Drone-based description and multi-physics measurement of samples

Abstract

Disclosed and described herein are systems, methods and device for drone-based multi-physics measurement and acquisition wherein a drone flies above samples to rapidly acquire a large volume of data (standard photos, multispectral images, acoustic reflection data, and electromagnetic reflection data, among others). By integrating data acquired by the conventional camera, multispectral camera, acoustic / electromagnetic systems, and other sensor systems, accurate information on lithology, composition, elastic properties, and the like of the samples can be obtained without the need for time-consuming manual analysis.

Description

Docket No. 10046-638WO1 Client Ref: 8434 TOR DRONE-BASED DESCRIPTION AND MULTI-PHYSICS MEASUREMENT OF SAMPLES CROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to and benefit of U.S. Provisional Patent Application Serial No. 63 / 683,813 filed August 16, 2024, which is fully incorporated by reference and made a part hereof. BACKGROUND

[0002] In the field of subsurface energy / resources exploration, accurate description and classification of rocks is crucial for optimal well and fluid / resource productivity. However, precise and fast rock classification and quantification to this day remains a significant technical challenge.

[0003] Conventional approaches to rock description / quantification are generally time-consuming and labor-intensive due to manual data processing and interpretation. Rock classification plays a key role across various industries, including mining, petroleum and numerous other fields. Traditional rock classification process has largely relied on visual assessments by geologists and mineralogists or labor-intensive laboratory tests (Chatterjee et al., 2010). Consequently, these types of manual classification are not only time-consuming but typically yield only a coarse approximation due to limited sampling (Guyot et al., 2004). To address these limitations, over the past two decades, efforts have been directed toward the development of automatic rock classification technologies capable of continuous monitoring throughout mining operations. Perez et al. (1999) focused on color image-based methods with neural networks (NNs) for classifying a limited set of rock types. Casali et al. (2001) introduced genetic algorithms for feature selection, leading to modest improvements in classification. Other approaches combine texture information and co-occurrence likelihoods for rock classification (Paclik etDocket No. 10046-638WO1 Client Ref: 8434 TOR al., 2005). Most effective models for general applications demand extensive datasets, often exceeding 300,000 traced examples (Lin et al., 2014; He et al., 2017).

[0004] A promising advantage in the context of automated rock element classification is the fact that geological materials are composed of minerals, each with distinct reflectance characteristics (Bishop et al., 2018) and birefringence properties (Cesare et al., 2022). The use of a multispectral camera, sensitive to a broader spectrum of color bands than the human eye can perceive, serves to capture these spectral properties within images. Consequentially, this approach reduces the volume of training examples required for the development of accurate classification models (Soomro et al., 2017).

[0005] Similarly, other fields such as farming, road / bridge / highway construction, general construction, groundwater, geothermal, environmental projects, mining, CO2- sequestration operations and others rely on quick yet accurate assessments of samples.

[0006] Therefore, what is desired is methods, systems and devices for gathering high-resolution data and analyzing sample properties in a rapid, economic, and accurate manner for quick turnaround that overcome challenges in the art, some of which are described above. SUMMARY

[0007] Disclosed and described herein are systems, methods and devices comprising drone-based systems equipped with advanced multi-physics sensors to gather high-resolution data from samples and analyze sample properties in a rapid, economic, and accurate manner for quick turnaround.

[0008] In one aspect, the disclosed systems, methods and devices address technicalin the context of rock classification and description / quantification by employing a multi-physicsDocket No. 10046-638WO1 Client Ref: 8434 TOR approach coupled with machine learning methods. Data collected by drones equipped with a range of multi-physics sensors tailored for specific rock property estimations is integrated to determine properties of the rock. In more detail, the disclosed systems, methods and devices comprise the combination of unmanned aerial vehicle (UAV) with several multi-physics sensors such as, for example, a regular camera, a multispectral camera, and an acoustic / electromagnetic system in order to achieve a complete rock description / quantification where the task of the regular camera is to acquire images of rock samples; these images are used as a training dataset for rock classification generated through a deep-learning algorithm. The purpose of the multispectral camera is to capture reflectance properties, the acoustic sensor is used for the estimation of rock elastic properties, while the electromagnetic sensor estimates dielectric properties. The deployment of drones facilitates and speeds up the acquisition process of the above rock properties.

[0009] In other aspects, the disclosed systems, methods and devices are employed for samples beyond rock samples, including concrete, cement, asphalt, soil, and the like.

[0010] The foregoing illustrative summary, as well as other exemplary objectives and / or advantages of the disclosure, and the manner in which the same are accomplished, are further explained within the following detailed description and its accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Various other objects, features and attendant advantages of the present invention will become fully appreciated as the same becomes better understood when considered in conjunction with the accompanying drawings, in which like reference characters designate the same or similar parts throughout the several views, and wherein:Docket No. 10046-638WO1 Client Ref: 8434 TOR

[0012] FIG. 1 is an overview illustration of the disclosed embodiments.

[0013] FIG. 2 is an image of an exemplary drone that can be used to implement aspects disclosed herein.

[0014] FIG. 3 illustrates examples of some rock samples acquired with a conventional camera.

[0015] FIGS. 4A and 4B illustrate an example of the segmentation algorithm where FIG. 4A illustrates an image before the segmentation and FIG. 4B illustrates an image after the application of the segmentation algorithm.

[0016] FIG. 5 illustrates additional details of the image processing procedure.

[0017] FIG. 6 illustrates steps in the training of an exemplary machine-learning algorithm for identifying properties of samples.

[0018] FIG. 7A illustrates a schematic of an exemplary drone-based system comprising a multispectral cameras, where the multispectral camera attached to the drone acquires multispectral images of the rock, and these images are converted in reflectance curve via signal processing.

[0019] FIG. 7B illustrates optical responses of rocks at specific wavelengths.

[0020] FIG. 8A illustrates matrix storage of multispectral data.

[0021] FIG. 8B illustrates exemplary structure of the images acquired using the multispectral camera.

[0022] FIG. 9 illustrates rock characterization of different rock types based on their spectral profiles.

[0023] FIGS. 10A and 10B illustrate a schematic of the acoustic drone system where the acoustic transducer emits an acoustic wave that is reflected by the rock surface.

[0024] FIGS. 11A and 11B illustrate examples of normalized reflection strength measured for five different rock types.Docket No. 10046-638WO1 Client Ref: 8434 TOR

[0025] FIG. 12A shows time-domain acoustic reflections recorded from aluminum, sandstone (hard rock), and a soft sample where the amplitude of the reflected wave varies with material type due to differences in acoustic impedance.

[0026] FIG.shows maximum amplitude extracted from each waveform of FIG. 12A.

[0027] FIGS. 13A and 13B illustrate examples of a custom-designed support platform

[0028] FIG. 14 shows an example computing environment in which example embodiments and aspects may be implemented. DETAILED DESCRIPTION

[0029] Before the present methods and systems are disclosed and described, it is to be understood that the methods and systems are not limited to specific synthetic methods, specific components, or to particular compositions. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0030] As used in the specification and the appended claims, the singular forms “a,” “an” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” one particular value, and / or to “about” another particular value. When such a range is expressed, another embodiment includesfrom the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by use of the antecedent “about,” it will be understood that the particular value forms another embodiment. It will be further understood that the endpoints of each of the ranges are significant both in relation to the other endpoint, and independently of the other endpoint.

[0031] “Optional” or “optionally” means that the subsequently described event or circumstance may or may not occur, and that theDocket No. 10046-638WO1 Client Ref: 8434 TOR description includes instances where said event or circumstance occurs and instances

[0032] Throughout the description and claims of this specification, the word “comprise” and variations of the word, such as “comprising” and “comprises,” means “including but not limited to,” and is not intended to exclude, for example, other additives, components, integers or steps. “Exemplary” means “an example of” and is not intended to convey an indication of a preferred or ideal embodiment. “Such as” is not used in a restrictive sense, but for explanatory purposes.

[0033] Disclosed are components that can be used to perform the disclosed methods and systems. These and other components are disclosed herein, and it is understood that when combinations, subsets, interactions, groups, etc. of these components are disclosed that while specific reference of each various individual and collective combinations and permutation of these may not be explicitly disclosed, each is specifically contemplated and described herein, for all methods and systems. This applies to all aspects of this application including, but not limited to, steps in disclosed methods. Thus, if there are a variety of additional steps that can be performed it is understood that each of these additional steps can be performed with any specific embodiment or combination of embodiments of the disclosed methods.

[0034] The present methods and systems may be understood more readily by reference to the following detailed description of preferred embodiments and the Examples included therein and to the Figures and their previous and following description.

[0035] Expedient, accurate, and reliable description and quantification of samples such as rock core samples is paramount for subsurface engineering projects which to date remains a significant technical challenge. The disclosed embodiments of systems, devices and methods addresses the related challenges by employing a multi-physics approach coupled with machine learningDocket No. 10046-638WO1 Client Ref: 8434 TOR methods. An integration of measurements are collected by drones equipped with a range of sensors designed for specific rock property estimations. In more detail, the disclosed embodiments comprises a combination of unmanned aerial vehicle (UAV) with four or more sensors, including for example a regular optical camera, a multispectral camera, a pulse acoustic system (sonar), and an electromagnetic system (radar) in order to achieve a complete rock description. The task of the regular camera is to acquire images of rock samples; these images are used as a training dataset for a rock classification method generated through a deep learning algorithm. The drone-based multi-physics acquisition and interpretation method is capable of describing tens of feet of rock core samples within a few minutes, thereby reducing analysis times and enabling the selection of important rock samples for subsequent detailed laboratory analysis.

[0036] FIG. 1 is an overview illustration of the disclosed embodiments. In FIG. 1, an unmanned aerial vehicle (i.e., a “drone”) 102 is equipped with a plurality of multi-physics measurement systems including sensors 104. The measurement systems may comprise, for example, optical (visible, ultraviolet and infrared ranges) systems including one or more conventional cameras and / or one or more multi-spectral cameras; acoustical systems such as for example sonar (kHz to ultrasonic frequencies), electromagnetic systems (e.g., radar, LiDAR, and terahertz radiation), photon radiation systems, and the like. The drone 102 may be equipped with multiple sources and sensors 104 for sensor- array detection. Although it is to be appreciated that in some instances the drone 102 may be equipped with a subset of the above listed measurement systems.

[0037] The drone 102 flies over one or more samples 106, using the multi-physical measurements systems and sensors 104 to acquire information about the one or more samples 106. In some instances, the drone 102 is equipped with a plurality of measurement systems when it flies over the one or more samples 106. In some instances,Docket No. 10046-638WO1 Client Ref: 8434 TOR the measurement systems may be interchangeable and the drone 102 flies multiple paths over the one or more samples 106 with different measurement systems. In some instances, the drone 102 may be equipped with only one measurement system for each flight over the one or more samples 106.

[0038] The drone 102 is in communication with a measurement / processing / interpretation system 108. Data acquired by the multi-physical measurements systems and sensors 104 is transmitted from the drone 102 to the measurement / processing / interpretation system 108 using a wireless protocol such as Blue Tooth or the like. In some instances, data, including for example control signals, may be transmitted from the measurement / processing / interpretation system 108 to the drone 102.

[0039] In some instances, the one or more samples 106 may comprise rock samples. The measurement / processing / interpretation system 108 processes and fuses all the acquired multi-physics measurements to estimate compositional (fluids and solids) properties of rocks (types of fluids and types of solids, including chemical elements, chemical compounds, minerals, and their relative concentrations(fluid storage (e.g., porosity)) and flow (e.g., permeability) properties of rocks, and elastic (e.g., P- and S-wave velocity, anisotropy, density) and mechanical (e.g., Young’s modulus, Poisson’s ratio, Lame’s parameter, bulk modulus, etc.) properties of rocks. In some instances, the measurement / processing / interpretation system 108 performs a quality control (QC) check of the received data and interprets all the available data in almost real time to transform the measurements into continuous displays of rock properties, and classifies rock segments into coincidental groups (i.e., rock classes sharing similar properties). In some instances, the measurement / processing / interpretation system 108 has an intelligent feedback loop to modify the drone’s trajectory and speed in order to acquire more detailed measurements in placesDocket No. 10046-638WO1 Client Ref: 8434 TOR where rocks exhibit higher spatial variability in their properties.

[0040] The disclosed systems provides rapid surveillance of rock fragments stemming from drilling operations (e.g., rock cutting whole core, slabbed core, etc.) either at the well site or once the rock fragments have been transported and laid out in a warehouse.

[0041] In some instances, the one or more samples 106 may comprise concrete, cement, soil, and the like. Applications are in drilling for subsurface energy resources such as hydrocarbons, lithium, geothermal, groundwater, ore mining, civil engineering infrastructure (e.g., concrete / cement, soil, road probes), environmental (e.g., soil remediation, bore samples), and agriculture (soil samples).

[0042] The embodiments described herein provide high speed and low cost with which measurements can be acquired, processed, and interpreted, e.g., tens of feet of rock samples in less than 10 minutes, compared to months of data acquisition and interpretation in the laboratory and tens of thousands of dollars in expenses. EXAMPLES

[0043] In one example of a method of determining properties of a sample, a drone 102 such as the one shown in FIG. 2 that is equipped with one or more multi-physical sensor systems flies above core samples obtaining data that is used to make determinations about the core samples. For example, the drone 102 may comprise 1 DJI Mini 3 drone (DJI, Shenzhen, China), though other drones are contemplated withing the scope of this disclosure. In one aspect, the drone 102 comprises a compact lightweight camera drone and includes a conventional camera (the features of the camera are discussed below). Doing so rapidly acquires high-resolution images of the core samples (FIG. 3). Alternatively and / or additionally, the drone 102 may also comprise several attached sensors and flies above the rock core samples (see FIG. 3) in order to acquireDocket No. 10046-638WO1 Client Ref: 8434 TOR several multi-physics measurements and to speed up the acquisition process and obtain a full quantitative description ofproperties.

[0044] In some instances, the conventional camera mounted on the drone may be a 1 / 1.3-inch CMOS with effective pixels of 48 MP, though other cameras with different resolutions may be used. The images acquired by the camera are subjected to image processing techniques using the measurement / processing / interpretation system 108. In particular, an algorithm that combines a watershed segmentation with a neighboring search to enable accurate segmentation of rock core samples from the background can be used in order to enhance the quality of the training dataset for subsequent rock classification. FIGS. 4A and 4B show an example of the application of the algorithm. The algorithm identifies the boundaries of the rocks 106 by treating the image as a topographic surface and "flooding" it from the minimum values, effectively removing the background and isolating the rock 106 features. FIG. 5 is a more detailed illustration of the image processing procedure.

[0045] Machine learning (e.g., Convolutional Neural Network (CNN), in particular a 2D convolutional UNET architecture for image translation and classification) may be used for image segmentation of the images acquired by the conventional camera using the measurement / processing / interpretation system 108. The U-Net architecture is effective for pixel classification, making it ideal for predicting the class of each pixel in the segmented images. The training of the U-Net model is based on a dataset comprising images of rock samples with known classifications, leveraging prior knowledge to enhance the accuracy of the predictions. This training process involves feeding the segmented images into the U-Net, allowing it to learn the distinct features and patterns associated with different rock types, as shown in FIG. 6.Docket No. 10046-638WO1 Client Ref: 8434 TOR

[0046] Alternatively, and / or additionally, the drone may be equipped with a multispectral camera system. In this exemplary method, rock characterization may be performed by the measurement / processing / interpretation system 108 using multispectral images of the core samples. Though other multispectral cameras may be used, in one instance, the multispectral camera comprises a Monarch Pro camera (Unispectral, Ramat Gan, Israel), which allows the capture of images at, for example, seven frequency bands. This camera has the capability of capturing images in 10 Near Infrared (NIR) bands, specifically from 680nm to 940nm. The multispectral camera mounted in the drone is used to acquire multispectral images (one for each wavelength), where the subsequent step is to extract spectral features from the multispectral data for each rock sample using the measurement / processing / interpretation system 108. These features include spectral reflectance values across specific spectral bands. By analyzing the spectral characteristics using the measurement / processing / interpretation system 108, distinctive reflectance properties of different rock elements can be identified.

[0047] FIG.7A shows a schematic of the measurement acquisition and processing method necessary to convert the multispectral images into a reflectance curve. The camera is mounted on the drone in order to fly above core sample and acquire multispectral images. The intensity of the multispectral images is related to the reflective properties of the rocks (samples). As shown in FIG. 7B, different rocks have different optical response at specific wavelengths. This can be used to classify the rock samples.

[0048] As shown in FIG. 8A, data is stored as a matrix of a digital number. Each pixel has a location (I, j) and a reflectance value. The I and j for a pixel are the same in all bands, but the pixel value may vary from band to band. In one particular example as shown in FIG. 8B, the acquired multispectral images comprise sevenDocket No. 10046-638WO1 Client Ref: 8434 TOR matrices, one for each frequency, where each cell represents the reflectance of the rock at a specific wavelength. Once the images are acquired it may be necessary to perform some image processing using the measurement / processing / interpretation system 108 in order to mitigate the effects of external noise (such as ambient light). This is achieved by using a black stripe in the acquired images. The reflectance of the rock is calculated using the following formula:

[0049] This formula allows for the normalization of the rock's intensity values against the black stripe's intensity, effectively reducing noise and providing more accurate reflectance data for each pixel.

[0050] The calibrated reflectance images are then analyzed to extract spectral signatures of the rocks. By comparing the reflectance data across the seven frequency bands, different rock types can be identified and characterized based on their spectral profiles, as shown in FIG. 9.

[0051] Alternatively, and / or additionally, the drone may be equipped with an acoustic system 1000. For example, the acoustic system may comprise a waveform generator, a lithium battery, and a Murata air transducer (MA40S4R) having a frequency of 40 kHz.

[0052] , though other acoustic systems are contemplated within the scope of this disclosure. The acoustic transducer emits an acoustic wave that is reflected by the rock (sample 106) surface.

[0053] As shown in FIG. 10A, the acoustic system 1000 records ultrasonic reflections from the rock samples 106. The recorded acoustic reflections are analyzed by the measurement / processing / interpretation system 108 to compute the acoustic impedance of the rock sample and calculate the P-wave velocity, Vp, of the rock sample via the formula:Docket No. 10046-638WO1 Client Ref: 8434 TOR Where Z is the acoustic impedance (function of the acoustic reflection) and ^ is the density of the rock sample.

[0054] FIG. 10B illustrates exemplary on-board circuitry of an acoustic system 1000. The circuit (FIG. 10B) is divided into three sections (A, B and C) based on function. In the following paragraphs, every section (A, B and C) is described individually.

[0055]

[0056] Section A of the circuit generates a square wave that activates the acoustic piezoelectric source. The technical name of the circuit is “Astable 555 Timer”. At the output of the circuit, it generates a square wave with the frequency of: ^^ ^^ ^ ^^^^^ ^ ^^^ ^ ^^^(2) With a square wave duty cycle of:

[0057] Section B

[0058] Section B of the circuit turns 2 alternating wave signals into two separate direct constant smooth signals. They are technically known as full-wave rectifiers. They work by installing a network of diodes to obtain an electrical node that is positively charged regardless of the source’s polarity. In the case of the signal emitted by the receiver acoustic piezoelectric transducer, the circuit works as an absolute value function. To smooth the resultant signal, a capacitor is applied to the output of the circuit. The capacitor stores excess signal and discharges it when the signal is close to zero volts. The latter ability makes the circuit a maximum amplitude signal processor.

[0059] Docket No. 10046-638WO1 Client Ref: 8434 TOR

[0060] Section C of the circuit amplifies 2 separate direct current voltage by using 2 separate operational amplifiers. They are technically known as non-inverting amplifiers. The core part of the circuit is the operational amplifier. With proper wiring, it equates the voltages at its inverting and non-inverting inputs and restricts current flow through the inputs. The voltage on the inverting inputs becomes equal to the signal drawn from the full- wave rectifier. The inverting input is connected to the ground through a resistor from the clipped potentiometer. A current of the following amplitude is formed:Because the current is not allowed to go through the amplifiers, the current is bypassed from the inverting input to the output using a resistor. The resultant voltage drop isThe output voltage becomes ^^ !^^ ^ ^^ ^ ^^^^ ^^ !^^ ^ ^^(6) The following is a simplified gain function to estimate the amplification ratio based on the values of resistors

[0061] In other aspects, the drone may alternatively and / or additionally be equipped with electromagnetic systems (e.g., radar, LiDAR (light detection and ranging), and terahertzDocket No. 10046-638WO1 Client Ref: 8434 TOR radiation), photon radiation systems, and the like. For example, in some instances the drone 102 may be equipped with a LiDAR system. In one instance, the LiDAR system may comprise a time- of-flight (TF) Mini sensor integrated with an Arduino microcontroller for real-time data acquisition (FIGS. 11A, 11B), though other LiDAR systems are contemplated within the scope of this disclosure. The TF Mini sensor emits near-infrared light pulses and records the time-of-flight and signal intensity of the reflected beam from the rock surface. Since each rock type exhibits a distinct reflectance profile depending on its surface texture, composition, and optical properties. By normalizing the return intensity values, it is possible to distinguish materials based on their reflected light strength. FIGS. 11A and 11B present the normalized reflection strength measured for five different rock types. The variation in intensity reflects material-dependent optical properties, confirming that LiDAR can serve as a rapid, non-contact method for rock classification.

[0062] The reflected signal strength is used as a proxy for the dielectric permittivity of the rock surface. Because materials with higher dielectric constants tend to reflect more energy, the recorded LiDAR intensity provides a preliminary means for distinguishing between rock types based on their electromagnetic properties. In order to obtain reliable results, it may be necessary to perform calibration using known reference materials to convert raw LiDAR intensity into permittivity estimates. This system offers a fast, non-contact method for characterizing surface properties.

[0063] Preliminary Results

[0064] To demonstrate the capability of the acoustic system in distinguishing between different materials, measurements were conducted on three distinct targets: aluminum, a soft sample, and a sandstone. The system successfully recorded ultrasonic reflections from each target, and clear differences in theDocket No. 10046-638WO1 Client Ref: 8434 TOR reflection coefficients were observed. FIGS. 12A and 12B show the results.

[0065] FIG. 12A shows time-domain acoustic reflections recorded from aluminum, sandstone (hard rock), and a soft sample. The amplitude of the reflected wave varies with material type due to differences in acoustic impedance. FIG. 12B shows maximum amplitude extracted from each waveform. F12A and 12B show that the acoustic system can effectively differentiate materials based on their reflection response, validating its potential for rapid, non-contact classification of exposed rock surfaces in field or laboratory settings.

[0066] Platform for the instruments

[0067] To accommodate the required circuits, a custom support platform was developed. This platform ensures stable housing for the electronics during both flight and stationary operation. The structure is constructed using thin-walled PVC pipes, offering a lightweight framework that holds the drone upright when not in flight and securely carries the circuit components. FIGS. 13A and 13B show an example of the designed platform. CONCLUSIONS

[0068] This above examples demonstrate the effectiveness of using a drone equipped with both conventional and multispectral cameras for rock classification and characterization. The conventional camera provided an initial classification accuracy of approximately 85%, enabling a robust first approach to identifying rock types. The multispectral camera allowed for detailed spectral analysis, revealing lithological changes within the frequency range of 713 to 830 nm.

[0069] In particular, the lowest frequency range (713-730 nm) showed enhanced sensitivity, penetrating deeper and providing more detailed lithological variations.Docket No. 10046-638WO1 Client Ref: 8434 TOR

[0070] Advantageously, the disclosed drone-based multi-physics measurement acquisition system. In comparison, in the oil and gas industry the drone-technologies used generally employ only one type of sensor. The disclosed drone-based multi-physics measurement acquisition systems include combinations of a conventional camera, a multispectral camera, and an acoustic / electromagnetics system may provide a comprehensive, multi-physics, and multi-dimensional approach to rock description and quantification.

[0071] Furthermore, conventional methods often involve the use of specific separate tools and techniques for data collection, processing, and analysis. The disclosed drone-based system instead offers a multi-physics approach integrating multiple sensors into a single drone, avoiding the need of using different machines to achieve full rock description / quantification. Also, conventional approaches to rock-core description are generally time-consuming and labor-intensive due to manual data processing and interpretation. The disclosed drone-system reduces the acquisition and the processing time.

[0072] Additional advantages of the disclosed embodiments include the following: 1. There is a substantial commercial need to perform fast and accurate measurements of rock core samples, especially on as- received samples from borehole drilling. There are no measurements available on site for this purpose, which makes it difficult to make real-time decisions concerning perforations for fluid production and directional drilling. The drone-based measurement system can be readily implemented on-site, including onshore and offshore operations around the world. Likewise, the drone-based measurement system is capable of acquiring a multitude of measurements (different physics such as acoustic, electromagnetic, nuclear, spectroscopy, optical, etc.) which can be blended into accurate rock / fluid descriptions based on machine-learning algorithms.Docket No. 10046-638WO1 Client Ref: 8434 TOR 2. The method can be readily implemented not only for hydrocarbon operations but also for groundwater, geothermal, environmental projects, mining, and CO2-sequestration operations. 3. The method is also ideal for rock-core repositories, where fast and extensive measurements are needed for fast turn- around measurements and diagnostics. Current measurement systems require that the samples be transported to specialized laboratories and that the samples be slowly scanned by expensive machines which also require expensive maintenance. The process is time-consuming and onerous. 4. The fast development of drones makes it feasible to constantly augment the types of measurements and sensors / actuators used to assess rock properties. It is also possible to constantly improve the pre-processing and interpretation of measurements based on machine learning because of the addition of measurements for algorithm training. 5. The method enables the real-time implementation of ad- hoc measurement acquisition strategies across rock segments which are found to be very heterogenous or which exhibit strong variability of properties. This is not possible with alternative methods (no intelligent measurement acquisition). COMPUTING ENVIRONMENT

[0073] FIG. 14 shows an example computing environment in which example embodiments and aspects may be implemented. The computing device environment is only one example of a suitable computing environment and is not intended to suggest any limitation as to the scope of use or functionality. The computing environment of FIG. 14 may be a computing device 500 used by a controller, and / or other hardware aspects of the disclosure, to implement aspects of the disclosure. For example, computing device 500 may be a component of or comprise a cloud computing and storage system. Computing device 500 may comprise all or a portion of a server or a controller. In one aspect, computing device 500 comprises allDocket No. 10046-638WO1 Client Ref: 8434 TOR or a portion of measurement / processing / interpretation system 108. Each computing device may have one or more processors. In various implementations, computing devices 500 used by the various parties may be interconnected with one another through various connections, including networks. Such networks may be wired (including fiber optic), wireless, or combinations thereof including parallel, RS-232 (all serial communication from point to point), Visual (Infra-Red), audible (modem for example). Other connections / communication standards can also be improved such as USB, PCI Express, Firewire, Fiber Channel, HDMI, I2C, SPI, etc. Some of these are important for applications such as wireless barcode readers, Credit Card Terminals, walkie-talkies, radios, video conferencing, etc.

[0074] Numerous other general purpose or special purpose computing devices environments or configurations may be used. Examples of well-known computing devices, environments, and / or configurations that may be suitable for use include, but are not limited to, personal computers, server computers, handheld or laptop devices, multiprocessor systems, cloud-based systems, microprocessor-based systems, network personal computers (PCs), minicomputers, mainframe computers, embedded systems, Internet of Things devices, network switches, network routers, network edge devices, Modulator-demodulators (modems), industrial control equipment, including distributed computing environments that include any of the above systems or devices, smart phones or smart devices, and the like.

[0075] Computer-executable instructions, such as program modules, being executed by a computer may be used. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. Distributed computing environments may be used where tasks are performed by remote processing devices that are linked through a communications network or other data transmission medium. In a distributedDocket No. 10046-638WO1 Client Ref: 8434 TOR computing environment, program modules and other data may be located in both local and remote computer storage media including memory storage devices.

[0076] With reference to FIG. 14, an example system for implementing aspects described herein includes a computing device, such as computing device 500. In its most basic configuration, computing device 500 typically includes one or more processing units 502 and one or more memory 504. Depending on the exact configuration and type of computing device, memory 504 may be volatile (such as random-access memory (RAM)), non-volatile (such as read-only memory (ROM), flash memory, etc.), or some combination of the two. This most basic configuration is illustrated in FIG. 11 by dashed line 506.

[0077] Computing device 500 may have additional features / functionality. For example, computing device 500 may include additional storage (removable and / or non-removable) including, but not limited to, magnetic or optical disks or tape. Such additional storage is illustrated in FIG. 11 by removable storage 508 and non-removable storage 510.

[0078] Computing device 500 typically includes a variety of computer readable media. Computer readable media can be any available media that can be accessed by the device 500 and includes both volatile and non-volatile media, removable and non-removable

[0079] Computer storage media include volatile and non-volatile, and removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Memory 504, removable storage 508, and non-removable storage 510 are all examples of computer storage media. Computer storage media include, but are not limited to, RAM, ROM, electrically erasable program read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other opticalDocket No. 10046-638WO1 Client Ref: 8434 TOR storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information, and which can be accessed by computing device 500. Any such computer storage media may be part of computing device 500.

[0080] Computing device 500 may contain communication connection(s) 512 that allow the device to communicate with other devices over networks. Such networks may be public or private, combinations thereof, and may include the internet. Computing device 500 may also have input device(s) 514 such as a keyboard, mouse, pen, voice input device, touch input device, etc. Output device(s) 516 such as a display, speakers, printer, etc. may also be included.

[0081] It should be understood that the various techniques described herein may be implemented in connection with hardware components or software components or, where appropriate, with a combination of both. Illustrative types of hardware components that can be used include Field-programmable Gate Arrays (FPGAs), Application-specific Integrated Circuits (ASICs), Application- specific Standard Products (ASSPs), System-on-a-chip systems (SOCs), Complex Programmable Logic Devices (CPLDs), etc. The methods and apparatus of the presently disclosed subject matter, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, such as floppy diskettes, CD-ROMs, hard drives, or any other machine- readable storage medium where, when the program code is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the presently disclosed subject matter.

[0082] Although exemplary implementations may refer to utilizing aspects of the presently disclosed subject matter in the context of one or more stand-alone computer systems, the subject matter is not so limited, but rather may be implemented in connection with any computing environment, such as a network or distributedDocket No. 10046-638WO1 Client Ref: 8434 TOR computing environment. Still further, aspects of the presently disclosed subject matter may be implemented in or across a plurality of processing chips or devices, and storage may similarly be effected across a plurality of devices. Such devices might include personal computers, network servers, and handheld devices, for example.

[0083] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

[0084] In the specification and / or figures, typical embodiments have been disclosed. The present disclosure is not limited to such exemplary embodiments. Those skilled in the art will also appreciate that various adaptations and modifications of the preferred and alternative embodiments described above can be configured without departing from the scope and spirit of the disclosure.

[0085] The use of the term “and / or” includes any and all combinations of one or more of the associated listed items. The figures are schematic representations and so are not necessarily drawn to scale. Unless otherwise noted, specific terms have been used in a generic and descriptive sense and not for purposes of limitation.

[0086] While the methods and systems have been described in connection with preferred embodiments and specific examples, it is not intended that the scope be limited to the particular embodiments set forth, as the embodiments herein are intended in all respects to be illustrative rather than restrictive.

[0087] Unless otherwise expressly stated, it is in no way intendedits steps be performed in a specific order. Accordingly, where aDocket No. 10046-638WO1 Client Ref: 8434 TOR method claim does not actually recite an order to be followed by its steps or it is not otherwise specifically stated in the claims or descriptions that the steps are to be limited to a specific order, it is no way intended that an order be inferred, in any respect. This holds for any possible non-express basis for interpretation, including: matters of logic with respect to arrangement of steps or operational flow; plain meaning derived from grammatical organization or punctuation; the number or type of embodiments described in the specification. REFERENCES

[0088] Throughout this application, various publications may be referenced. The disclosures of these publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art to which the methods and systems pertain. These publications, which are incorporated by reference, include the following: Bishop, C., Turner, A., and Read, P. 2018. “Training methods and considerations for practitioners to reduce interlimb asymmetries”. The Journal of Strength and Conditioning 40, 40–46. Casali, A., Gonzalez, G., Vallebuona, G., Perez, C., and Vargas, R. 2001. “Grindability soft sensors based on lithological composition and on-line measurements”. Minerals Engineering, 14(7), 689-700. Cesare, B., Campomenosi, N., and Shribak, M. 2022. “Polychromatic polarization: Boosting the capabilities of the good old petrographic microscope”. Geology, 50, pp. 137–141. https: / / doi.org / 10.1130 / G49303.1. Chatterjee, S., Bandopadhyay, S., and Machuca, D. 2010. “Ore grade prediction using a genetic algorithm and clustering based ensemble neural network model”. Mathematical Geosciences, 42, 309- 326.Docket No. 10046-638WO1 Client Ref: 8434 TOR Guyot, O., Monredon, T., LaRosa, D., and Broussaud, A. 2004. “VisioRock, an integrated vision technology for advanced control of comminution circuits”. Minerals Engineering, 17(11-12), 1227- 1235.al. 2017. “Mask R-CNN“. Proceedings of the 2017 IEEE International Conference on Computer Vision, pp. 2961–2969. https: / / 10.1109 / ICCV.2017.322.Saartenoja, A. et al. 2019. “Developing Multi-Sensor Drones for Geological Mapping and Mineral Exploration: Setup and First Results from the MULSEDRO Project”. GEUS Bulletin 43 (July). https: / / doi.org / 10.34194 / GEUSB-201943- 03-02. Jackisch, R., Heincke, B. H., Zimmermann, R. et al. 2022. “Drone-Based Magnetic and Multispectral Surveys to Develop a 3D Model for Mineral Wxploration at Qullissat, Disko Island, Greenland”. Solid Earth, 13, 793–825, https: / / doi.org / 10.5194 / se- 13-793-2022, 2022. Gyo-Cheol J., and Jong-Tae K. 2021. “Analysis and Comparison of Rock Spectroscopic Information Using Drone-Based Hyperspectral Sensor.” The Journal of Engineering Geology 31, no. 4: 479–92. doi:10.9720 / KSEG.2021.4.479. Lin, T.Y., Maire, M., Belongie, S. et al. 2014. “Microsoft COCO: Common objects in context, in Fleet”. Lecture Notes in Computer Science: Berlin, Springer, v. 8693, p. 740–755 https: / / doi.org / 10.1007 / 978-3-319-10602-1_48. Mattar, R. A., and Kalai R. 2018. “Development of a Wall- Sticking Drone for Non-Destructive Ultrasonic and Corrosion Testing.” Drones 2, no. 1: 8. https: / / doi.org / 10.3390 / drones2010008 Migliazza, M., Carriero, M. T., Lingua, A., et al. 2021. “Rock Mass Characterization by UAV and Close-Range Photogrammetry: A Multiscale Approach Applied along the ValloneDocket No. 10046-638WO1 Client Ref: 8434 TOR dell’Elva Road (Italy).” Geosciences 11, no. 11: 436. https: / / doi.org / 10.3390 / geosciences11110436 Nooralishahi, P., Ibarra-Castanedo, C., Deane, S., et al. 2021. “Drone-Based Non-Destructive Inspection of Industrial Sites: A Review and Case Studies.” Drones 5, no. 4: 106. https: / / doi.org / 10.3390 / drones5040106 Oliveira, M. J. R., Savastano, V., Matos, G., et al. 2019. “The Use of Drones and Deep Learning to Identify Igneous Rocks and Fractures.” Paper presented at the Offshore Technology Conference Brasil, Rio de Janeiro, Brazil, October 2019. doi: https: / / doi.org / 10.4043 / 29829-MS Paclík, P., Verzakov, S., and Duin, R. P. 2005. “Improving the maximum-likelihood co-occurrence classifier: a study on classification of inhomogeneous rock images”. Paper presented at the Image Analysis: 14th Scandinavian Conference, SCIA 2005, Joensuu, Finland, June 19-22, 2005. Proceedings 14 (pp. 998- 1008). Springer Berlin Heidelberg. Paskaleva, B.Hayat, M. M., Moya, M. M., et al. 2004. “Multispectral rock-type separation and classification”. In Infrared Spaceborne Remote Sensing XII (Vol. 5543, pp. 152-163). Perez, C., Casali, A., Gonzalez, G., Vallebuona, G., and Vargas, R. 1999. “Lithological composition sensor based on digital image feature extraction, genetic selection of features and neural classification”. Paper Presented at 1999 International Conference on Information Intelligence and Systems (Cat. No. PR00446) (pp. 236-241). IEEE. Sinaice, B. B., et al. 2021. “Coupling NCA dimensionality reduction with machine learning in multispectral rock classification problems.” Minerals, 2021, 11.8: 846. Soomro, T.A., Khan, M.A., Gao, et al. 2017. “Contrast normalization steps for increased sensitivity of a retinal imageDocket No. 10046-638WO1 Client Ref: 8434 TOR segmentation method”. Signal, Image and Video Processing, 11, pp. 1509–1517. https: / / doi.org / 10.1007 / s11760 -017-1114-7. Yang, P., Kamran, E., Goodfellow, S., et al. 2023. “Mine Pit Wall Geological Mapping Using UAV-Based RGB Imaging and Unsupervised Learning.” Remote Sensing 15, no. 6: 1641. https: / / doi.org / 10.3390 / rs15061641

[0089] It will be apparent to those skilled in the art that various modifications and variations can be made without departing from the scope or spirit. Other embodiments will be apparent to those skilled in the art from consideration of the specification and practice disclosed herein. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit being indicated by the following claims.

Claims

Docket No. 10046-638WO1 Client Ref: 8434 TOR CLAIMS 1. A system for obtaining and analyzing data from one or more samples, said system comprising: an unmanned aerial vehicle (a drone), wherein the ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ a computing device comprising at least one or more processors and a memory in communication with the one or more processors, wherein the computing device is in communication with the drone over a wireless network, and wherein the one or more processors execute computer readable instructions stored in the memory to: receive data from the one or more sensor systems ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ analyze the received data to make a determination about the samples.

2. The system of claim 1, wherein one or more sensor systems comprise one or more multi-physics sensor systems.

3. The system of claim 2, wherein the multi-physics sensor systems comprise one or more of optical (visible, ultraviolet and infrared ranges) systems including one or more conventional cameras and / or one or more multi-spectral ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^h as for example sonar (kHz to ultrasonic frequencies), electromagnetic systems (e.g., radar, LiDAR, and terahertz radiation), and photon radiation systems.

4. The system of claim 3, wherein at least one multi-physics sensor is a LiDAR sensor comprising a time-of-flight (TF) sensor with a microcontroller for real-time data acquisition, wherein the TF sensor emits near-infrared light pulses and records a time-of-flight and signal intensity of a reflected beam from a surface of at least one of the samples, wherein the signal intensity of theDocket No. 10046-638WO1 Client Ref: 8434 TOR reflected beam is normalized and used to distinguish materials based on a strength of the reflected beam.

5. The system of any one of claims 1-4, wherein the one or more samples comprise rock samples and wherein the computing device processes and fuses all the received data to estimate compositional (fluids and solids) properties of the rock samples (types of fluids and types of solids, including chemical elements, chemical compounds, minerals, and their relative concentrations(fluid storage (e.g., porosity)) and flow (e.g., permeability) properties of the rock samples, and elastic (e.g., P- and S-wave velocity, anisotropy, density) and mechanical (e.g., Young’s modulus, Poisson’s ratio, Lame’s parameter, bulk modulus, etc.) properties of the rock samples.

6. The system of any one of claims 1-5, wherein the one or more sensor systems comprise an optical camera mounted on the drone and images acquired by the optical camera are transmitted to the computing device where they are subjected to image processing techniques, wherein the image processing techniques comprise computer-executable instructions executed by the one or more processors that combines a watershed segmentation with a neighboring search to enable accurate segmentation of images of rock core samples from a background, and wherein the computer- executable instructions executed by the one or more processors identify the boundaries of the rock core samples by treating the image as a topographic surface and "flooding" it from the minimum values, effectively removing the background and isolating the rock core sample features.

7. The system of claim 6, wherein the image processing techniques comprise computer-executable instructions executed by the one or more processors to execute a machine learning algorithm, wherein the machine learning algorithm segments the images acquired by the optical camera, whereinDocket No. 10046-638WO1 Client Ref: 8434 TOR the machine learning algorithm comprises a Convolutional U- Net architecture, wherein the a Convolutional U-Net architecture performs pixel classification, wherein it predicts a class of each pixel in the segmented images, and wherein the U-Net architecture is trained is based on a dataset comprising images of rock samples with known classifications, leveraging prior knowledge to enhance the accuracy of the predictions, wherein segmented images are fed into the U-Net architecture , allowing it to learn distinct features and patterns associated with different rock types.

8. The system of any one of claims 1-7, wherein the one or more sensor systems comprise a multispectral camera system, wherein a multispectral camera mounted in the drone acquires multispectral images (one for each of a plurality of wavelengths), wherein spectral features from the multispectral data for each rock sample using the computing device, wherein the spectral features comprise spectral reflectance values across specific spectral bands, and wherein the computing device identifies distinctive reflectance properties of different rock elements, wherein the computing device converts the multispectral images into a reflectance curve, wherein an intensity of the multispectral images is related to reflective properties of the rocks (samples), and wherein different rocks have different optical response at specific wavelengths which is used to classify the rock samples, wherein data from the multispectral images is stored as a matrix of a digital number, wherein each pixel has a location (i, j) and a reflectance value, The i and j for a pixel are the same in all bands, but the pixel value may vary from band to band, wherein the acquired multispectral images comprise seven matrices, one for each frequency, where each cell represents the reflectance of the rock at a specificDocket No. 10046-638WO1 Client Ref: 8434 TOR wavelength, and wherein image processing is performed on the multispectral images by the computing device in order to mitigate effects of external noise (such as ambient light).

9. The system of any one of claims 1-8, wherein the one or more sensor systems comprise an acoustic system, wherein the acoustic system comprises a waveform generator, a lithium battery, and an Murata air transducer (MA40S4R) having a frequency of 40 kHz, wherein the acoustic system records ultrasonic reflections from the rock samples, wherein the recorded acoustic reflections are transmitted to the computing device and analyzed by the computing device to compute an acoustic impedance of the rock sample and calculate P-wave velocity, Vp, of the rock sample.

10. A method of obtaining and analyzing data from one or more samples using an unmanned aerial vehicle (drone), comprising: receiving, by a computing device, data from the one or more sensor systems as the drone flies over the one or more samples, wherein the drone is equipped with the one or more sensor systems; and analyzing, by the computing device, the received data to make a determination about the samples, wherein one or more sensor systems comprise one or more multi-physics sensor systems, wherein the multi-physics sensor systems comprise one or more of optical (visible, ultraviolet and infrared ranges) systems including one or more conventional cameras and / or one or more multi-spectral ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ to ultrasonic frequencies), electromagnetic systems (e.g., radar, LiDAR, and terahertz radiation), and photon radiation systems, wherein at least one multi-physics sensor is a LiDAR sensor comprising a time-of-flight (TF) sensor with a microcontroller for real-time data acquisition, wherein the TF sensor emits near-infraredDocket No. 10046-638WO1 Client Ref: 8434 TOR light pulses and records a time-of-flight and signal intensity of a reflected beam from a surface of at least one of the samples, wherein the signal intensity of the reflected beam is normalized and used to distinguish materials based on a strength of the reflected beam, and wherein the multi-physics sensor systems are interchangeable on the drone.

11. The method of claim 10, wherein the control signals comprise an intelligent feedback loop to modify a trajectory and / or a speed of the drone in order to acquire more detailed measurements in places where the one or more samples exhibit higher spatial variability in their properties.

12. The method of any one of claims 10-11, wherein the one or more samples comprise rock samples, and wherein the computing device processes and fuses all the received data to estimate compositional (fluids and solids) properties of the rock samples (types of fluids and types of solids, including chemical elements, chemical compounds, minerals, and their relative concentrations(fluid storage (e.g., porosity)) and flow (e.g., permeability) properties of the rock samples, and elastic (e.g., P- and S-wave velocity, anisotropy, density) and mechanical (e.g., Young’s modulus, Poisson’s ratio, Lame’s parameter, bulk modulus, etc.) properties of the rock samples.

13. The method of any one of claims 10-12, wherein the one or more sensor systems comprise an optical camera mounted on the drone and images acquired by the optical camera are transmitted to the computing device where they are subjected to image processing techniques, wherein the image processing techniques comprise computer-executable instructions executed by one or more processors of the computing device that combines a watershed segmentation with a neighboring search to enable accurate segmentationDocket No. 10046-638WO1 Client Ref: 8434 TOR of images of rock core samples from a background, wherein the computer-executable instructions executed by the one or more processors identify the boundaries of the rock core samples by treating the image as a topographic surface and "flooding" it from the minimum values, effectively removing the background and isolating the rock core sample features, wherein the image processing techniques comprise computer- executable instructions executed by one or more processors of the computing device to execute a machine learning algorithm, wherein the machine learning algorithm segments the images acquired by the optical camera.

14. The method of claim 13, wherein the machine learning algorithm comprises a Convolutional U-Net architecture, wherein the a Convolutional U-Net architecture performs pixel classification, wherein it predicts a class of each pixel in the segmented images, and wherein the U-Net architecture is trained is based on a dataset comprising images of rock samples with known classifications, leveraging prior knowledge to enhance the accuracy of the predictions, wherein segmented images are fed into the U- Net architecture , allowing it to learn distinct features and patterns associated with different rock types.

15. The method of any one of claims 10-14, wherein the one or more sensor systems comprise a multispectral camera system, wherein a multispectral camera mounted in the drone acquires multispectral images (one for each of a plurality of wavelengths), wherein spectral features from the multispectral data for each rock sample using the computing device, wherein the spectral features comprise spectral reflectance values across specific spectral bands, and wherein the computing device identifies distinctive reflectance properties of different rock elements, wherein the computing device converts the multispectral images into a reflectance curve, wherein an intensity of theDocket No. 10046-638WO1 Client Ref: 8434 TOR multispectral images is related to reflective properties of the rocks (samples), and wherein different rocks have different optical response at specific wavelengths which is used to classify the rock samples, wherein data from the multispectral images is stored as a matrix of a digital number, wherein each pixel has a location (i, j) and a reflectance value, the i and j for a pixel are the same in all bands, but the pixel value may vary from band to band, wherein the acquired multispectral images comprise seven matrices, one for each frequency, where each cell represents the reflectance of the rock at a specific wavelength, and wherein image processing is performed on the multispectral images by the computing device in order to mitigate effects of external noise (such as ambient light).

16. The method of any one of claims 10-15, wherein the one or more sensor systems comprise an acoustic system, wherein the acoustic system comprises a waveform generator, a lithium battery, and an air transducer having a frequency of 40 kHz., wherein the acoustic system records ultrasonic reflections from the rock samples, wherein the recorded acoustic reflections are transmitted to the computing device and analyzed by the computing device to compute an acoustic impedance of the rock sample and calculate P-wave velocity, Vp, of the rock sample.#

Images