Indoor UAV Flight Control and Navigation Performance Evaluation Method Based on Autonomous Positioning GNSS

Abstract

This invention proposes a method for evaluating the flight control and navigation performance of indoor unmanned aerial vehicles (UAVs) based on autonomous positioning GNSS, comprising the following steps: Step 1: Installation of an external UAV real-time status perception device; Step 2: Installation of indoor and external positioning devices; Step 3: Activation of a GNSS / RTK signal generator based on an autonomous positioning system; Step 4: UAV flight testing; Step 5: UAV flight data processing and performance evaluation. This invention ensures that UAVs maintain efficient and safe flight performance in various operating environments and potentially non-standard conditions. Through these comprehensive navigation performance evaluations, UAV operators and manufacturers can obtain detailed data on equipment performance, which helps optimize design and improve the reliability of flight strategies. This not only improves the practical application efficiency of UAVs but also strengthens their application capabilities in complex environments, promoting the widespread application of UAV technology in civilian and commercial fields.

Description

Technical Field

[0001] This invention belongs to the field of flight control and navigation performance evaluation of rotary-wing unmanned aerial vehicles (UAVs), including indoor autonomous positioning, GNSS / RTK signal generation, real-time state estimation, multi-sensor fusion, computer vision, etc. In particular, it can realize real-time attitude detection and flight control and navigation performance evaluation of closed-source UAVs in indoor environments. Specifically, it involves an indoor satellite positioning test method based on signal simulation and a UAV flight control and navigation performance evaluation system. Background Technology

[0002] Multirotor aircraft are aircraft that rely on multiple fixed-pitch propellers to provide control force and torque. They can easily complete basic flight maneuvers such as vertical takeoff and landing, hovering, and turning. They have wide applications in fields such as aerial photography, exploration and detection, agricultural plant protection, and rescue and search, and also have a certain market and development prospects in emerging fields such as short-distance freight logistics and short-distance manned transportation.

[0003] Compared to fixed-wing aircraft, rotorcraft's actuators consist of motors and propellers. A malfunction in a rotorcraft can directly affect its flight status, causing it to fail to fly according to mission instructions or even crash, potentially resulting in economic losses and even personal injury in outdoor scenarios. Therefore, the safety and reliability of drone flight control and navigation performance have become crucial performance indicators for drones used in large venues. To verify this performance, regular flight control and navigation performance evaluations are typically required. However, most drone applications on the market use closed-source drone equipment from established companies like DJI and Autel Robotics, making it difficult to directly obtain drone status information, implement GPS navigation flight indoors, and conduct regular evaluations of drone flight control and navigation performance.

[0004] Therefore, this invention primarily addresses the current lack of flight control and navigation performance evaluation capabilities for unmanned aerial vehicles (UAVs). It proposes an indoor UAV flight control and navigation performance evaluation method based on GNSS / RTK signal generation from an autonomous positioning system. This method can be deployed on UAVs with a high degree of closed-source operation, reducing the difficulty of retrieving monitoring data. An indoor UAV positioning system based on a satellite positioning device is deployed indoors, enabling satellite navigation and allowing for the evaluation and verification of UAV flight control and navigation performance in an indoor environment. Furthermore, a highly versatile flight control and navigation performance evaluation system for rotary-wing UAVs is constructed, using open-source systems to unify related mathematical definitions, descriptions, partitions, and interfaces, thereby improving the system's versatility. Summary of the Invention

[0005] Because drone components are susceptible to invisible damage during flight due to environmental factors such as temperature and humidity, malfunctions directly affect the drone's flight status, causing it to fail to fly according to mission instructions or even crash, potentially resulting in economic losses and even personal injury in outdoor scenarios. Therefore, the safety and reliability of drone flight control and navigation performance have become crucial performance indicators for drones used in large venues. To verify this performance, regular flight control and navigation performance evaluations are typically required. However, commercially available drone applications often use closed-source drone equipment from established companies like DJI and Autel Robotics, making it difficult to directly obtain drone status information, implement GPS navigation flight indoors, or evaluate drone flight control and navigation performance in an indoor laboratory environment.

[0006] To evaluate the flight control and navigation performance of outdoor unmanned aerial vehicles (UAVs) using GNSS / RTK signals in an indoor laboratory environment, this invention constructs an indoor UAV flight control and navigation performance evaluation method based on GNSS / RTK signal generation from an autonomous positioning system. The overall system composition is as follows: Figure 1 As shown. It includes the following steps:

[0007] Step 1: Installation of external drone real-time status sensing device

[0008] Attach the external drone real-time status sensing device according to the mounting holes on the DJI Mavic 3E drone, and start the sensing device to record the current drone status data.

[0009] The external drone real-time status perception device is primarily designed for aircraft from DJI, Autel, and other manufacturers with closed-source modules. It acquires real-time flight status information by deploying multi-dimensional information sensors on the aircraft. The device includes three different types of accelerometers and gyroscopes, and processes and transmits real-time data via an STM32 main processor.

[0010] It mainly includes the following three types of status information:

[0011] 1. UAV attitude information: Real-time angular velocity and acceleration information of the UAV are obtained by using an inertial sensor. The estimated values ​​of the angle, angular velocity and angular acceleration of the UAV in the pitch, roll and yaw directions are obtained by using a Kalman filter.

[0012] 2. UAV position information: The UAV position information is obtained by using a position acquisition device (satellite positioning system) and fused with information from the airborne inertial sensor and altitude sensor. An extended Kalman filter is used to fuse the multi-dimensional sensor information to obtain more accurate and smooth 3D position estimation information.

[0013] 3. UAV vibration information: By using inertial sensor information to generate power spectrum density information and real-time vibration intensity information of the UAV, the vibration state of the UAV during flight can be assessed.

[0014] After denoising the data, the information is sent to the ground-based online UAV performance evaluation system via a remote communication module to conduct in-depth UAV performance evaluation.

[0015] Step 2: Installation of indoor and outdoor positioning devices

[0016] A lidar is used as an indoor / outdoor positioning device to obtain the drone's real-time indoor location, enabling the GNSS / RTK signal generator to produce accurate indoor GNSS / RTK positioning signals. The lidar is connected to the DJI Mavic 3E drone via a proprietary connector, the lidar positioning system is activated to obtain the drone's indoor coordinates, and these coordinates are transmitted wirelessly to the GNSS / RTK signal generator.

[0017] Indoor high-precision positioning systems based on lidar are a cutting-edge technology primarily used for accurately tracking and locating objects or people in indoor environments. These systems have wide applications in various fields, including autonomous navigation, robotics, security monitoring, and intelligent transportation systems. LiDAR systems typically use one or more laser emitters and receivers to emit laser beams towards a target from different angles and capture the reflected laser light to obtain the target's precise location information.

[0018] The indoor high-precision positioning system based on lidar employs SLAM (Simultaneous Localization and Mapping) technology, an advanced technology capable of simultaneously performing self-localization and environmental mapping in unknown environments. It has wide applications in various fields, including autonomous navigation, robotics, security monitoring, and intelligent transportation systems. SLAM technology enables real-time updates to the drone's position through continuous lidar observations, while simultaneously constructing a detailed 3D map of the environment.

[0019] SLAM technology is mainly based on two core mathematical models: the state estimation equation and the observation equation.

[0020] Motion Model (State Estimation Equation)

[0021] x t =f(x) t-1 ,u t ,w t )

[0022] Where, x tThe drone's state at time t (typically including position and orientation). t-1 This represents the state at the previous moment. u represents the control input (such as changes in speed and direction). w t Process noise represents the uncertainty introduced by model imperfections or external influences.

[0023] The state estimation equation describes how the state of the UAV evolves from one time step to the next, depending on its control commands and inherent uncertainties.

[0024] Measurement Model

[0025] z t =h(x t ,n t )

[0026] Among them, z t x represents the observation data obtained at time t using sensors such as lidar. t This represents the robot's current estimated state. t The noise level represents the measurement error.

[0027] The observation equation links the robot's predicted state with the actual data observed through sensors, helping to correct biases in state estimation.

[0028] Based on this principle, we can calculate the precise location of the drone indoors using a lidar indoor positioning system, and then realize it through a GPS baseband signal data stream generation system and a GPS signal transmission system.

[0029] Step 3: Start the GNSS / RTK signal generator based on the autonomous positioning system

[0030] After obtaining the drone's precise location indoors, we can activate the GNSS / RTK signal generator based on the autonomous positioning system to generate accurate GNSS / RTK positioning signals to enable GPS navigation for the drone. This allows the drone to rely on its own satellite positioning system to achieve position control and flight mode in indoor scenarios.

[0031] GPS analog signal data stream generation system based on GPS broadcast ephemeris files:

[0032] Step 3.1: Satellite Data Acquisition

[0033] Before generating a simulated GPS signal, the GPS ephemeris file for the target time must first be obtained. The ephemeris file primarily contains precise satellite orbit data, which is crucial for the Global Positioning System (GPS). This data includes: 1. Satellite orbital parameters: These parameters describe the satellite's exact position and motion in space, including its right ascension of the ascending node, orbital inclination, perigee distance, eccentricity, mean perigee angle, and orbital period. 2. Clock data: The deviation between the satellite's atomic clock and ground time, as well as clock error correction parameters. 3. Health status: Indicates whether the satellite is suitable for navigation and positioning services. 4. Atmospheric delay parameters: These parameters help the receiver correct for ionospheric and tropospheric delays during signal propagation. This data is essential for generating the GPS baseband signal data stream. These parameters are then used to calculate the satellite's position at any given time using Kepler's equations.

[0034] E = M + e sin E

[0035] Where E is the eccentric anomaly; M is the mean anomaly, calculated as: M = n(t - t0); n is the mean motion, representing the average angular velocity of the satellite in one orbit around the Earth. t is the observation time. t0 is the reference time, typically the time it takes for the satellite to pass its perigee. e is the orbital eccentricity, describing the flattening of the satellite's orbit.

[0036] Based on the anomalous angle, the satellite's position in the orbital plane can be further calculated:

[0037] x′=a(cos Ee)

[0038]

[0039] Where x′ and y′ are the coordinates of the satellite in the orbital plane. 'a' is the semi-major axis, representing the size of the satellite's orbit.

[0040] These two formulas are used to calculate the satellite's position on its orbital plane, where x′ is along the direction from the Earth's center to the perigee, and y′ is perpendicular to x′. In reality, the satellite's orbit is inclined relative to the Earth's equatorial plane, so it's also necessary to transform the orbital plane position to a geocentric coordinate system:

[0041] x=x′cos(Ω)-y′cos(i)sin(Ω)

[0042] y=x′sin(Ω)+y′cos(i)cos(Ω)

[0043] z = y′sin (i)

[0044] Where Ω is the right ascension of the ascending node, and i is the orbital inclination.

[0045] Step 3.2: Signal Generation

[0046] After acquiring satellite data, the system uses indoor positioning information and base point coordinates obtained through motion capture to calculate the line-of-sight direction and pseudorange from the satellite to the set receiver position. Then, it calculates and generates the corresponding GPS signal from the satellite data, including generating pseudo-random noise code (PRN) and encoding navigation messages.

[0047] Each GPS satellite is assigned a unique PRN code, which is a repeating binary sequence used in receivers to help identify and synchronize with the specific satellite signal.

[0048]

[0049] Where C(t) is the output code at time T, and G i (t) represents the state of the code generator at time t, D code It is a data code, and Δt is the duration of the code element.

[0050] Navigation messages contain information such as satellite orbital parameters, clock correction parameters, and health status, which are crucial for GPS receivers to calculate their position.

[0051]

[0052] Where D(t) is the navigation data word at time t, b k It is the k-th bit of binary data, and n is the length of the data word.

[0053] Step 3.3: Signal Modulation

[0054] The generated digital baseband signal needs to be converted into an analog signal through modulation so that it can be transmitted through SDR hardware. BPSK (Binary Phase Shift Keying) is mainly used to modulate navigation data and PRN codes onto the carrier frequency to form a transmittable signal.

[0055] First, we need to integrate the baseband signal containing navigation information and the PRN code. During modulation, both the PRN code and the navigation message affect the phase change of the carrier wave. The PRN code provides a unique identifier for each satellite, while the navigation message contains crucial information about the satellite's position and time. This integration method ensures that the signal has high anti-interference capability and unique identification characteristics, enabling the receiver to accurately distinguish and lock onto a single satellite from multiple satellite signals.

[0056] φ(t)=π×(d(t)+c(t))mod2

[0057] Where d(t) is the navigation data bit and c(t) is the PRN code.

[0058] After acquiring the integrated signal, we use BPSK modulation to encode the data onto the RF carrier. This method represents the data bits (0 or 1) by changing the phase of the carrier.

[0059] s(t)=A·cos(2πf c t+φ(t))

[0060] Where s(t) is the modulated signal, A is the signal amplitude, and f is the amplitude of the signal. c φ(t) is the carrier frequency, and φ(t) is the phase of time t, which depends on the number of data bits (typically 0 or π for GPS).

[0061] During BPSK modulation, we can extract the in-phase (I) and quadrature (Q) components from the modulated signal to form the IQ signal. This step is accomplished by analyzing the modulated signal into its sine (quadrature component) and cosine (in-phase component) components, allowing the signal to carry more information and improving transmission efficiency.

[0062] I(t)=d(t)cos(2πf c t)

[0063] Q(t)=d(t)sin(2πf c t)

[0064] The generated analog signal needs to be sampled and converted into a format suitable for SDR hardware processing. The mathematical processing involved in the sampling process mainly involves discretizing the analog signal, converting the continuous analog signal into a discrete signal for digital storage and processing. The phase and amplitude of each sample point are calculated according to its corresponding time point to ensure that the integrity and accuracy of the signal are preserved.

[0065] s[n]=A·cos(2πf c nT+φ[n])

[0066] Where s[n] is the signal value at sampling time n, and T is the sampling interval.

[0067] Step 3.4: Transmit GNSS / RTK positioning signals using a software-defined radio-based GPS signal transmission system.

[0068] For the GPS signal transmission system, we selected HACKRF as the Software Defined Radio (SDR) hardware, providing an efficient and flexible signal transmission solution for GPS applications. This system can receive GPS signals generated by the GPS analog signal data stream generation system, including navigation data and pseudo-random noise (PRN) codes. Then, utilizing HACKRF's high-performance digital-to-analog conversion capabilities, these baseband signals are up-converted to the GPS operating frequency and transmitted through the RF interface.

[0069] First, the SDR device needs to receive the generated digital I / Q data from the GPS analog signal data stream generation system. This data is the baseband representation of the GPS signal. Then, the in-phase and quadrature components in the data stream are combined into a complex signal for subsequent processing and modulation.

[0070] s[n] = I[n] + jQ[n]

[0071] Where s[n] is the baseband signal in complex form, and I[n] and Q[n] are the in-phase and quadrature components at sampling point n, respectively.

[0072] After combining the signals, we can convert the complex signal into an analog signal for subsequent frequency conversion and transmission. This step is achieved through a digital-to-analog converter (DAC), which mainly converts the processed digital signal into an analog signal to prepare for transmission. The formula below... This represents the mixing process of the signal and carrier, which upscals the baseband signal to the RF band.

[0073]

[0074] Where v(t) is an analog signal, f c is the carrier frequency, and s(t) is a continuous-time complex signal obtained by interpolation and filtering from the digital baseband signal.

[0075] After completing the analog-to-digital conversion of the signal, we will need to perform frequency conversion on the analog signal, upscaling the baseband frequency to the GPS operating frequency. This is a crucial step before signal transmission, ensuring the signal is sent within the correct frequency band. The following formula describes the baseband signal passing through an analog mixer and the local oscillator (LO) frequency f. RF The mixing process:

[0076] x(t)=v(t)cos(2πf RF t)-u(t)sin(2πf RF t)

[0077] Where x(t) is the up-converted RF signal, u(t) and v(t) are the I and Q components of the baseband signal, respectively, and f RF It is the target radio frequency.

[0078] Finally, the signal power is amplified to a level sufficient for transmission through the antenna by the RF amplifier of the SDR device, ensuring that the signal can be transmitted effectively.

[0079] P out =Gain × P in

[0080] Among them, P out P is the amplified output power, Gain is the amplifier input power, and P is the output power. in It is the amplifier gain.

[0081] In this way, the system can generate customized GPS signals required by researchers, enabling testing and performance evaluation of flight control systems and GPS receiving devices. It also provides researchers with a practical platform for experimenting with and validating new algorithms or technologies.

[0082] Step 4: Drone Flight Test

[0083] After the installation and startup of the real-time status perception device, indoor and outdoor positioning device and GNSS / RTK signal generator are completed, the UAV can be started to begin indoor flight testing.

[0084] First, the accuracy of the outdoor flight test device for the UAV should be calibrated. The outdoor flight test device for the UAV should be reprojected error calibrated before each use. The measuring device should be placed in place and the measuring device should continuously read the position value within one minute to perform the measurement accuracy convergence calibration. After the calibration of the test device and the UAV itself is completed, the flight test can begin.

[0085] Step 4.1: Hovering Endurance and Positioning Accuracy Test

[0086] First, hovering endurance and positioning accuracy tests were conducted. The drone took off, hovered in a fixed-point mode, and landed after 30 minutes of flight.

[0087] Step 4.2: Hovering Test under Wind Resistance Conditions

[0088] A wind field was deployed in an indoor test scenario, and the drone was tested for hovering in the wind field. After flying for 10 minutes, it landed.

[0089] Step 4.3: Virtual Waypoint Flight Test

[0090] Before takeoff, waypoints are planned based on the size of the site, and the flight path mode is activated during flight, allowing the drone to autonomously complete the flight mission based on the waypoints.

[0091] Step 5: Drone flight data processing and performance evaluation

[0092] Step 5.1: Data Acquisition

[0093] The drone's flight data is acquired in real time during flight and saved to a log file.

[0094] The data acquisition module mainly acquires the real-time status information of the UAV, and realizes functions such as real-time data collection, transmission and storage. It is an important connection between the physical world and the UAV performance evaluation system. It needs to consider the classification and characteristics of the data, as well as various related issues of data acquisition behavior, to ensure the accuracy of the acquired data.

[0095] Step 5.2: Data Processing

[0096] After acquiring the flight data, the data is processed to facilitate the subsequent evaluation of the drone's flight performance.

[0097] To efficiently utilize data and filter out noise interference, stored data needs to be processed, including data preprocessing and feature extraction, as shown in the figure. Data preprocessing involves various aspects of processing the raw data from a data science perspective, such as statistical analysis. This mainly includes outlier and anomaly detection, sample alignment and imputation, and noise reduction. Feature extraction involves finding the most informative features in the data and discarding less informative ones. By removing unimportant features and redundant information, data compression can be achieved, effectively reducing data volume and alleviating storage pressure. This mainly includes feature extraction for time-series signal data, feature extraction based on frequency domain analysis methods, and feature extraction based on time domain analysis methods. For example... Figure 2 As shown.

[0098] Step 5.3: Unmanned Aerial Vehicle (UAV) Flight Control Performance Evaluation

[0099] After obtaining the processed drone flight data, we can use this data to evaluate the drone's flight control performance.

[0100] In evaluating the performance of UAV navigation systems, the core focus is ensuring navigation accuracy and response speed under various environments and conditions. The evaluation of UAV navigation system performance involves several key indicators, including positioning accuracy, path tracking capability, adaptability to environmental changes, and stability under different operating conditions. These evaluation indicators are divided into several parts, as follows:

[0101] 1. Basic performance evaluation:

[0102] Flight stability: Evaluating the control precision of a drone while maintaining a fixed altitude and speed is crucial for performing precise aerial missions such as aerial photography or data acquisition.

[0103] Environmental adaptability: Test the flight performance of the drone under different weather conditions, including operational stability in strong winds, low temperatures or high temperatures, and its response speed to environmental changes.

[0104] 2. Emergency response capabilities:

[0105] Emergency obstacle avoidance: Simulates sudden situations, such as unexpected obstacles, to test the drone's obstacle avoidance reaction time and effectiveness, and to verify the sensitivity of its sensors and the effectiveness of its processing algorithms.

[0106] Autonomous landing: In the event of loss of control signal or insufficient power, whether the drone can land safely, and during emergency landing, test its positioning accuracy and the smoothness of ground contact.

[0107] 3. Accuracy Test:

[0108] Position holding capability: By conducting short-term stationary hovering tests, the position stability of the drone under windless and windy conditions is evaluated, which is crucial for performing tasks that require high positioning accuracy.

[0109] Path accuracy: On a preset flight path, detect the deviation between the actual flight trajectory of the UAV and the predetermined trajectory to evaluate the accuracy and reliability of its flight control system.

[0110] 4. Endurance and stability test:

[0111] Continuous flight performance: Evaluate the performance degradation of the drone after long-term flight, including battery endurance and the durability of various components.

[0112] Control stability: During flight, how consistent is the UAV's response, especially during complex maneuvers or long-duration missions, and how stable and responsive is its control system?

[0113] These comprehensive performance evaluations ensure the efficiency and safety of drones during missions. These tests not only help manufacturers improve drone design but also provide operators with crucial information about machine performance, enabling better strategic decisions in practice. For example, endurance test results can guide operators in making reasonable expectations about battery life, while assessments of emergency response capabilities directly impact mission planning and risk management strategies. This rigorous evaluation process ensures the continuous advancement of drone technology and significantly enhances its application in both civilian and commercial sectors. For critical areas such as public safety and environmental monitoring, the optimization and reliability of this technology further expands the breadth and depth of its applications, driving innovation and standardization of drone technology globally.

[0114] Step 5.4: Unmanned Aerial Vehicle (UAV) Navigation Performance Evaluation

[0115] After obtaining the processed drone flight data, we can use this data to evaluate the drone's navigation performance.

[0116] In evaluating the performance of UAV navigation systems, the core focus is ensuring navigation accuracy and response speed under various environments and conditions. The evaluation of UAV navigation system performance involves several key indicators, including positioning accuracy, path tracking capability, adaptability to environmental changes, and stability under different operating conditions. These evaluation indicators are divided into several parts, as follows:

[0117] 1. Basic navigation capabilities:

[0118] In our basic navigation capability tests for drones, we focus on the drone's flight accuracy on fixed routes. This includes two key aspects: first, the drone's ability to maintain its position on a predetermined path, i.e., its ability to remain stable along a set flight path; and second, the accurate tracking of a preset route, which involves how the drone precisely adjusts its flight path to meet new route requirements after receiving route adjustment commands. These tests help ensure that drones can maintain efficient and accurate operation in complex routes and changing mission requirements.

[0119] 2. Altitude and speed control:

[0120] This evaluation focuses on testing the drone's stability in maintaining a specific flight altitude and speed. This includes not only the drone's ability to fly stably at a set altitude, but also its tolerance for altitude changes during flight and how quickly and accurately it adjusts its altitude to adapt to unexpected situations. The evaluation also tests the drone's acceleration and deceleration capabilities, assessing its dynamic response capabilities in handling emergencies or completing specific flight missions.

[0121] 3. Attitude stability:

[0122] Attitude stability testing of drones focuses on evaluating the machine's stability under various flight attitudes, especially its performance when faced with challenges such as sudden changes in wind direction or abrupt changes in operational commands. The test observes the drone's response speed and adjustment precision, ensuring that the drone can quickly and accurately adjust its attitude during flight to maintain stability and safety.

[0123] 4. Sensor performance evaluation:

[0124] Sensor performance evaluation primarily focused on IMUs (Inertial Measurement Units) and GNSS (Global Navigation Satellite Systems). These sensors are critical components of UAV navigation systems, therefore the testing aimed to verify their ability to provide accurate and reliable position and orientation data in real-world operations. Ensuring high sensor performance is fundamental to achieving precise navigation and stable operation of UAVs.

[0125] 5. Emergency Operation Response:

[0126] Emergency response assessments focus on a drone's ability to handle navigation errors or sudden changes in flight plans. This includes the drone's emergency stop and obstacle avoidance capabilities, testing how it quickly and safely adjusts its flight to avoid accidents when faced with potential collisions or restricted airspace. These capabilities are crucial for improving a drone's emergency response and overall safety.

[0127] The advantages and benefits of this invention are as follows: these tests ensure that the drone maintains efficient and safe flight performance in various operating environments and under potentially non-standard conditions. Through these comprehensive navigation performance evaluations, drone operators and manufacturers can obtain detailed data on equipment performance, which helps optimize design and improve the reliability of flight strategies. This not only improves the practical application efficiency of drones but also strengthens their ability to operate in complex environments, promoting the widespread application of drone technology in both civilian and commercial sectors. Attached Figure Description

[0128] Figure 1 This is a schematic diagram of an unmanned aerial vehicle (UAV) flight control and navigation performance evaluation system.

[0129] Figure 2 This is a schematic diagram of the general data processing process.

[0130] Figure 3 This is a schematic diagram of the connector for the real-time sensing device.

[0131] Figure 4 It is a 3D flight trajectory for flight control performance evaluation.

[0132] Figure 5 This is a schematic diagram of the X-coordinate data analysis results for flight control performance evaluation.

[0133] Figure 6 This is a schematic diagram of the Y-coordinate data analysis results for flight control performance evaluation.

[0134] Figure 7 This is a schematic diagram of the Z-coordinate data analysis results for flight control performance evaluation.

[0135] Figure 8 This is a schematic diagram of the X-coordinate data analysis results for evaluating the navigation performance of a drone.

[0136] Figure 9 This is a schematic diagram of the Y-coordinate data analysis results for evaluating the navigation performance of a drone.

[0137] Figure 10 This is a schematic diagram of the Z-coordinate data analysis results for evaluating the navigation performance of a drone. Detailed Implementation

[0138] This invention enables the evaluation of the flight control and navigation performance of the DJI Mavic 3E aircraft, specifically as follows:

[0139] Step 1: Installation of external drone real-time status sensing device

[0140] Attach the external drone real-time status sensor to the mounting holes on the DJI Mavic 3E drone and start recording the current drone status data. The real-time status sensor connector is as follows: Figure 3 As shown: Our homemade real-time sensing device is installed in the middle of the connector and connected to the connector via a shock-absorbing mount and screws. The bottom of the connector is connected to the mounting hole on the DJI Mavic 3E drone to achieve the mounting of the external drone real-time status sensing device.

[0141] Step 2: Installation of indoor and outdoor positioning devices

[0142] A lidar is used as an indoor / outdoor positioning device to obtain the drone's real-time indoor location, enabling the GNSS / RTK signal generator to produce accurate indoor GNSS / RTK positioning signals. The lidar is connected to the DJI Mavic 3E drone via a proprietary connector, the lidar positioning system is activated to obtain the drone's indoor coordinates, and these coordinates are transmitted wirelessly to the GNSS / RTK signal generator.

[0143] Step 3: Start the GNSS / RTK signal generator based on the autonomous positioning system

[0144] After obtaining the precise location of the drone indoors, we can activate the GNSS / RTK signal generator based on the autonomous positioning system to generate a precise GNSS / RTK positioning signal to enable GPS navigation for the drone. The indoor GNSS / RTK positioning signal is generated and transmitted through four steps: satellite data acquisition, signal generation, signal modulation, and signal transmission, allowing the drone to achieve position control and flight mode in indoor scenarios by relying on its own satellite positioning system.

[0145] Step 4: Drone Flight Test

[0146] After the installation and startup of the external drone real-time status perception device, indoor and outdoor positioning device and GNSS / RTK signal generator are completed, the drone can be used for flight testing.

[0147] Before flight testing, the UAV outdoor flight test equipment should undergo reprojection error correction before each power-on. The measuring equipment should be placed stationary, and its position values ​​should be continuously read for up to one minute to perform measurement accuracy convergence correction. Afterwards, hovering endurance and positioning accuracy tests, wind-resistant hovering tests, and virtual waypoint flight tests should be performed sequentially.

[0148] First, hovering endurance and positioning accuracy tests were conducted. The drone took off, hovered in a fixed-point mode, and landed after 30 minutes of flight. Then, a wind field was deployed in an indoor test scenario, and the drone was tested for hovering in the wind field. It landed after 10 minutes of flight. Finally, before takeoff, square waypoints were planned according to the size of the site, and the flight path mode was activated during the flight, allowing the drone to autonomously complete the flight mission based on the waypoints.

[0149] Step 5: Drone flight data processing and performance evaluation

[0150] During flight, the drone's flight data is acquired in real time via wireless communication and saved to a log file.

[0151] After acquiring the flight data, it is processed to facilitate subsequent evaluation of the drone's flight performance. Once the processed drone flight data is obtained, we can use this data to assess the drone's flight control performance.

[0152] Evaluation of UAV flight control performance, flight trajectory such as Figure 4 As shown in the figure, the drone took off from a designated indoor location, entered hovering and stationary mode, and landed after hovering for 30 minutes. The figure shows that the drone maintained a good hovering and stationary effect near the takeoff point for most of the 30 minutes. Even when encountering gusts of wind, a deviation of about 0.5m occurred, but this was quickly corrected. The drone hovering flight analysis results are as follows: Figures 5 to 7 As shown; Figure 5This shows the drone's tracking status in the X-coordinate. The green line represents our set value, and the red line represents the drone's status value. As you can see, the green and red lines completely overlap in the X-coordinate, indicating that the drone has excellent tracking and control capabilities in that area. Figure 6 This shows the drone's tracking performance in the Y-coordinate. The green line represents our set value, and the red line represents the drone's status value. As you can see, the green and red lines completely overlap in the Y-coordinate, indicating that the drone has excellent tracking and control capabilities in this area. Figure 7 This shows the drone's altitude tracking performance. The green line represents our set value, and the red line represents the drone's status value. As shown in the figure, the green and red lines completely overlap in altitude, indicating that the drone has excellent tracking and control capabilities. The conclusion is that the drone can achieve good self-control under the guidance of GNSS / RTK positioning signals, meeting the qualification standard. For the performance evaluation of drone navigation, the drone flies along a designated route guided by GNSS / RTK navigation signals. During flight, a route flight mode is activated, allowing the drone to autonomously complete its flight mission based on waypoints. The drone route flight analysis results are as follows... Figures 8 to 10 As shown, Figure 8 This shows the drone's X-coordinate tracking. The green line represents our set value, and the red line represents the drone's status value. As shown in the figure, the green and red lines completely overlap on the X-coordinate. Under GNSS / RTK navigation signal guidance, the drone can follow the flight path on the X-coordinate, demonstrating good flight path capability. Figure 9 This shows the drone's Y-coordinate tracking. The green line represents our set value, and the red line represents the drone's status value. As shown in the figure, the green and red lines completely overlap on the Y-coordinate. Under GNSS / RTK navigation signal guidance, the drone can follow the flight path on the Y-coordinate, demonstrating good flight path capability. Figure 10 This shows the drone's Z-coordinate tracking. The green line represents our set value, and the red line represents the drone's status value. As shown in the figure, the green and red lines completely overlap on the Z-coordinate. Under the guidance of GNSS / RTK navigation signals, the drone can follow the flight path on the Z-coordinate, demonstrating good flight path capability. Therefore, we can conclude that the drone can achieve good tracking of the target position under the guidance of GNSS / RTK positioning signals, meeting the qualified standard.

Claims

1. A method for evaluating the flight control and navigation performance of indoor unmanned aerial vehicles (UAVs) based on autonomous positioning GNSS, characterized in that: Includes the following steps: Step 1: Installation of external drone real-time status sensing device An external drone real-time status perception device is mounted on the drone, and multi-dimensional information sensors are deployed on the aircraft to obtain the drone's flight status information in real time. The device contains three different types of accelerometers and gyroscopes, and processes and sends real-time data through an STM32 main processor. Step 2: Installation of indoor and outdoor positioning devices Using a lidar as an indoor / outdoor positioning device, the drone's real-time indoor location is obtained so that the GNSS / RTK signal generator can generate accurate indoor GNSS / RTK positioning signals. The lidar is connected to the drone via a connector, the lidar positioning system is activated to obtain the drone's indoor coordinates, and the indoor coordinates are sent to the GNSS / RTK signal generator via wireless communication. Step 3: Start the GNSS / RTK signal generator based on the autonomous positioning system After obtaining the precise location of the drone indoors, the GNSS / RTK signal generator based on the autonomous positioning system is activated to generate precise GNSS / RTK positioning signals to enable GPS navigation for the drone, allowing the drone to achieve position control flight mode in indoor scenarios by relying on its own satellite positioning system. Step 4: Drone Flight Test After completing the installation and startup of the real-time status perception device, indoor and outdoor positioning device and GNSS / RTK signal generator, the UAV was started to begin indoor flight testing. Step 5: Drone flight data processing and performance evaluation This includes data acquisition, data processing, drone flight control performance evaluation, and drone navigation performance evaluation.

2. The method for evaluating the flight control and navigation performance of indoor unmanned aerial vehicles based on autonomous positioning GNSS according to claim 1, characterized in that: Step 1 includes the following three types of status information: UAV attitude information: Real-time angular velocity and acceleration information of the UAV are obtained by using an inertial sensor. The estimated values ​​of the angle, angular velocity and angular acceleration of the UAV in the pitch, roll and yaw directions are obtained by using a Kalman filter. UAV position information: The UAV position information is obtained by the onboard position acquisition device and fused with the information from the airborne inertial sensor and altitude sensor. An extended Kalman filter is used to fuse the multi-dimensional sensor information to obtain more accurate and smooth three-dimensional position estimation information. UAV vibration information: Using inertial sensors to sense information, generate power spectrum density information and real-time vibration intensity information of the UAV to assess the vibration state of the UAV in flight.

3. The method for evaluating the flight control and navigation performance of indoor unmanned aerial vehicles based on autonomous positioning GNSS according to claim 1, characterized in that: In step 2, SLAM is based on two core mathematical models: the state estimation equation and the observation equation. State estimation equation: x t =f(x t-1 ,u t ,w t ) Where, x t The state of the drone at time t, including position and orientation; x t-1 The state at the previous moment; u t For control input; w t This is process noise; The state estimation equation describes how the state of the UAV evolves from one time step to the next, depending on its control commands and inherent uncertainties; Observation equation: z t =h(x t ,n t ) Among them, z t x represents the observation data obtained by the lidar sensor at time t; t n represents the robot's current estimated state. t The noise level represents the measurement error.

4. The method for evaluating the flight control and navigation performance of indoor UAVs based on autonomous positioning GNSS according to claim 1, characterized in that: Step 3 also includes: satellite data acquisition; Before generating a simulated GPS signal, the GPS ephemeris file for the target time must first be obtained. The ephemeris file contains precise orbital data of the satellites. This data is used to calculate the satellite's position at any given time using Kepler's equations. E = M + e sin E Where E is the anomalous angle; M is the mean anomalous angle, calculated as: M = n(t - t0); n is the average motion, representing the average angular velocity of the satellite orbiting the Earth once; t is the observation time; t0 is the reference time, usually the time it takes for the satellite to pass the perigee; e is the orbital eccentricity, describing the flattening of the satellite's orbit. Based on the angle of asymmetry, the satellite's position in the orbital plane is further calculated: x′=a(cos Ee) Where x′ and y′ are the coordinates of the satellite in the orbital plane; a is the semi-major axis, representing the size of the satellite's orbit; These two formulas are used to calculate the satellite's position on its orbital plane, where x′ is along the direction from the Earth's center to the perigee, and y′ is perpendicular to x′. In reality, the satellite's orbit is inclined relative to the Earth's equatorial plane, so it is also necessary to transform the position of the orbital plane to the geocentric coordinate system: x=x′cos(Ω)-y′cos(i)sin(Ω) y=x′sin(Ω)+y′cos(i)cos(Ω) z = y′sin (i) Where Ω is the right ascension of the ascending node, and i is the orbital inclination.

5. The method for evaluating the flight control and navigation performance of indoor unmanned aerial vehicles based on autonomous positioning GNSS according to claim 4, characterized in that: Step 3 also includes: signal generation; after acquiring satellite data, the line-of-sight direction and pseudorange from the satellite to the set receiver position are calculated using the indoor positioning information and base point coordinates obtained by the motion capture system, and then the corresponding GPS signal is calculated and generated from the satellite data, including generating pseudo-random noise code PRN and encoding navigation messages. Each GPS satellite is assigned a unique PRN code, which is a repeating binary sequence used in receivers to help identify and synchronize with specific satellite signals; Where C(t) is the output code at time T, and G i (t) represents the state of the code generator at time t, D code It is a data code, and Δt is the duration of the code element; The navigation message contains the satellite's orbital parameters, clock correction parameters, and health status information; Where D(t) is the navigation data word at time t, b k It is the k-th bit of binary data, and n is the length of the data word.

6. The method for evaluating the flight control and navigation performance of indoor unmanned aerial vehicles based on autonomous positioning GNSS according to claim 5, characterized in that: Step 3 also includes: signal modulation; using BPSK to modulate navigation data and PRN codes onto a carrier frequency to form a transmittable signal; φ(t)=π×(d(t)+c(t))mod2 Where d(t) is the navigation data bit and c(t) is the PRN code; After acquiring the integrated signal, BPSK modulation is used to encode the data onto the RF carrier. This method represents the data bits by changing the phase of the carrier. s(t)=A·cos(2πf c t+φ(t)) Where s(t) is the modulated signal, A is the signal amplitude, and f is the amplitude of the signal. c φ(t) is the carrier frequency, and φ(t) is the phase of time t, which depends on the number of data bits. During BPSK modulation, in-phase I and quadrature Q components are extracted from the modulated signal to form IQ signals; this is accomplished by analyzing the modulated signal into its sine and cosine components, enabling the signal to carry more information and improving transmission efficiency. I(t)=d(t)cos(2πf c t) Q(t)=d(t)sin(2πf c t) The generated analog signal needs to be sampled and converted into a format suitable for SDR hardware processing; the mathematical processing involved in the sampling process is to discretize the analog signal, converting the continuous analog signal into a discrete signal for digital storage and processing; the phase and amplitude of each sampling point are calculated according to its corresponding time point to ensure that the integrity and accuracy of the signal are preserved; s[n]=A·cos(2πf c nT+φ[n]) Where s[n] is the signal value at sampling time n, and T is the sampling interval.

7. The method for evaluating the flight control and navigation performance of indoor unmanned aerial vehicles based on autonomous positioning GNSS according to claim 6, characterized in that: Step 3 also includes: transmitting GNSS / RTK positioning signals using a software radio-based GPS signal transmission system; First, the SDR device needs to receive the generated digital I / Q data from the GPS analog signal data stream generation system. This data is the baseband representation of the GPS signal. Then, the in-phase and quadrature components in the data stream are combined into a complex signal for subsequent processing and modulation. s[n] = I[n] + jQ[n] Wherein, s[n] is the baseband signal in complex form, and I[n] and Q[n] are the in-phase and quadrature components at sampling point n, respectively; After signal compositing, the complex signal is converted into an analog signal for subsequent frequency conversion and transmission. This step is achieved through a digital-to-analog converter (DAC), which converts the processed digital signal into an analog signal to prepare for transmission. The formula below... This represents the mixing process of the signal and carrier, which upscals the baseband signal to the RF band. Where v(t) is an analog signal, f c It is the carrier frequency, and s(t) is the continuous-time complex signal obtained by interpolation and filtering from the digital baseband signal; After the analog signal conversion is completed, the analog signal needs to be frequency-converted to the GPS operating frequency; this is a crucial step before signal transmission, ensuring the signal is sent within the correct frequency band; the following formula describes the interaction between the baseband signal and the local oscillator (LO) frequency f through the analog mixer. RF The mixing process: x(t)=v(t)cos(2πf RF t)-u(t)sin(2πf RF t) Where x(t) is the up-converted RF signal, u(t) and v(t) are the I and Q components of the baseband signal, respectively, and f RF It is the target radio frequency; Finally, the signal power is amplified to a level sufficient for transmission through the antenna by the RF amplifier of the SDR device, ensuring effective signal transmission; P out =Gain×P in Among them, P out P is the amplified output power, Gain is the amplifier input power, and P is the output power. in It is the amplifier gain.

8. The method for evaluating the flight control and navigation performance of indoor unmanned aerial vehicles based on autonomous positioning GNSS according to claim 1, characterized in that: In step 4, the accuracy of the UAV outdoor flight test device is calibrated. The UAV outdoor flight test device should be reprojected error calibrated before each power-on. The measuring device is placed in place and the measuring device continuously reads the position value within one minute to perform measurement accuracy convergence calibration. After the calibration of the test device and the UAV itself is completed, the flight test begins.

9. The method for evaluating the flight control and navigation performance of indoor unmanned aerial vehicles based on autonomous positioning GNSS according to claim 1, characterized in that: Step 5, regarding the evaluation of the UAV flight control performance, is divided into several parts, as follows: Basic performance assessment, including: flight stability and environmental adaptability; Emergency response capabilities include: emergency obstacle avoidance and autonomous landing; Accuracy testing includes: position holding capability and path accuracy; Endurance and stability testing, including continuous flight performance and control stability.

10. The method for evaluating the flight control and navigation performance of indoor unmanned aerial vehicles based on autonomous positioning GNSS according to claim 1 or 9, characterized in that: Step 5, regarding the evaluation of UAV navigation performance, is divided into several parts, as follows: Basic navigation capability: In the basic navigation capability test of the UAV, ensure the flight accuracy of the UAV on a fixed route; Altitude and speed control: ensuring the stability of the drone in maintaining a specific flight altitude and speed; Attitude stability: The attitude stability test for UAVs focuses on evaluating the stability of the machine under various flight attitudes; Sensor performance evaluation: Sensor performance evaluation focuses on inertial measurement units (IMUs) and global navigation satellite systems (GNSS). Emergency response: Ensure the drone's ability to handle navigation errors or sudden changes to its flight plan.

Images