed approximately every 100 calculation steps.

6. Navigation instrument (10) according to claim 3, 4, or 5, wherein the sensor fusion algorithm further comprises a bias adaptation module configured to:- identify time intervals of stationarity; and- performing, during the stationary time intervals, a procedure for recalculating and updating the bias parameters of the at least one gyroscope (30) for yaw angle correction.

7. Navigation instrument (10) according to any of the preceding claims, configured to operate with a sampling frequency of 5 kHz and to ensure a cycle execution time of less than 1 ms.

8. Navigation instrument (10) according to any of the preceding claims, configured to provide orientation data with a roll and pitch accuracy less than or equal to 0.01° RMS and a wave motion estimate with anaccuracy equal to the greater of 5 cm and 5% of the wave amplitude.

9. Navigation instrument (10) according to any preceding claim, further comprising an integrated GNSS receiver module or a serial communication interface for an external GNSS receiver.

10. Navigation instrument (10) according to any preceding claim, constructed as an electronic module integrated on a printed circuit board without an external casing (PCB Core), for OEM integration.

11. Navigation system (100) comprising a navigation instrument (10) according to any of claims 1 to 10, the navigation system further being provided with:- a plurality of magnets (20) to ensure the correct orientation of other components,- at least one device (60) configured to operate according to GPS satellite technology, and- a terminal (70) for monitoring the acquired data and provided with a control panel and control software.

12. Navigation system (100) according to claim 11, wherein the control software is configured to:- manage an inertial measurement unit (IMU) using a serial port or an Ethernet interface for configuration and logging;- allow real-time display of 2D and 3D data; and- allow remote calibration of the navigation instrument (10).

Images