Multi-sensor fusion unmanned aerial vehicle environment monitoring and early warning system

Abstract

The invention discloses a multi-sensor fusion unmanned aerial vehicle environment monitoring and early warning system, and relates to the technical field of environment monitoring and early warning, and the system comprises the following components: an unmanned aerial vehicle platform which is used as a carrying basis of sensors and data processing equipment and has excellent flight stability and strong cruising ability; according to the invention, by introducing a task planning mechanism based on autonomous learning, the unmanned aerial vehicle can continuously learn and optimize the flight strategy when executing the monitoring task, the innovation point significantly improves the efficiency of task execution, and the unmanned aerial vehicle can perform real-time monitoring according to the actual environment feedback and historical task data. The unmanned aerial vehicle can automatically adjust the parameters of the flight path, speed and height to avoid adverse factors and select the optimal path, and meanwhile, the unmanned aerial vehicle is endowed with strong scene adaptability by the autonomous learning ability, so that the unmanned aerial vehicle can quickly respond to a complicated and changeable natural environment or a sudden emergency, and smooth completion of a task is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring and early warning, and in particular to a multi-sensor fusion unmanned aerial vehicle environmental monitoring and early warning system. Background Art

[0002] With the rapid development of UAV technology, UAVs have shown great application potential in many fields such as environmental monitoring, target tracking, and disaster relief. In order to improve the mission execution efficiency and adaptability of UAVs, mission planning technology has become a key link in the UAV technology system. Mission planning aims to plan the optimal flight path and action strategy for UAVs based on mission requirements and environmental conditions to ensure efficient and safe completion of the mission.

[0003] Although traditional mission planning technology has achieved certain results in UAV applications, it still has many shortcomings in autonomous learning-based mission planning. Traditional technology often relies on preset algorithm models and manual intervention, lacks flexibility and autonomy, and is difficult to automatically adjust flight strategies based on real-time environmental feedback and historical mission data when faced with complex and changeable monitoring scenarios. This results in low mission execution efficiency and even inability to adapt to sudden environmental changes. In addition, traditional mission planning technology lacks the ability to cultivate autonomous decision-making capabilities of UAVs, limiting the application and development of UAVs in a wider range of fields.

[0004] Therefore, a multi-sensor fusion UAV environmental monitoring and early warning system was developed. This system not only overcomes the limitations of traditional mission planning technology, but also significantly improves the mission execution efficiency and intelligence level of UAVs, providing more reliable and efficient technical support for the application of UAVs in environmental monitoring and target tracking. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a multi-sensor fusion UAV environmental monitoring and early warning system, which can realize the continuous learning and self-optimization of the UAV in the process of mission execution through deep fusion algorithm formula and autonomous learning mechanism. It not only significantly improves the adaptability of the UAV to complex and changing environments, but also greatly improves the efficiency and accuracy of mission execution. Through in-depth analysis of historical mission data and environmental feedback, the UAV can autonomously adjust its flight strategy to ensure that it can maintain the optimal state in various monitoring scenarios.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a multi-sensor fusion UAV environment monitoring and early warning system, which includes the following components: UAV platform: As the basis for carrying sensors and data processing equipment, it has excellent flight stability and strong endurance, and can perfectly adapt to long-term and large-scale monitoring tasks; Multi-sensor module: covers a variety of detection equipment including high-definition cameras, infrared thermal imagers, lidars, meteorological sensors and gas detectors, which can capture various environmental information in real time; Data acquisition and processing unit: receives data from multiple sensor modules, filters, denoises and fuses the data, and integrates the information from different sensors; Self-learning module: The internal algorithm formula can automatically identify the optimal monitoring strategy under different terrain and meteorological conditions by deeply learning historical mission data, environmental feedback and real-time monitoring data; Mission planning optimization unit: Based on the output results of the self-learning module, combined with the current environmental conditions and target monitoring requirements, it automatically generates and optimizes the flight mission plan, and uses algorithm formulas to find the best flight path and monitoring points; Early warning output module: Based on the results of data processing and analysis, it provides real-time early warnings for existing environmental risks, such as excessive pollution, fire hazards, and geological disasters, and quickly sends the early warning information to the ground control center through wireless communication technology.

[0007] Furthermore, the autonomous learning module of the UAV platform integrates an autopilot, attitude sensor, gyroscope, and accelerometer integrated flight control system, uses algorithms to accurately control the attitude of the power system, and can fly stably in complex weather conditions and changing environments. It supports high-precision, long-term monitoring tasks. The platform is equipped with an efficient energy system, sensors and data processing equipment, and can collect, process and analyze data in real time during flight. It is also equipped with a complete fault warning and self-diagnosis system, emergency landing and recovery system, and anti-collision and obstacle avoidance system.

[0008] Furthermore, the autonomous learning module collects historical mission data, environmental feedback data and real-time monitoring data, and carries out data cleaning, feature extraction and data standardization. In the model selection and training stage, a preliminary monitoring strategy index is selected based on the preprocessed data to generate preliminary monitoring strategies for different terrains and meteorological conditions, and the strategy evolution index is used to iteratively optimize the search strategy. When the UAV performs a mission, the flight altitude, speed, heading and sensor parameters are dynamically adjusted according to the generated monitoring strategy through the dynamic adaptation adjustment index. At the same time, monitoring data and environmental feedback are collected in real time to evaluate the current strategy. After the mission is completed, the autonomous learning and strategy execution process is comprehensively evaluated, and it is automatically iterated and upgraded based on the evaluation results to optimize the internal algorithm and model structure.

[0009] Furthermore, the autonomous learning module generates preliminary monitoring strategies for different terrains and meteorological conditions through the preliminary monitoring strategy index, collects terrain height data H, slope data S, meteorological temperature data T, humidity data RH, and wind speed data V under various terrains and meteorological conditions, and pre-processes these data, considering terrain factors and terrain complexity index TI, and the calculation formula is: TI= *H+bS, where a and b are weight coefficients, taking into account meteorological factors and the meteorological impact index MI, the calculation formula is: MI=cT+d*RH+e*V, where c, d and e are weight coefficients, comprehensive terrain and meteorological factors, and the preliminary monitoring strategy index MSI is obtained, and the calculation formula is: MSl *TI *MI+h, where f and g are weight coefficients and h is a constant term.

[0010] Furthermore, the autonomous learning module iteratively optimizes the search strategy through the strategy evolution index. Assume that the initial monitoring strategy is , the goal is to find the optimal monitoring strategy , the calculation formula of the strategy evolution index is defined as: ,in, It is An evaluation function is used to measure the monitoring strategy In different aspects of performance, is the corresponding weight coefficient, randomly generating the initial monitoring strategy , set the number of iterations , maximum number of iterations , through The decision variables in the random perturbation generate a new monitoring strategy , the calculation formula is: ,in, Indicated in The current monitoring strategy at iteration , is a small random vector, computing and , , accept the new strategy, otherwise, with probability , accept the new strategy, when When , stop the iteration. For the optimized monitoring strategy .

[0011] Furthermore, the autonomous learning module adjusts the flight altitude, speed, heading and sensor parameters of the drone in real time through a dynamic adaptive adjustment index, and the dynamic adaptive adjustment index formula is defined as: ,in: is the current flight altitude, is the target flight altitude, is the current flight speed, is the target flight speed, is the heading angle at the current moment, is the target heading angle, is the sensor parameter value at the current moment, is the target sensor parameter value, , ,γ, is the weight coefficient, which corresponds to the importance of flight altitude, speed, heading and sensor parameters, and calculates the current Values, for the adjustment of flight altitude, speed, heading and sensor parameters, according to The adjustment amount is determined based on the size and change trend of the flight altitude. The calculation formula for the adjustment amount of the flight altitude is: ,in is an adjustment factor, and the adjustment of speed, heading and sensor parameters is calculated similarly , and apply the adjustment amount to the UAV’s control system to adjust the flight altitude, speed, heading, and sensor parameters in real time.

[0012] Furthermore, the mission planning optimization unit receives the optimal monitoring strategy provided by the autonomous learning module, and integrates the current environmental data and specific requirements of target monitoring. Based on the target monitoring requirements and environment, a preliminary flight path is generated using a comprehensive attractiveness index. On a known map, the shortest path from the starting point to the end point is found using an optimization algorithm through lidar on the map data. The flight path is further optimized, and the energy consumption and monitoring efficiency under different path planning are evaluated based on the UAV flight performance parameters and current environmental conditions. The final flight mission plan is generated based on the evaluation results.

[0013] Furthermore, the mission planning optimization unit receives the preliminary flight path covering all monitoring points generated by the autonomous learning module through the comprehensive attraction index, and has monitoring points, each with three-dimensional coordinates Indicates that Represents altitude information. The drone starts from a starting point. Start and find a flight path that covers all monitoring points. For each monitoring point , and define its comprehensive attractiveness index as: ,in Is the starting point To the monitoring point The distance is calculated using the three-dimensional Euclidean distance formula: , It is a monitoring point The average distance to all other monitoring points, It is a monitoring point The weight of , , γ is the weight coefficient, which is used to adjust the relative importance of the three factors. The monitoring point with the highest comprehensive attractiveness index is selected as the first point to be visited. For the selected monitoring points, the comprehensive attractiveness index of other unvisited monitoring points is calculated. Among the unvisited monitoring points, the point with the highest comprehensive attractiveness index is selected as the next point to be visited. This process is repeated until all monitoring points are visited, and the last monitoring point and the starting point are connected to form a closed path.

[0014] Furthermore, the mission planning optimization unit receives the specific steps of optimizing the preliminary flight path by the autonomous learning module using the optimization algorithm; (1) Receive information on the initial flight path’s starting point, end point, waypoints, estimated flight altitude, and speed limit; (2) Preprocessing the preliminary path, removing redundant points, smoothing to reduce sharp turns, and marking potential risk areas; (3) Construct a comprehensive cost model based on the specific requirements of the flight mission and assign weights to each cost factor; (4) Using an optimization algorithm to iteratively optimize the preprocessed path with the goal of minimizing the total cost; (5) Evaluate multiple candidate optimization paths generated by the algorithm and select the best path as the final flight path; (6) Output the optimized flight path to the user and the flight control system in a visual form and provide detailed parameters of the path; (7) During the flight, the flight status is monitored in real time and the flight path is dynamically adjusted based on the real-time data.

[0015] Furthermore, the early warning output module receives environmental parameter data including air quality, temperature, humidity, fire detector signals and geological activity monitoring in real time through multiple environmental monitoring sensors and data sources, performs preliminary processing and analysis to identify potential environmental risk events, further analyzes the identified risk events, evaluates the environmental risk level, and sends the early warning information to the ground control center, emergency department and receiving terminal in real time according to preset communication protocols and coding rules, receives confirmation receipts from the receiving end, and records all operations, data changes and communication records of the early warning process in detail.

[0016] Compared with the existing technology, the multi-sensor fusion UAV environmental monitoring and early warning system has the following beneficial effects: 1. The present invention introduces a task planning mechanism based on autonomous learning, so that the UAV can continuously learn and optimize its flight strategy when performing monitoring tasks. This innovation significantly improves the efficiency of task execution, because the UAV can automatically adjust the flight path, speed, and altitude parameters according to actual environmental feedback and historical task data to avoid unfavorable factors and select the optimal path. At the same time, the autonomous learning ability also gives the UAV strong scene adaptability. Whether facing complex and changeable natural environments or sudden emergencies, the UAV can respond quickly to ensure the smooth completion of the task.

[0017] 2. The present invention realizes the intelligence and autonomy of UAV decision-making through self-learning mission planning. During the execution of the mission, the UAV can continuously collect and analyze environmental information, and combine historical mission data to form a deep understanding of the monitoring scene. Based on this, the UAV can make more accurate and efficient decisions independently, such as dynamically adjusting the distribution of monitoring points and optimizing flight paths, thereby greatly reducing the burden of manual operation and improving the intelligence level of monitoring work. This autonomous decision-making ability not only improves the flexibility of mission execution, but also provides strong technical support for the future application of UAVs in a wider range of fields.

[0018] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 This is the process operation diagram of the multi-sensor fusion UAV environmental monitoring and early warning system; Figure 2 This is a schematic diagram of the UAV structure; Figure 3 This is the process framework diagram of the self-learning module. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Embodiment 1: This embodiment provides a multi-sensor fusion UAV environmental monitoring and early warning system. The system integrates multiple sensors such as high-definition cameras and infrared thermal imagers, and has excellent flight stability and long-term endurance. Through data collection and processing, autonomous learning, mission planning optimization and early warning output modules, it realizes all-round, real-time monitoring and early warning of complex environments. The autonomous learning module combines historical data with real-time monitoring data, and uses algorithms to continuously optimize flight strategies to ensure efficient monitoring of UAVs under different terrain and meteorological conditions. The mission planning optimization unit automatically generates and optimizes flight paths based on autonomous learning results to improve monitoring efficiency. The system has broad application prospects in the fields of environmental monitoring and disaster early warning, and sets a new benchmark for the intelligent development of UAV technology.

[0023] In the specific implementation, the multi-sensor module covers a variety of equipment, including high-definition cameras, which are responsible for capturing high-resolution image data and recording visual information of surface coverage, vegetation growth, and human activities in the monitored area in real time. Infrared thermal imagers use infrared radiation detection technology to obtain temperature distribution maps of the monitored area in real time and identify abnormally high and low temperature areas, such as fire hazards and heat leaks. LiDAR accurately measures the three-dimensional terrain structure, vegetation height, and building distribution information of the monitored area by emitting laser pulses and receiving echo signals. Meteorological sensors integrate temperature, humidity, air pressure, wind speed, and wind direction to monitor multiple meteorological parameters and collect meteorological data of the monitored area in real time. Gas detectors are equipped with high-sensitivity gas sensors to detect the concentration of harmful gases in the monitored area in real time, such as sulfur dioxide, nitrogen oxides, and ozone. Through the collaborative work of multiple sensor modules during the flight of the drone, various environmental information is captured and transmitted to the data acquisition and processing unit for subsequent processing and analysis.

[0024] The data acquisition and processing unit receives raw data from multi-sensor modules including high-definition cameras, infrared thermal imagers, lidars, meteorological sensors and gas detectors, and processes the received data using digital filtering technology to remove high-frequency noise and random errors. It further removes interference components in the data through statistical methods and signal processing technology, performs time synchronization processing on data from different sensors, and calibrates data deviations between sensors. It extracts key features from the preprocessed data, such as edge information in the image, temperature distribution in the infrared image, and point cloud data of the lidar. It uses data association algorithms to match and associate feature information from different sensors, and deeply fuses data from different sensors to comprehensively generate more comprehensive and accurate environmental information. It outputs the fused data to the autonomous learning module and other required data processing units to provide basic data support for subsequent monitoring strategy optimization, task planning and early warning output.

[0025] The autonomous learning module collects information from the UAV mission history database, real-time environmental monitoring data and real-time sensor data, performs data cleaning to remove noise and outliers, extracts features for fire hotspot identification and pollution area definition, and standardizes the data to ensure data consistency and comparability. Based on the preprocessed data, the preliminary monitoring strategy index is used to construct preliminary monitoring strategies for different terrains, such as mountainous areas, plains, and meteorological conditions, such as sunny days, rainy days, and foggy days, and the strategy evolution index is used to iteratively optimize the search strategy. According to the generated monitoring strategy, the flight altitude, speed, heading and sensor parameters are dynamically adjusted through the dynamic adaptation adjustment index to adapt to the complex and changeable monitoring environment. During the task execution, monitoring data and environmental feedback information are collected in real time to evaluate the effectiveness of the current monitoring strategy. If the strategy is not suitable for the current environment, the emergency adjustment mechanism is immediately activated for subsequent analysis and optimization. After the task is completed, the system conducts a comprehensive evaluation of the autonomous learning and strategy execution process, including indicators of monitoring accuracy, response time, and resource consumption. Based on the evaluation results, the algorithm and model structure are automatically iterated and upgraded to improve the overall performance and adaptability of the system.

[0026] The mission planning module receives the optimal monitoring strategy from the autonomous learning module, combines the current environmental data, such as wind direction, wind speed, terrain, and specific requirements for target monitoring, such as the distribution of monitoring points and monitoring frequency, and generates a preliminary flight path covering all monitoring points through a comprehensive attraction index. On a known map, the laser radar is used on the map data to find the shortest path from the starting point to the end point using an optimization algorithm, and the flight path is further optimized to reduce flight time and energy consumption. The energy consumption and monitoring efficiency under different path planning are evaluated based on the flight performance parameters of the UAV, such as maximum speed, climb rate, and endurance, and the current environmental conditions, such as wind influence and flight altitude limit. Based on the evaluation results, the optimal path is selected as the final flight mission planning, and specific mission parameters such as flight altitude, speed, heading, and sensor parameters are set. During the execution of the mission, the system continuously monitors environmental changes and UAV status, such as battery power, sensor status, and flight attitude, and uses real-time data for dynamic evaluation. If environmental changes are found, such as new monitoring needs, sudden environmental events, and abnormal UAV status, the dynamic adjustment mechanism is immediately activated to dynamically adjust the flight path, monitoring strategy, and mission parameters according to real-time data to ensure that the mission is completed efficiently and safely.

[0027] The early warning output module receives the raw data from the monitoring sensors of the air quality monitor, temperature sensor, humidity sensor, fire detector, and geological activity monitor in real time through the communication interface on the drone, and performs preliminary processing on the received raw data, including data cleaning to remove noise, outliers, data verification, and data formatting, in preparation for subsequent analysis. It uses the preset algorithm to perform preliminary analysis on the processed data to identify possible environmental risk events, such as air quality exceeding the standard, fire hazards, and geological disasters. According to the nature, severity, and possible impact range of the risk event, it further evaluates the environmental risk level, determines the urgency of issuing the early warning signal, and According to the communication protocols and coding rules set, the warning information, including key information such as risk type, level, location, and time, is encoded and packaged to ensure the accuracy and readability of the information, and sent to the ground control center, emergency department and designated receiving terminal in real time to ensure the timely communication of information, receive confirmation receipts from the receiving end to confirm that the warning information has been successfully received and understood, and keep detailed records of all operations and data changes in the warning process) and communication records (such as for subsequent inquiries and audits. After the warning information is issued, the warning output module continues to receive feedback information from the on-site receiving end to understand the warning response situation, adjust the warning strategy and provide further support as needed.

[0028] Embodiment 2: The multi-sensor fusion UAV environmental monitoring and early warning system in this embodiment places special emphasis on using algorithm formulas to allow the UAV to continuously learn and optimize mission planning during mission execution. By analyzing historical mission data and environmental feedback, the UAV automatically adjusts its flight strategy to better adapt to different monitoring scenarios.

[0029] Historical mission data, environmental feedback data and real-time monitoring data are comprehensively collected, and a series of data preprocessing steps are carried out, including data cleaning, removal of invalid and erroneous data, extraction of feature information useful for monitoring tasks, and data standardization, so that data of different magnitudes can be compared on the same scale.

[0030] Based on the preprocessed data, the autonomous learning module uses algorithm formulas to generate preliminary monitoring strategies for different terrains and meteorological conditions. These preliminary strategies are designed to formulate preliminary flight altitude, speed, heading and sensor parameter settings according to different environmental characteristics to ensure that the UAV can complete the monitoring task efficiently and accurately.

[0031] In order to further improve the effect of the monitoring strategy, the autonomous learning module uses the preliminary monitoring strategy index to iteratively optimize the preliminary monitoring strategy. The preliminary monitoring strategy index simulates the process of human problem solving and searches for the optimal solution in the search space. It collects terrain height data H, slope data S, meteorological temperature data T, humidity data RH, and wind speed data V under various terrain and meteorological conditions, pre-processes these data, removes outliers, and standardizes them, taking into account terrain factors and terrain complexity index TI. The calculation formula is: TI= *H+bS, where a and b are weight coefficients, which are adjusted according to actual conditions, taking into account meteorological factors and the meteorological impact index MI, the formula is: MI=cT+d*RH+e*V, where c, d and e are weight coefficients, comprehensive terrain and meteorological factors, and the preliminary monitoring strategy index MSI is obtained, and the calculation formula is: MSl *TI *MI+h, where f and g are weight coefficients and h is a constant term that is adjusted according to specific monitoring objectives and needs. A higher MSI value indicates the need for a stricter and more frequent monitoring strategy, and a corresponding increase in the density of monitoring equipment and monitoring frequency. A lower MSI value can appropriately reduce the intensity of monitoring.

[0032] The search strategy is iteratively optimized through the strategy evolution index. The initial monitoring strategy is set as , the goal is to find the optimal monitoring strategy , the calculation formula of the strategy evolution index is defined as: ,in, It is An evaluation function is used to measure the monitoring strategy Performance in different aspects, such as monitoring effect, resource utilization efficiency, is the corresponding weight coefficient, which is adjusted according to actual needs and randomly generates the initial monitoring strategy , set the number of iterations , maximum number of iterations , through The decision variables in the random perturbation generate a new monitoring strategy , the calculation formula is: ,in, Indicated in The current monitoring strategy at iteration , is a small random vector, computing and , , then accept the new strategy, otherwise, with probability , Acceptance of new strategy When the comprehensive performance of the new strategy is not as good as the current strategy, instead of rejecting the new strategy directly, we use the probability Decide whether to accept the new strategy. The probability is calculated as , which helps the algorithm avoid falling into local optimal solutions and accept new strategies. When , stop the iteration. For the optimized monitoring strategy When the UAV is performing a mission, the autonomous learning module will adjust the flight altitude, speed, heading and sensor parameters of the UAV in real time according to the generated monitoring strategy through the dynamic adaptive adjustment index. The dynamic adaptive adjustment index formula is defined as: ,in: is the current flight altitude, is the target flight altitude, is the current flight speed, is the target flight speed, is the heading angle at the current moment, is the target heading angle, is the sensor parameter value at the current moment, is the target sensor parameter value, , ,γ, is the weight coefficient, which corresponds to the importance of flight altitude, speed, heading and sensor parameters, and calculates the current Value, if If the value is large and shows an upward trend, it means that the current state is far from the target state and is deteriorating, and a large adjustment is needed; if The value is small and stable, indicating that it is close to the target state. Small adjustments are made to the flight altitude, speed, heading and sensor parameters according to the The adjustment amount is determined based on the size and change trend of the flight altitude. The calculation formula for the adjustment amount of the flight altitude is: ,in is an adjustment factor, determined experimentally, and the adjustment amounts for speed, heading, and sensor parameters are similarly calculated , the adjustment amount is applied to the control system of the UAV, and the flight altitude, speed, heading and sensor parameters are adjusted in real time. At the same time, monitoring data and environmental feedback are collected in real time to evaluate the effectiveness of the current strategy. If problems and poor results are found in the current strategy during the execution process, adjustments will be made immediately to ensure the smooth progress of the monitoring mission.

[0033] After the task is completed, the autonomous learning module will conduct a comprehensive evaluation of the entire learning and strategy execution process. The evaluation content includes but is not limited to the strategy execution effect, resource utilization efficiency and system adaptability. Based on the evaluation results, the system can automatically iterate and upgrade, optimize the internal algorithm and model structure to improve the overall performance and adaptability. This process is cyclical. With the continuous execution of tasks and the continuous accumulation of data, the system's monitoring strategy will become more and more perfect, and performance will be continuously improved.

[0034] Receive the optimal monitoring strategy provided by the self-learning module, integrate the current environmental data and the specific requirements of the target monitoring, and use the comprehensive attractiveness index to generate a preliminary flight path based on the target monitoring requirements and environment. monitoring points, each with three-dimensional coordinates Indicates that Represents altitude information. The drone starts from a starting point. Start and find a flight path that covers all monitoring points. For each monitoring point , and define its comprehensive attractiveness index as: ,in Is the starting point To the monitoring point The distance is calculated using the three-dimensional Euclidean distance formula: , It is a monitoring point The average distance to all other monitoring points is calculated by The distance to each other monitoring point is averaged and obtained. It is a monitoring point The weight is determined according to the importance of the monitoring point and the complexity of the terrain. , , γ is the weight coefficient, which is used to adjust the relative importance of the three factors. The monitoring point with the highest comprehensive attractiveness index is selected as the first point to be visited. For the selected monitoring points, the comprehensive attractiveness index of other unvisited monitoring points is calculated. Among the unvisited monitoring points, the point with the highest comprehensive attractiveness index is selected as the next point to be visited. This process is repeated until all monitoring points are visited. The last monitoring point and the starting point are connected to form a closed path.

[0035] On a known map, the laser radar is used on the map data to find the shortest path from the starting point to the end point using an optimization algorithm, and the flight path is further optimized. The information of the starting point, end point, waypoints, expected flight altitude, and speed limit of the preliminary flight path is received. The preliminary path is preprocessed to remove redundant points and smooth to reduce sharp turns, and potential risk areas are marked. According to the specific needs of the flight mission, a comprehensive cost model is constructed, and weights are assigned to each cost factor. The preprocessed path is iteratively optimized using an optimization algorithm, with the goal of minimizing the total cost. Multiple candidate optimization paths generated by the algorithm are evaluated, and the optimal path is selected as the final flight path. The optimized flight path is output to the user and the flight control system in a visual form, and detailed parameters of the path are provided. During the flight, the flight status is monitored in real time, the flight path is dynamically adjusted according to the real-time data, and the energy consumption and monitoring efficiency under different path planning are evaluated according to the UAV flight performance parameters and current environmental conditions, and the final flight mission plan is generated based on the evaluation results.

[0036] Embodiment three: In the multi-sensor fusion UAV environmental monitoring and early warning system of the present invention, the UAV platform, as the core carrier of the entire system, integrates key equipment such as autopilot, attitude sensor, gyroscope and accelerometer to form a highly integrated flight control system. The system controls the power system through precise algorithms to achieve stable flight of the UAV in complex meteorological conditions and changing environments, ensuring high precision and long-term sustainability of monitoring tasks.

[0037] During the specific implementation process, the UAV platform is equipped with an efficient energy system, which can flexibly select the optimal power configuration according to the specific needs of the mission, effectively extend the flight time and improve the efficiency of mission execution. The platform design adopts a modular approach, which is convenient for rapid configuration and upgrading according to the specific requirements of environmental monitoring tasks. At the same time, it simplifies the maintenance process and reduces operating costs.

[0038] In order to further improve the compatibility and scalability of the system, the UAV platform is designed with standardized interfaces and data processing units, which can easily carry various high-precision sensors and monitoring equipment. During the flight, these sensors and devices can collect environmental data in real time and conduct instant analysis through the integrated data processing unit, providing strong support for early warning output.

[0039] In addition, the UAV platform is also equipped with a complete fault warning and self-diagnosis system, an emergency landing and recovery system, and an anti-collision and obstacle avoidance system to ensure the safe and stable operation of the UAV in complex environments. The addition of these safety measures not only improves the reliability of the system, but also reduces the risks during mission execution.

[0040] In summary, the UAV platform of the present invention realizes high-precision and long-term environmental monitoring tasks in complex environments through a highly integrated flight control system, modular design and a complete safety assurance system, providing a solid foundation for the stable operation of the entire system.

[0041] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. Multi-sensor fusion UAV environmental monitoring and early warning system, characterized by: The system consists of the following components: UAV platform: As the basis for carrying sensors and data processing equipment, it has excellent flight stability and strong endurance, and can perfectly adapt to long-term and large-scale monitoring tasks; Multi-sensor module: covers a variety of detection equipment including high-definition cameras, infrared thermal imagers, lidars, meteorological sensors and gas detectors, which can capture various environmental information in real time; Data acquisition and processing unit: receives data from multiple sensor modules, filters, removes noise, and performs data fusion processing, and integrates information from different sensors; Self-learning module: The internal algorithm formula can automatically identify the optimal monitoring strategy under different terrain and meteorological conditions by deeply learning historical mission data, environmental feedback and real-time monitoring data; Mission planning optimization unit: Based on the output results of the self-learning module, combined with the current environmental conditions and target monitoring requirements, it automatically generates and optimizes the flight mission plan, and uses algorithm formulas to find the best flight path and monitoring points; Early warning output module: Based on the results of data processing and analysis, it provides real-time early warnings for existing environmental risks, such as excessive pollution, fire hazards, and geological disasters, and quickly sends the early warning information to the ground control center through wireless communication technology.

2. The multi-sensor fusion UAV environment monitoring and early warning system according to claim 1 is characterized in that: The autonomous learning module of the UAV platform integrates an autopilot, attitude sensor, gyroscope, and accelerometer integrated flight control system. It uses algorithms to accurately control the attitude of the power system, can fly stably in complex weather conditions and changing environments, and supports high-precision, long-term monitoring tasks. The platform is equipped with an efficient energy system, sensors and data processing equipment, and can collect, process and analyze data in real time during flight. It is also equipped with a complete fault warning and self-diagnosis system, emergency landing and recovery system, and anti-collision and obstacle avoidance system.

3. The multi-sensor fusion UAV environment monitoring and early warning system according to claim 1 is characterized in that: The autonomous learning module collects historical mission data, environmental feedback data and real-time monitoring data, and carries out data cleaning, feature extraction and data standardization. In the model selection and training stage, a preliminary monitoring strategy index is selected based on the preprocessed data to generate preliminary monitoring strategies for different terrains and meteorological conditions, and the strategy evolution index is used to iteratively optimize the search strategy. When the UAV performs a mission, the flight altitude, speed, heading and sensor parameters are dynamically adjusted according to the generated monitoring strategy through the dynamic adaptation adjustment index. At the same time, monitoring data and environmental feedback are collected in real time to evaluate the current strategy. After the mission is completed, the autonomous learning and strategy execution process is comprehensively evaluated, and it is automatically iterated and upgraded based on the evaluation results to optimize the internal algorithm and model structure.

4. The multi-sensor fusion UAV environment monitoring and early warning system according to claim 3 is characterized in that: The autonomous learning module generates preliminary monitoring strategies for different terrains and meteorological conditions through the preliminary monitoring strategy index, collects terrain height data H, slope data S, meteorological temperature data T, humidity data RH, and wind speed data V under various terrains and meteorological conditions, and pre-processes these data, considering terrain factors and terrain complexity index TI. The calculation formula is: TI= *H+bS, where a and b are weight coefficients, taking into account meteorological factors and the meteorological impact index MI, the calculation formula is: MI=cT+d*RH+e*V, where c, d and e are weight coefficients, comprehensive terrain and meteorological factors, and the preliminary monitoring strategy index MSI is obtained, and the calculation formula is: MSl *TI *MI+h, where f and g are weight coefficients and h is a constant term.

5. The multi-sensor fusion UAV environment monitoring and early warning system according to claim 3 is characterized in that: The autonomous learning module iteratively optimizes the search strategy through the strategy evolution index. Suppose the initial monitoring strategy is , the goal is to find the optimal monitoring strategy , the calculation formula of the strategy evolution index is defined as: ,in, It is An evaluation function is used to measure the monitoring strategy In different aspects of performance, is the corresponding weight coefficient, randomly generating the initial monitoring strategy , set the number of iterations , maximum number of iterations , through The decision variables in the random perturbation generate a new monitoring strategy , the calculation formula is: ,in, Indicated in The current monitoring strategy at iteration , is a small random vector, computing and , , accept the new strategy, otherwise, with probability , accept the new strategy, when When , stop the iteration. For the optimized monitoring strategy .

6. The multi-sensor fusion UAV environment monitoring and early warning system according to claim 3 is characterized in that: The autonomous learning module adjusts the flight altitude, speed, heading and sensor parameters of the UAV in real time through the dynamic adaptive adjustment index. The dynamic adaptive adjustment index formula is defined as: ,in: is the current flight altitude, is the target flight altitude, is the current flight speed, is the target flight speed, is the heading angle at the current moment, is the target heading angle, is the sensor parameter value at the current moment, is the target sensor parameter value, , ,γ, is the weight coefficient, which corresponds to the importance of flight altitude, speed, heading and sensor parameters, and calculates the current Values, for the adjustment of flight altitude, speed, heading and sensor parameters, according to The adjustment amount is determined based on the size and change trend of the flight altitude. The calculation formula for the adjustment amount of the flight altitude is: ,in is an adjustment factor, and the adjustment of speed, heading and sensor parameters is calculated similarly , and apply the adjustment amount to the UAV’s control system to adjust the flight altitude, speed, heading, and sensor parameters in real time.

7. The multi-sensor fusion UAV environment monitoring and early warning system according to claim 1 is characterized in that: The mission planning optimization unit receives the optimal monitoring strategy provided by the autonomous learning module, and integrates the current environmental data and specific requirements of target monitoring. Based on the target monitoring requirements and environment, it uses the comprehensive attractiveness index to generate a preliminary flight path. On a known map, it uses a laser radar to find the shortest path from the starting point to the end point on the map data, and further optimizes the flight path. It also evaluates the energy consumption and monitoring efficiency under different path planning based on the UAV flight performance parameters and current environmental conditions, and generates a final flight mission plan based on the evaluation results.

8. The multi-sensor fusion UAV environment monitoring and early warning system according to claim 7 is characterized in that: The mission planning optimization unit receives the preliminary flight path covering all monitoring points generated by the autonomous learning module through the comprehensive attraction index, and has monitoring points, each with three-dimensional coordinates Indicates that Represents altitude information. The drone starts from a starting point. Start and find a flight path that covers all monitoring points. For each monitoring point , and define its comprehensive attractiveness index as: ,in Is the starting point To the monitoring point The distance is calculated using the three-dimensional Euclidean distance formula: , It is a monitoring point The average distance to all other monitoring points, It is a monitoring point The weight of , , γ is the weight coefficient, which is used to adjust the relative importance of the three factors. The monitoring point with the highest comprehensive attractiveness index is selected as the first point to be visited. For the selected monitoring points, the comprehensive attractiveness index of other unvisited monitoring points is calculated. Among the unvisited monitoring points, the point with the highest comprehensive attractiveness index is selected as the next point to be visited. This process is repeated until all monitoring points are visited, and the last monitoring point and the starting point are connected to form a closed path.

9. The multi-sensor fusion UAV environment monitoring and early warning system according to claim 7 is characterized in that: The mission planning optimization unit receives the specific steps of the autonomous learning module optimizing the preliminary flight path using the optimization algorithm; (1) Receive information on the initial flight path’s starting point, end point, waypoints, estimated flight altitude, and speed limit; (2) Preprocessing the preliminary path, removing redundant points, smoothing to reduce sharp turns, and marking potential risk areas; (3) Construct a comprehensive cost model based on the specific requirements of the flight mission and assign weights to each cost factor; (4) Using an optimization algorithm to iteratively optimize the preprocessed path with the goal of minimizing the total cost; (5) Evaluate multiple candidate optimization paths generated by the algorithm and select the best path as the final flight path; (6) Output the optimized flight path to the user and the flight control system in a visual form and provide detailed parameters of the path; (7) During the flight, the flight status is monitored in real time and the flight path is dynamically adjusted based on the real-time data.

10. The multi-sensor fusion UAV environment monitoring and early warning system according to claim 1 is characterized in that: The early warning output module receives environmental parameter data including air quality, temperature, humidity, fire detector signals and geological activity monitoring in real time through multiple environmental monitoring sensors and data sources, performs preliminary processing and analysis to identify potential environmental risk events, further analyzes the identified risk events, evaluates the environmental risk level, and sends the early warning information to the ground control center, emergency department and receiving terminal in real time according to preset communication protocols and coding rules, receives confirmation receipts from the receiving end, and records all operations, data changes and communication records of the early warning process in detail.

Images