Unmanned aerial vehicle with spectrometric detection module for wall characterization

Abstract

An unmanned aerial vehicle (1) for wall characterization incorporates a NaI or CeBr3 scintillation detector (6), a vehicle positioning system (9), and distance sensors (4) with respect to the wall to maintain the vehicle at a horizontal distance perpendicular to the wall of between 3 and 20 cm, an onboard computer (8), and a multichannel analyzer (7) for the direct, real-time acquisition of detector data and its processing for automatic decision-making regarding the characterization of said walls. The unmanned aerial vehicle (1) further comprises a front stabilization bar (3) that houses the scintillation detector (6) and the distance sensors (4).

Description

Unmanned aerial vehicle with spectrometric detection module for wall characterization. Technical sector The invention pertains to the field of nuclear facilities. Specifically, it relates to an autonomous unmanned aerial vehicle with a spectrometric detection module for radiological characterization and declassification campaigns of interior walls of any nuclear facility, particularly during its decommissioning phase. Background of the invention Throughout the decommissioning phase of a nuclear facility, much of the effort is focused on the preliminary radiological characterization of the buildings, the classification of areas, and decommissioning planning, as well as the final characterizations for declassifying materials. To carry out these characterizations, a large number of measurements covering the entire wall are required, which must be performed at different times: an initial characterization to determine the wall's condition, intermediate characterizations each time contaminated areas are removed, and a final characterization to demonstrate that the wall meets the declassification levels specified in the regulations. Currently, the characterization and / or declassification of walls is carried out using handheld contamination meters, operated by a single person. The work is entirely manual and involves the continuous assembly and disassembly of scaffolding. Furthermore, measurements must be taken almost in contact with the surface, and the results may not be reliable, as they do not distinguish between natural and artificial isotopes. Based on the dose rate data reported by the contamination meter, and knowing the isotopic value of the location, the activity of each radioisotope present in the wall is determined. It is worth highlighting the significant human factor in performing the measurements (e.g., the uncertainty associated with measurement time), the manual recording of the data, and its analysis. The high uncertainty, along with the low reliability mentioned above, can lead to the non-declassification of declassifiable material, with the consequent increase in waste that must be stored, managed and controlled for hundreds of years. Commercial systems exist that use drone-mounted detectors to perform characterization tasks in environments with high radiation levels, such as those associated with accidents like the Fukushima nuclear disaster. However, no system exists specifically designed for declassification tasks. Documents KR 102 019891 B1 (SUNKWANG ATOMIC ENERGY SAFETY) and XP034258321 [DOI: 10.1109 / IROS47612.2022.9981415, "Unmanned aircraft system-Based Radiological Mapping of Buildings", 2022 IEEE / RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), IEEE, October 23, 2022, pages 1794-1801, by Tomás LAZNA et al.], disclose unmanned aerial vehicles with a spectrometric detection module for the characterization of walls, according to the preamble of appended independent claim 1. Therefore, it is desirable to have an automated system for performing the measurements and enabling the declassification of the walls. Summary of the invention The object of the present invention is to provide an unmanned aerial vehicle according to claim 1, which enables the automatic radiological characterization, by gamma spectrometry, of walls and interior surfaces of nuclear or radioactive facilities with very low levels of contamination, the ultimate goal being the declassification of these walls. Therefore, the invention proposes an unmanned aerial vehicle equipped with an indoor positioning system for areas without GNSS signal, based, among other systems, on LIDAR laser technology. The vehicle is also provided with a front stabilization bar housing a detection module and wall distance sensors, independent of the positioning system. The detection module is equipped with a scintillation detector and a multichannel analyzer.It should be noted that all the electronic components for position estimation are located within the vehicle; that is, no external positioning devices are required. The detection module comprises an interchangeable 2 x 2" Nal or 1.5 x 1.5" CeBr 3 scintillator, combining lightness, sensitivity, and easy integration of its electronics into the vehicle. Thanks to the combination of the detector, the laser positioning system, and the wall distance sensors—features not included in the latest generation of drones—it is possible to perform final wall declassification, something that was not previously possible with existing drones. The use of the invention also implies a reduction in both radiological risks and risks associated with other aspects of conventional safety. Thanks to the invention's vehicle, the human factor is avoided both in taking measurements and in their traceability, since the proposed system allows for the orderly and real-time recording of radiological information. The invention also provides a declassification system comprising the autonomous vehicle and a field computer that communicates with the on-board computer located in the chassis. The on-board computer in the chassis and the field computer communicate with each other via a radio antenna. The vehicle is designed to maintain the detector at a distance of 3 to 20 cm from the surface being measured throughout the inspection, while the vehicle moves at speeds of approximately 2 to 15 cm / s. The distance to the wall is calculated and maintained solely through vehicle stabilization while in motion; there are no mechanical means to maintain this distance. Monte Carlo simulations and experimental tests have been carried out to ensure that, with these inspection parameters, the selected detectors can reach the detection limits required by the regulations for the declassification of walls. The result is a file that correlates the position on the wall with the gamma spectrum, which is then analyzed to transform the point cloud into a radiation map, allowing hot spots to be located and the results to be compared with the limits allowed in the regulations. Brief description of the figures In order to facilitate a better understanding of the characteristics of the invention, the following figures are attached as an integral part thereof, the nature of which is illustrative and not limiting: Figure 1 shows the essential elements of the invention. Figure 2 is a bottom view of the vehicle of the invention. Figure 3 is a detail of the front stabilizer bar. Figure 4 shows the sensors on both sides of the stabilizer bar. Detailed description With reference to Figure 1, the proposed vehicle 1 aims to characterize and declassify walls using a radiation detection module housed in a front stabilization bar 3, a vehicle positioning system 9, and distance sensors 4 on the bar 3, in order to automatically perform data collection, measurement recording, and results analysis. A detection module 5, comprising a scintillation detector 6, was selected due to the high efficiency of this type of detector for gamma emissions, as the objective is to characterize areas with low or very low pollution levels. The module also includes a multichannel analyzer (MCA) 7, which houses the electronics responsible for analyzing the pulses recorded on the detector's scintillation crystal. The selected detectors are scintillators with crystals ranging in size from 1.5 x 1.5 inches to 3 x 3 inches. In particular, the 2 x 2 inch Nal and 1.5 x 1.5 inch CeBr 3 scintillators combine lightness, sensitivity, and easy integration of their electronics into the vehicle. Despite the small size of the glass, the selected detectors can detect at least 50% of the declassification level indicated in Table 1. It is found that, with inspection speeds on the order of cm / s and a distance to the wall of 3 to 20 cm, the detection capacity complies with the regulations. The inspection parameters (inspection speed, distance to the wall, measurement time) must be defined for each campaign, as they depend on the background in the area. In any case, the inspection will be carried out at speeds of 2 to 15 cm / s and distances to the surface of between 3 and 20 cm. Since the system records its position in (X, Y, Z) at all times, the results of the recorded activity are analyzed taking into account the distance between the detector and the wall.The following table shows the declassification levels for the demolition and reuse of buildings according to Instruction IS-13 of March 21, 2007 of the Nuclear Safety Council, on radiological criteria for the release of sites of nuclear facilities. From the above criteria, it is concluded that it is imperative that the vehicle be able to perform inspections at a distance of 20 cm or less from the wall and at inspection speeds on the order of cm / s, while supporting the weight associated with the detection module (approximately 1.5 kg), guaranteeing uninterrupted operation and high reliability from a safety standpoint. Furthermore, the system must be capable of inspecting walls up to 20 meters high from the ground. The invention comprises a main chassis 2 housing the propellers, motors (preferably four to six), and on-board computer 8 (Figures 1 and 2), as well as a vehicle positioning system 9. A front stabilizer bar 3 (Figures 3 and 4) houses the detection module 5, which includes a Nal or CeBr 36 detector, and a multichannel analyzer 7. The Nal detector is used for dynamic inspections due to its high efficiency, and the CeBr3 for static inspections due to its high resolution. Both detectors are interchangeable. The distance sensors 4, located on either side of the bar, are independent of the vehicle positioning system 9. During flight, for altitude positioning, the vehicle is equipped with a laser positioning system (9) for 3D ground scanning. To avoid potential obstacles on the wall and maintain the desired distance from it, the system has distance sensors (4) to scan the wall. This allows it to detect any protrusions during flight, move far enough away to avoid them, and then return to the set distance once the obstacle is clear. Chassis 2 is made of carbon fiber and laser-cut for precision. The chassis is constructed primarily of lightweight, impact-resistant braided carbon fiber, making it ideal for the aircraft's chassis and other components subjected to significant stress. It consists of an H-shaped plate, 2 to 4 mm thick, with dimensions between 100 x 300 mm and 300 x 700 mm, and two round carbon fiber tubes, 300 to 900 mm long, 25 mm in outer diameter, and 1-2 mm thick, which house the four motors at their ends (e.g., brushless motors between 100 and 500 kv). This plate is reinforced with square carbon fiber tubes on the back (depending on the flight direction), measuring 10 x 10 mm and 1-3 mm thick. The diagonal dimension (without propellers) is between 400 and 1100 mm.The selected folding propellers range from 375 to 700 mm. The main power supply is provided by a battery or cable. If wired, performance increases, as the vehicle can take measurements for longer periods without requiring a ground connection to change the battery. Since the power cable length can be considerable (estimated at around 30 meters), a conductor with the following characteristics has been identified: Multi-conductor core to provide greater cable flexibility. 4 independent conductors, for redundancy in case of breakage (a significant level of fatigue is expected). 1 mm² cross-section (US "AWG" gauge between 12 and 20). Silicone case or similar to increase flexibility. Safety connectors both on the vehicle and at the connection to the power supply. Regarding the power supply, a redundant power supply is used to prevent loss of vehicle lift in the event of a power system failure. The power supply will provide the vehicle with a voltage between 54 V DC and 60 V DC. Overcurrent protection mechanisms are in place to prevent further damage in case of a robot malfunction. Optionally, the vehicle is equipped with a 10,000–20,000 mAh battery that will provide approximately 30 minutes of autonomy (depending on the final weight of the detection module, which is between 1 kg and 5 kg, excluding the battery and the detection module). Communication between the on-board computer 8 and the external processing means (not shown in the figure) is carried out at radio frequencies within the ISM bands with a radio range between 433 MHz and 5.8 GHz, allowing the equipment to be operated and the maneuver monitored even in large installations. The front stabilization bar 3 (Figure 3), in addition to housing the detection module 5, protects the propellers from potential collisions with the surface being characterized. The front stabilization bar 3 has unique distance sensors 4 (Figure 4), independent of those used for vehicle navigation, which maintain the vehicle at the most constant distance possible from the surface being scanned. These distance sensors 4 can be optical time-of-flight distance detectors and provide distance measurements at a much higher rate than the vehicle's other sensors, thus improving vehicle stability when positioned along the axis normal to the scanned surface.It should be noted that the distance between the surface being measured and the vehicle is maintained without the use of mechanical means; that is, it is achieved solely through vehicle stability control during flight. This provides the system with an additional degree of versatility, as it is less dependent on irregularities in the surfaces being measured, which are common in work environments. Distance control without mechanical elements (a "stop" type of control) is achieved through parameterization of specific control loops based on: i) the weight of the detector, and ii) the presence or absence of a battery. The specific control loop that regulates the distance between the detector and the surface uses as an estimate the information gathered by laser scans (LiDAR) and by the Time-of-Flight (ToF) sensors installed on the front stabilizer bar.While LiDAR data feeds the control loop at a frequency between 5 Hz and 20 Hz, the distance sensors installed on the front bar do so at a frequency between 150 Hz and 300 Hz, providing additional information about the distance between LiDAR scans. Vehicle control is achieved using an Extended Kalman Filter (EKF) dedicated to tilt (attitude) control, which is fed by one or more thermally calibrated inertial measurement units (IMUs). This filter operates with a data input frequency between 1000 Hz and 2000 Hz from the IMU(s). As a reference for the control system, tilt instructions are provided, calculated based on the distance error to the surface being measured. This error is derived from the velocity, which in turn is derived from the acceleration, the latter being the determining factor for the reference tilt provided to the control system. The detection module 5, consisting of a scintillation detector 6 and a multichannel analyzer 7, is located on the front stabilizer bar 3 at the front of the vehicle. By means of a mechanical adjustment mechanism (when the vehicle is stationary), the relative distance between the sensor and the crossbar is adjusted within an approximate range of 0 to 10 cm, in order to optimize the detector's position in terms of distance to the target surface. The vehicle optionally has a side protection system consisting of two pieces made by 3D printing with reinforced material, mainly Kevlar and carbon fiber, each located on one side of the main chassis. The vehicle's electronic core is based on a PCB (printed circuit board). This design can power up to six propulsion motors, image capture sensors, high-power lighting systems, an onboard computer, and other peripherals. It features overcurrent protection, battery input voltage measurement circuits, overall power consumption control, communication port distribution, and auxiliary connections for integrating optional components such as gas sensors or thermal imaging. The indoor positioning system uses Time-of-Flight (ToF) sensors to estimate the distance to environmental boundaries (walls, ceiling, and floor, as well as intermediate elements such as columns). The position estimation system uses two different reference points to estimate its position and avoid potential slippage or loss of reference that could significantly affect vehicle behavior. While the global coordinates are calculated using an ICP-based laser scan comparison system with cell occupancy, the relative distance between the detector and the wall uses a separate reference point. This is because, while ICP works with occupancy cells between 5 and 20 cm, the distance between the detector and the wall must be calculated in the millimeter range. The absolute positioning system also uses data from the inertial measurement unit (IMU) to refine its position, as using 20 cm occupancy cells is not feasible for obtaining reliable position data.On the other hand, the vehicle's forward speed is calculated using differential comparison of laser scans (ICP) and does not depend, under any circumstances, on the calculation of the overall position based on cell occupancy. Position estimation allows for the programming of autonomous navigation patterns. During the scan, the vehicle records its position at any given time, as well as the distance between the detection module, located on the front stabilizer bar, and the surface of the wall being measured. The system can inspect wall areas up to 20 meters high. The invention has two modes of operation for this purpose: • Dynamic mode: Performing scans, sweeping in an "S" shape across the entire wall to be inspected. • Static mode: Performing static measurements, precisely positioning itself at pre-selected points. It is essential to have these two modes of operation, as this is what the regulations require in order to declassify the surfaces and walls present in nuclear facilities. The system acquires data, performs calculations, presents the results in graphs or maps, has decision-making criteria, and produces reports that can be used to characterize / declassify walls. During the inspection, the driver does not interact with the vehicle. The points to be inspected and the route to be taken are entered into the computer, and the system performs the inspection automatically. Data transmission between the field computer and the onboard computer is carried out in real time via radio, as described above. The onboard computer controls data acquisition by the detection module using different algorithms to start and stop detection, differentiating between dynamic and static measurements. Software integrated into the vehicle's onboard computer has been developed to manage the detection module's control functions and configure the inspection parameters.This application allows the execution of various Python code snippets using functions available in the Multi-Channel Analyzer (MCA) SDK (Software Development Kit). These snippets are responsible for applying the appropriate voltage and gain to each detector, starting and stopping data acquisition in both operating modes, and recording the data acquired during inspections. Furthermore, this application includes a configuration window where the user can select the detector type for each task, choose the operating mode (dynamic or static), configure measurement times, the detector-to-wall distance, the scan interval, and the wall's lateral limits.If a dynamic scan is performed, after entering the boundaries of the area to be inspected and the intrinsic flight parameters (sweep speed, vertical pitch, and distance to the wall) into the field computer, the vehicle will traverse the inspection area in an S-shape, recording a spectrum every second. At the end of the inspection, the complete file containing all the spectra will be sent to the field computer, and these spectra will be analyzed to provide the coordinates of the center of the 1 × 1 m grid squares considered points of interest. This analysis can be performed using a Python script developed for this purpose or using the GENIE 2000 software. When taking a static measurement, the vehicle will sequentially move to each of the selected points and take measurements for a preselected time, recording one spectrum per second.As with dynamic scanning, the analysis of the spectra and the determination of activity can be performed using a Python script developed for this purpose. Data processing is carried out automatically on the field computer, generating a customized report where the results information appears in the chosen format: Color map, graphs, reports, etc. The designed system can be used in two types of radiological characterization campaigns: • Preliminary characterization of interior walls, in order to serve as a first radiograph of the wall, as well as input data for the classification of release units, for their subsequent declassification • Characterization for the declassification of the wall, in order to guarantee with sufficient reliability that the wall has radiation levels below the declassification limits and that it can be demolished as conventional material.

Claims

1. An unmanned aerial vehicle (1) with a spectrometric detection module (5) for wall characterization, comprising: - a chassis (2) equipped with propellers, motors, and a vehicle positioning system (9), as well as an on-board computer (8); - a detection module (5) comprising a Nal or CeBr 3 scintillation detector (6) for the spectrometric detection of radioisotopes and a multichannel analyzer (7) for analyzing and transmitting the data obtained by the detector to the on-board computer (8); and - distance sensors (4) for measuring the distance to the wall; the vehicle (1) being characterized in that it further comprises a front stabilization bar (3) in which the detection module (5) and the distance sensors (4) are housed, for maintaining the vehicle at a horizontal distance in a line perpendicular to the wall of between 3 and 20 cm. 2.Unmanned aerial vehicle according to claim 1, wherein the distance sensors (4) are optical time-of-flight sensors.

3. Spectrometric system for wall declassification comprising an unmanned vehicle (1) according to any of the preceding claims provided with a radio antenna and a field computer that communicates with the on-board computer via said antenna.

Images