Position correction method and device applied to unmanned aerial vehicle laser SLAM system

Abstract

The invention provides a position correction method and device applied to an unmanned aerial vehicle laser SLAM system, and relates to the field of data processing. The method comprises the following steps: acquiring initial positioning data and environment data sent by sensor equipment located on an unmanned aerial vehicle; determining a first position of the unmanned aerial vehicle according to the initial positioning data; according to the environment data, environment types are determined, and the environment types comprise a dense building area, an open area and a complex area; based on the environment type, a weight combination of positioning modes for the unmanned aerial vehicle is determined, the positioning modes comprise a laser radar positioning mode, an IMU positioning mode and a visual positioning mode, and one positioning mode corresponds to one weight value; performing position correction on the first position according to the positioning mode and the weight combination to obtain a second position; and positioning the unmanned aerial vehicle based on the second position. By implementing the technical scheme provided by the invention, the position correction accuracy of the SLAM system can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a position correction method and device applied to a UAV laser SLAM system. Background Art

[0002] In drone laser SLAM (Simultaneous Localization and Mapping) systems, using it for position correction is a common positioning method.

[0003] At present, traditional laser SLAM systems usually rely on a single sensor data source and fail to fully consider the different effects of different environments on the sensor positioning accuracy, resulting in the system's lack of flexibility in dealing with dynamic environments. Especially in UAV missions with high positioning requirements, such as urban inspections, disaster area search and rescue, or mountain mapping, the limitations of a single sensor significantly increase the system's error accumulation and reduce the position correction accuracy of the SLAM system.

[0004] Therefore, there is an urgent need for a position correction method and device for use in UAV laser SLAM systems. Summary of the invention

[0005] The present application provides a position correction method and device applied to a UAV laser SLAM system, which is convenient for improving the position correction accuracy of the SLAM system.

[0006] In the first aspect of the present application, a position correction method applied to a drone laser SLAM system is provided, the method comprising: obtaining initial positioning data and environmental data sent by a sensor device located on the drone; determining a first position of the drone based on the initial positioning data; determining an environment type based on the environmental data, the environment type including densely built-up areas, open areas, and complex areas; based on the environment type, determining a weight combination of positioning methods for the drone, the positioning methods including a laser radar positioning method, an IMU positioning method, and a visual positioning method, and one positioning method corresponds to one weight value; performing position correction on the first position according to the positioning method and the weight combination to obtain a second position; and positioning the drone based on the second position.

[0007] By adopting the above technical solution, by identifying the type of environment in which the drone is located, the method enables the system to adjust the weights of different sensors according to the characteristics of the environment to achieve more accurate positioning. The system uses multi-sensor data fusion of three positioning methods, namely, lidar, IMU and vision, which can better adapt to different environments and enhance the stability of positioning. In the case of failure or reduced accuracy of one sensor data, the system can dynamically adjust the data weights of other sensors to reduce the impact of failed data. This method enables the system to automatically adapt to different types of environments without human intervention to set weights, thereby improving the autonomy and task flexibility of the drone. The system will adjust in real time according to the type of environment, without the need to manually set the sensor combination, reducing operating costs. This real-time weight adaptive adjustment enables the drone to quickly switch to the best positioning strategy, adapt to various task requirements, and improve operating efficiency and overall positioning accuracy. When the sensor is interfered or the data quality is reduced, the weight ratio of the sensor can be reduced according to the situation, and the weight of other sensors can be increased to balance the error impact. Traditional SLAM systems are susceptible to error accumulation, especially when the drone is flying for a long time. By adjusting the weight distribution according to environmental characteristics, the system can effectively reduce the error accumulation of a single data source. In the case of accumulated errors, the system will switch to a more accurate sensor combination according to different environmental characteristics, thereby avoiding the cumulative deviation of positioning data. Therefore, it is easy to improve the position correction accuracy of the SLAM system.

[0008] Optionally, determining the environment type according to the environmental data specifically includes: determining multiple feature points of the environment in which the drone is located according to the environmental data; inputting the multiple feature points into a neural network, calculating the feature scores corresponding to the feature points, and obtaining the total feature value by statistics; if it is determined that the total feature value is within a first preset interval, the environment type is determined to be the densely built-up area; if it is determined that the total feature value is within a second preset interval, the environment type is determined to be the open area; if it is determined that the total feature value is within a third preset interval, the environment type is determined to be the complex area, wherein the first preset interval, the second preset interval and the third preset interval are three continuous intervals, and the first preset interval, the second preset interval and the third preset interval are all left-closed and right-open intervals.

[0009] By adopting the above technical solution, by using feature points and neural networks, the system can quickly and accurately identify the type of the environment in which the drone is located, and automatically classify it. Based on multiple feature points to evaluate environmental features, the one-sidedness of single feature judgment is avoided, thereby improving the accuracy of environmental recognition. Dense building areas, open areas and complex areas have different requirements for drone navigation. By accurately identifying the type of environment, the system can better match the navigation requirements of the drone and achieve higher positioning accuracy. The method automatically judges the environment through a neural network, and determines the type of environment in different intervals according to the total value of the feature, so as to achieve dynamic positioning mode adjustment. The system does not need to rely on pre-set scene information, but can automatically adjust according to the changes of real-time environmental feature points. It can adapt autonomously in any flight environment, thereby enhancing the environmental adaptability of the drone. In complex flight missions, drones will encounter switching between multiple scenes. The method can respond quickly when the environment changes, and has strong adaptability, thereby ensuring the robustness and stability of positioning. Using neural networks to analyze feature points and calculate the scores of each feature makes the recognition result more accurate. The neural network can automatically learn environmental features according to the distribution of feature points, without relying on traditional threshold settings or manual rules, reducing manual intervention and human errors. The neural network converts feature points into scores and determines the environment type by the sum of the scores, ensuring the efficiency and accuracy of the evaluation process, which is suitable for the needs of drones to quickly identify the environment. This method uses continuous intervals to divide different environment types, which can reduce misjudgment caused by boundary problems. By dividing the continuous intervals, the drone can still accurately judge the environment type when the total feature value is close to the boundary, avoiding misjudgment caused by discontinuous intervals. If the total value of some features is at the boundary, the flexibility of identification can be adjusted by expanding or narrowing the range of the interval to adapt to actual application needs, making the environment recognition system more adaptable.

[0010] Optionally, determining a weight combination of the positioning method for the drone based on the environment type specifically includes: if the environment type is determined to be the densely built-up area, determining that the weight value corresponding to the lidar positioning method is greater than the weight value corresponding to the IMU positioning method, and greater than the weight value corresponding to the visual positioning method; if the environment type is determined to be the open area, determining that the weight value corresponding to the IMU positioning method is greater than the weight value corresponding to the lidar positioning method, and greater than the weight value corresponding to the visual positioning method; if the environment type is determined to be the complex area, determining that the weight value corresponding to the IMU positioning method is less than the weight value corresponding to the lidar positioning method, and less than the weight value corresponding to the visual positioning method.

[0011] By adopting the above technical solutions, in densely built-up areas, the building features are rich and severely obscured. LiDAR can provide high-precision positioning data, and the weight allocation takes precedence over IMU and visual sensors, which can reduce positioning errors caused by light interference and occlusion. In open areas, feature points are sparse, and LiDAR may drift, while IMU can provide stable inertial data. Therefore, the weight of IMU takes precedence over LiDAR, enhancing the robustness of the system. In complex regional environments such as woods and mountains, the feature points are complex and irregular, and visual sensors can supplement the data of LiDAR. Giving the visual sensor a higher weight than IMU helps the system quickly identify irregular environmental features and improve positioning accuracy. By dynamically adjusting the weight combination of sensors, drones can reduce data interference in different environments and improve the reliability of overall positioning. In densely built-up areas, visual sensors are easily obscured, and IMU is not as accurate as LiDAR in position, so the system gives LiDAR the highest weight, reducing the occlusion effects that may be caused by other sensors. In scenes with insufficient lighting or high contrast, the data of the visual sensor may be inaccurate. After the weight is reduced, the system can rely more stably on the data of the lidar and IMU, reducing the interference of lighting changes on positioning. The dynamically adjusted weight combination can switch the priority of the sensor when the positioning error accumulates, thereby reducing the impact of the error of a single sensor on the system. In open areas, the weight of the lidar is reduced and the weight of the IMU is increased, which can reduce the drift problem of the lidar when there are no feature points and reduce the accumulation of errors. By selecting the optimal sensor combination in different environments, the system can more easily distribute the errors evenly among multiple sensors, thereby reducing the risk of accumulated errors of a single sensor and improving the long-term accuracy of positioning. This method automatically adjusts the sensor weight according to the environment, giving the system a strong autonomy and reducing the need for manual adjustment of sensor parameters. The drone can automatically identify the environment type during cruising and select the appropriate positioning method based on the weight combination to adapt to complex task requirements without human intervention.

[0012] Optionally, the performing position correction on the first position according to the positioning mode and the weight combination to obtain the second position specifically includes: ; in, For the second position, is the position of the drone measured by lidar positioning. is the position of the drone measured by IMU positioning. is the position of the drone measured by visual positioning. is the weight value corresponding to the laser radar positioning method, is the weight value corresponding to the IMU positioning method, is the weight value corresponding to the visual positioning method, is a weighted combination.

[0013] By adopting the above technical solutions, by combining the multi-source information of LiDAR, IMU and visual positioning, the system integrates the positioning advantages of multiple sensors and can more accurately calculate the actual position of the drone. LiDAR is superior in feature-rich areas, but it will drift in feature-sparse environments, while IMU can provide stable inertial data, and visual positioning has advantages in identifying environmental structures. The complementarity of the three helps to reduce the positioning deviation of a single sensor. By integrating multiple positioning methods, the system can obtain more accurate positioning results in various environments, avoid errors caused by the limitations of a single positioning method, and ensure the accuracy of position correction. The formula uses weights to adjust the priorities of different positioning methods to enhance the robustness of positioning. The system automatically adjusts the weight values ​​of LiDAR, IMU and visual positioning according to the type of environment. The weight distribution can be dynamically adjusted according to the real-time data quality of the sensor, improve the stability of the system under environmental changes or sudden interference, reduce error accumulation, and ensure the reliability of positioning. Through the weight combination formula, the system can appropriately reduce the weight of a certain sensor when its data is unreliable, and use the positioning results of other sensors to compensate for insufficient data. In the event of fluctuations or short-term failures in sensor data, by reducing its weight, the system can reduce its reliance on its data and avoid positioning errors caused by single sensor problems. In different mission scenarios, drones can adaptively adjust positioning weights, such as increasing the weight of visual positioning in post-disaster search and rescue missions to enhance the perception of complex environments, and increasing the weight of lidar in areas with dense high-rise buildings to achieve precise navigation. In specific tasks, the system can give priority to the use of the most appropriate sensor data, improve the adaptability of drones under different mission requirements, and ensure the timeliness and accuracy of positioning.

[0014] Optionally, the method further includes: obtaining a take-off position of the UAV; performing mileage correction on an odometer of the UAV according to the take-off position and the second position to obtain target mileage data; and sending the target mileage data to a user device.

[0015] By adopting the above technical solution, the take-off position, as a known initial reference point, can be used as an accurate positioning basis, and the actual flight path of the drone can be corrected by combining the current second position. In the long-distance flight of the drone, relying solely on the data of the odometer may cause position drift and error accumulation. Using the take-off position as a reference, the accumulated error is recalibrated in each correction, which effectively reduces the accumulation of errors. The mileage calculation based on the take-off position and the corrected position can accurately reflect the actual flight path of the drone, ensure that each mileage data is more in line with the actual flight distance, and improve the accuracy of the flight trajectory. In the case of environmental changes or unstable sensor data, the stability of the system can be enhanced by setting a fixed take-off position as a reference. The take-off position is fixed and can maintain consistency in any environment without being affected by external environmental interference. Even in complex terrain or densely built areas, the system can still calculate accurate mileage data based on the distance from the take-off position. This mileage correction method does not rely on a single sensor and can adapt to a variety of geographical environments. Even if the sensor data such as lidar or IMU fluctuates greatly, the correction method based on the take-off position can still provide accurate mileage information. Using the take-off position and the current second position to calculate the target mileage can reduce complex calculation steps and simplify the mileage correction process. The fixed reference of the take-off position simplifies the calculation process of the position data, avoids the frequent analysis of historical data, and enables the system to make mileage corrections in real time. By reducing unnecessary calculations, the system can generate target mileage data more quickly and provide timely feedback to users, improving the overall computing efficiency of the drone system.

[0016] Optionally, the method also includes: obtaining a target map and a historical positioning error distribution from the UAV laser SLAM system, and comparing them with the second position to obtain a comparison result; if it is determined that the comparison result indicates that the UAV positioning error is greater than an error threshold, then correcting the weight combination to generate a target weight configuration; and relocating the UAV according to the target weight configuration.

[0017] By adopting the above technical solution, by comparing the target map with the historical positioning error distribution, the system can continuously monitor and detect the positioning error of the drone during flight. This method can help the system understand in real time whether there is a deviation in the current positioning and promptly discover the situation where the error exceeds the threshold. Once it is detected that the positioning error exceeds the preset error threshold, the system will automatically correct the weight combination. By adjusting the weight, the system can optimize the positioning priority of different sensors, so as to make positioning corrections more accurately. This correction mechanism based on real-time error comparison can effectively avoid the cumulative error caused by a single sensor or environmental changes, so that the drone can maintain a relatively stable positioning in a complex environment. In different flight environments, the positioning error may be different due to different sensor performance. By making corrections based on the historical positioning error distribution, the system can adaptively adjust the weight configuration according to the needs of different environments. By comparing the error and the target map, the system can automatically adjust the weight combination of each sensor according to the real-time data, eliminating manual intervention. Such automated adjustment reduces the risk of human operation errors and improves the intelligence level of the drone system. The process of dynamically adjusting the weight combination will be optimized based on the real-time comparison results, so that the system can make the most appropriate response in the positioning tasks at different stages. This method makes the positioning algorithm more flexible and adaptable to different flight mission requirements. By comparing the target map with the historical positioning error distribution, the system can accurately identify the source of the error and compensate for these errors by adjusting the sensor weights, thereby improving positioning accuracy. When the drone is performing long-term flight missions, traditional positioning methods may cause errors to gradually accumulate due to sensor drift or environmental factors. By performing real-time error comparison and weight correction during flight, the system can continuously correct errors and avoid the reduction in positioning accuracy caused by long-term flight. For missions that require long-term continuous flight, this correction method can ensure that the drone always maintains high-precision positioning throughout the flight and can accurately perform tasks according to the predetermined trajectory.

[0018] Optionally, the sensor equipment includes a lidar, an IMU and a visual camera.

[0019] By adopting the above technical solutions, the use of multiple sensor systems can improve the robustness of the system, because different sensors are suitable for different environmental conditions. For example, LiDAR is insensitive to environmental occlusion, IMU performs well in dynamic situations, and visual cameras perform well in environments with rich features. Through multi-sensor fusion, the system can overcome the limitations of a single sensor and improve its adaptability in various environments. If a sensor fails or the data quality deteriorates, other sensors can still provide supplementary data to maintain the stability of positioning. By complementing each other between sensors, the positioning error caused by sensor failure or abnormality can be effectively reduced. By combining the data of LiDAR, IMU and visual cameras, the system can generate high-precision positioning data in real time, which can not only improve the positioning accuracy of drones in complex environments, but also better solve the positioning problems indoors and in areas without GPS signals.

[0020] In a second aspect, the present application provides a position correction device applied to a drone laser SLAM system, the position correction device comprising an acquisition module and a processing module, wherein the acquisition module is used to acquire initial positioning data and environmental data sent by a sensor device located on the drone; the processing module is used to determine the first position of the drone based on the initial positioning data; the processing module is also used to determine the environment type based on the environmental data, and the environment type includes densely built-up areas, open areas, and complex areas; the processing module is also used to determine a weight combination of positioning methods for the drone based on the environment type, the positioning methods include laser radar positioning methods, IMU positioning methods, and visual positioning methods, and one positioning method corresponds to one weight value; the processing module is also used to perform position correction on the first position according to the positioning method and the weight combination to obtain a second position; the processing module is also used to position the drone based on the second position.

[0021] In the third aspect of the present application, an electronic device is provided, which includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device performs the method described above.

[0022] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, the method described above is executed.

[0023] In summary, one or more technical solutions provided in this application have at least the following technical effects or advantages: By identifying the type of environment the drone is in, this method enables the system to adjust the weights of different sensors according to the characteristics of the environment to achieve more accurate positioning. The system uses multi-sensor data fusion of three positioning methods: lidar, IMU and vision, which can better adapt to different environments and enhance the stability of positioning. In the case of failure or reduced accuracy of one sensor data, the system can dynamically adjust the data weights of other sensors to reduce the impact of failed data. This method enables the system to automatically adapt to different types of environments without human intervention to set weights, thereby improving the autonomy and mission flexibility of the drone. The system will adjust in real time according to the type of environment, without the need to manually set the sensor combination, reducing operating costs. This real-time weight adaptive adjustment enables the drone to quickly switch to the best positioning strategy, adapt to various mission requirements, and improve operational efficiency and overall positioning accuracy. When a sensor is interfered with or the data quality is reduced, the weight ratio of the sensor can be reduced according to the situation, and the weight of other sensors can be increased to balance the impact of errors. Traditional SLAM systems are susceptible to error accumulation, especially when drones are flying for a long time. By adjusting the weight distribution according to environmental characteristics, the system can effectively reduce the error accumulation of a single data source. In the case of accumulated errors, the system will switch to a more accurate sensor combination according to different environmental characteristics, thereby avoiding the cumulative deviation of positioning data. Therefore, it is easy to improve the position correction accuracy of the SLAM system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flow chart of a position correction method applied to a UAV laser SLAM system provided in an embodiment of the present application.

[0025] Figure 2 Another flow chart of a position correction method applied to a UAV laser SLAM system provided in an embodiment of the present application.

[0026] Figure 3 A schematic diagram of a module of a position correction device applied to a UAV laser SLAM system provided in an embodiment of the present application.

[0027] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0028] Explanation of the reference numerals: 31, acquisition module; 32, processing module; 41, processor; 42, communication bus; 43, user interface; 44, network interface; 45, memory. DETAILED DESCRIPTION

[0029] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0030] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0031] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0032] In the UAV laser SLAM (Simultaneous Localization and Mapping) system, position correction is an important step to improve positioning accuracy and ensure the smooth progress of the mission. The laser SLAM system scans the environment through laser radar and uses the SLAM algorithm for real-time positioning and map construction. In recent years, with the widespread application of drones in many fields, especially for tasks with high positioning accuracy requirements such as urban inspections, disaster area search and rescue, and mountain mapping, the reliability and accuracy of the laser SLAM system have become particularly important.

[0033] However, traditional laser SLAM systems usually rely on a single sensor data source, mainly the measurement data of LiDAR. Although LiDAR can provide relatively accurate positioning data in open or relatively simple environments, the use of this single sensor is insufficient to meet the challenges in complex or dynamic environments. In particular, the differences in sensor positioning accuracy in different environments have not been fully considered. For example, in urban areas, the measurement effect of LiDAR may be disturbed due to the obstruction of high-rise buildings, resulting in reduced positioning accuracy. In low-light environments or environments with fewer visual features, visual sensors may not be able to effectively provide useful positioning information. Therefore, the limitations of a single sensor significantly increase the error accumulation of the system and reduce the position correction accuracy of the SLAM system.

[0034] In order to solve the above technical problems, the present application provides a position correction method for a UAV laser SLAM system, referring to Figure 1 , Figure 1 The present invention provides a flow chart of a method for position correction applied to a UAV laser SLAM system. The method for position correction is applied to a UAV controller, and includes steps S110 to S160, which are as follows: S110: Acquire initial positioning data and environmental data sent by a sensor device located on the drone.

[0035] Specifically, the drone controller is one of the core components located on the drone, responsible for processing data from various sensors and controlling the flight of the drone based on this data. The controller's tasks include positioning, navigation, flight stability, obstacle avoidance, etc. It is the "brain" of most of the drone's automated operations. Sensor devices are tools for drones to collect environmental data and their own status. Sensors include: LiDAR: used to measure the distance between the drone and surrounding objects and obtain accurate three-dimensional environmental information. IMU: composed of accelerometers and gyroscopes, used to measure the acceleration, angular velocity, attitude changes and other dynamic information of the drone. Visual camera: captures the characteristics of the surrounding environment through images, helps with visual positioning (for example, SLAM technology), and detects obstacles and targets. GPS: provides the geographic location and altitude data of the drone to help it determine its current spatial position. Initial positioning data refers to the first batch of data obtained by the drone from sensor devices (such as GPS, IMU, etc.) to determine the starting position and status of the drone. This data set usually contains information about the drone's position (such as latitude and longitude, altitude), attitude (such as heading, pitch angle and roll angle), and motion status. These data are the basis for subsequent flight control and path planning. Environmental data refers to the surrounding environment information obtained by the drone through sensors (such as lidar and visual cameras). This data can include: Obstacle location: whether there are buildings, trees and other obstacles around the drone that may affect flight. Terrain characteristics: whether the ground is flat, whether there is a slope, etc. Lighting conditions: such as light intensity and whether the environment is suitable for the visual sensor to work.

[0036] For example, suppose a drone is performing a city inspection mission. To perform this mission, the drone controller needs to obtain initial positioning data and environmental data at takeoff to determine the drone's current location and surrounding environment. The specific process is as follows: GPS provides the drone's longitude, latitude and altitude information. Assume that the drone's longitude is -73.9857, latitude is 40.7484, and altitude is 50 meters when it just takes off. IMU provides the drone's attitude information, such as the drone's current heading is 90° (eastward), the pitch angle is 0°, and the roll angle is 0°. Lidar measures the distance between the drone and surrounding objects. Assume that there are no obstacles within 10 meters in front of the drone. Lidar scans and returns three-dimensional environmental data around the drone, which may show nearby buildings, streets, and other objects. This data helps the drone avoid collisions in subsequent flights. The visual camera captures images to identify surrounding features, such as buildings, landmarks, etc., to help the drone locate and navigate. The IMU records the drone's acceleration and angular velocity in real time to detect whether the drone maintains stable flight or vibrates violently. Based on this initial data, the controller can determine the current location of the drone, whether there are obstacles around it, and the flight status. Based on this information, the drone controller can make preliminary flight control decisions, execute missions, plan inspection paths, avoid obstacles, etc.

[0037] In one possible implementation, the sensor device includes a lidar, an IMU, and a visual camera.

[0038] Specifically, LiDAR is a device that measures the distance between an object and a sensor by emitting a laser beam and receiving the reflected laser signal. It scans the environment and quickly generates three-dimensional point cloud data of the surrounding environment, helping drones identify obstacles, terrain, and other features. LiDAR can provide high-precision distance data and is particularly suitable for building three-dimensional environmental models. LiDAR is still effective when working in low-light environments or in the dark because it does not rely on light or visible light. However, the scanning range of LiDAR may be affected by environmental conditions (such as rain, fog, and other weather). For example: When drones conduct urban inspections, LiDAR can effectively scan and identify the locations of objects such as buildings, roads, and trees, helping drones avoid collisions and build environmental maps.

[0039] An IMU is a device that measures the acceleration, angular velocity, and direction of a drone to estimate its motion state. It usually includes an accelerometer, gyroscope, and magnetometer, which can provide real-time information on the drone's position change, angle change, and speed change. An IMU can provide high-frequency motion data and can quickly track the drone's motion state in a short period of time. It does not rely on external signals, so it can still work effectively when GPS is unavailable or the signal is weak, for example, in indoor or underground environments. However, IMU data tends to drift over time, which may lead to error accumulation when used for a long time. Its accuracy is greatly affected by sensor quality and external interference (such as vibration). For example: In mountain surveying, an IMU can help a drone keep track of its own motion state when GPS signals cannot be received, thereby maintaining a relatively stable flight attitude.

[0040] Vision cameras capture images or videos of the environment to extract feature points, helping drones identify the surrounding environment and locate themselves. It is usually combined with computer vision algorithms to analyze features in images (such as objects, textures, edges, etc.) for environmental perception and position estimation. Vision sensors are relatively cheap, lightweight, and can accurately provide environmental details through image recognition. In a well-lit environment, the information provided by vision sensors is very rich and can effectively assist in positioning and mapping. However, vision sensors are greatly affected by lighting, weather (such as haze or rain and snow), and viewing angle limitations, and image quality is easily affected. It has high requirements for computing power and requires a lot of image processing and calculation. For example: In disaster area search and rescue, vision cameras can identify features such as collapsed buildings and road cracks, helping drones determine flight paths, avoid collisions, and provide useful real-time information.

[0041] S120: Determine a first position of the UAV according to the initial positioning data.

[0042] Specifically, the first position refers to the specific position of the drone after receiving the initial positioning data during the flight. When the drone just takes off, the controller will combine the data from different sensors through data fusion technology to calculate the position of the drone. For example, the controller will derive an accurate preliminary position based on the latitude and longitude data provided by the GPS, the attitude information of the IMU, and possible environmental data. This position is called the "first position". The drone controller calculates the first position of the drone by combining the initial data provided by various sensors. This position is crucial for the subsequent flight, path planning, positioning correction and other operations of the drone. It provides an accurate starting point for the entire flight mission, enabling the drone to accurately locate and effectively avoid obstacles in subsequent flights.

[0043] S130. Determine the environment type according to the environment data, where the environment type includes a densely built-up area, an open area, and a complex area.

[0044] Specifically, the environment type is a classification of the environment in which the drone is located. Different environment types have different effects on the flight control and positioning systems, so the positioning strategies and flight behaviors adopted for each environment type will be different. Densely built areas refer to areas in cities with dense high-rise buildings. In such areas, the lidar signal may be blocked by buildings, and the signal reflection may be uneven, which reduces the positioning accuracy of the lidar. The visual sensor may be greatly affected in an environment with tall buildings. Open areas refer to areas with no obstacles or few buildings, such as grasslands, open spaces, or open fields. In such areas, the signal propagation of the lidar is relatively smooth, and relatively accurate distance data can be obtained, with high positioning accuracy. At the same time, the visual sensor and IMU can also work effectively. Complex areas refer to some environments with complex terrain and dynamic changes. For example, mountainous areas, forests, or disaster areas, these places may have complex terrain and various obstacles (such as trees, mountains, snow, etc.). This environment is highly dependent on sensors, which may cause unstable sensor data and require complex positioning and flight control algorithms to ensure accuracy.

[0045] In a possible implementation, the environment type is determined according to the environmental data, specifically including: determining multiple feature points of the environment in which the drone is located according to the environmental data; inputting the multiple feature points into a neural network, calculating the feature scores corresponding to each feature point, and obtaining the total feature value; if the total feature value is determined to be within a first preset interval, the environment type is determined to be a densely built-up area; if the total feature value is determined to be within a second preset interval, the environment type is determined to be an open area; if the total feature value is determined to be within a third preset interval, the environment type is determined to be a complex area, wherein the first preset interval, the second preset interval, and the third preset interval are three continuous intervals, and the first preset interval, the second preset interval, and the third preset interval are all left-closed and right-open intervals.

[0046] Specifically, in environmental data, feature points are representative key points that can be used to reflect the geometric features of the environment. For example, the point cloud data returned by the lidar sensor may contain the edges of walls, buildings, or obstacles. These point clouds can be used as feature points to reflect the structure of the environment. The drone controller inputs these feature points into the neural network, and the neural network analyzes each feature point based on the trained model and calculates a feature score. These feature scores reflect certain properties of the feature points, such as the density, shape, and distribution of the object. During training, the neural network model has learned how to judge the type of environment based on different environmental features. For example, in a densely built area, feature points may be more concentrated and densely distributed, while in an open area, the distribution of feature points may be more dispersed. The sum of the feature scores corresponding to all feature points is calculated to obtain the total feature value. This total value reflects the characteristics of the entire environment, which can help determine which type the environment belongs to. Based on the total feature value, it is compared with the preset interval to determine the type of environment.

[0047] Among them, the set interval is a standard for dividing different types of environments: if the total value of the feature falls within this interval, the environment where the drone is located is considered to be a "dense building area". If the total value of the feature falls within this interval, the environment is considered to be an "open area". If the total value of the feature falls within this interval, the environment is considered to be a "complex area". These preset intervals are continuous, but do not overlap, and are left-closed and right-open intervals, which means that the lower bound of the interval is included, but the upper bound is not included. For example, the first preset interval may be [0, 50), the second preset interval may be [50, 100), and the third preset interval may be [100, 150). Through the above steps, the system can accurately determine the type of environment in which the drone is located through environmental data and neural network calculations. Different environment types correspond to different flight strategies and positioning methods. In open areas, the positioning effect of lidar may be the best; in densely built areas, it may be necessary to rely more on sensors such as vision and IMU; and in complex areas, it may be necessary to combine multiple sensors to improve positioning accuracy. In this way, the system can dynamically adjust the positioning strategy according to different environments, thereby improving the success rate and accuracy of flight missions.

[0048] S140. Based on the environment type, determine a weight combination of positioning methods for the UAV, where the positioning methods include a laser radar positioning method, an IMU positioning method, and a visual positioning method, and one positioning method corresponds to one weight value.

[0049] Specifically, the positioning method refers to the way the drone is positioned through different sensors. The distance information of the surrounding environment is obtained through the lidar, and the distance between the drone and the surrounding objects is accurately measured. Lidar performs better in open areas or places with clear line of sight. The IMU, combined with sensors such as accelerometers and gyroscopes, measures the acceleration and angular velocity of the drone to estimate the position and attitude. The advantage of IMU is that it can continue to provide estimates when the signal is lost, but errors may accumulate over time. The visual sensor captures the surrounding image information and identifies the position of objects such as the ground and buildings for positioning. Visual positioning is sensitive to conditions such as lighting and texture, and is suitable for areas with rich visual features. The weight value is a measure of the effectiveness and reliability of different positioning methods in a specific environment. For example, in some environments, lidar positioning may be the most accurate, while in other environments, IMU or visual positioning may be more reliable. The system dynamically adjusts the type of sensor it relies on by assigning different weight values ​​to each positioning method.

[0050] In one possible implementation, based on the environment type, a weight combination of positioning methods for the drone is determined, specifically including: if the environment type is determined to be a densely built-up area, then the weight value corresponding to the lidar positioning method is determined to be greater than the weight value corresponding to the IMU positioning method, and greater than the weight value corresponding to the visual positioning method; if the environment type is determined to be an open area, then the weight value corresponding to the IMU positioning method is determined to be greater than the weight value corresponding to the lidar positioning method, and greater than the weight value corresponding to the visual positioning method; if the environment type is determined to be a complex area, then the weight value corresponding to the IMU positioning method is determined to be less than the weight value corresponding to the lidar positioning method, and less than the weight value corresponding to the visual positioning method.

[0051] Specifically, in densely built-up areas, buildings are dense and there are many obstacles, which may affect the signal reflection of the lidar, resulting in large positioning errors. In this case, the lidar can be positioned relatively more effectively, and the contribution of the IMU and visual sensors is relatively small. In open areas, the area is open and there are not many obstructions, so the lidar signal can be well reflected. Therefore, the lidar positioning in this environment will show higher accuracy. At the same time, the IMU can provide stable attitude estimation. Complex areas contain various terrains, obstacles or very irregular environments. The lidar signal may not be able to obtain clear reflections. The visual sensor may also be affected by factors such as lighting and weather. The IMU can provide a relatively stable position correction to a certain extent. In densely built-up areas, due to building occlusion and complex environment, the accuracy of the lidar will be higher, so the weight value of the lidar is larger, while the weight of the IMU and vision is lower. In open areas, due to the open environment, the lidar positioning can obtain more accurate reflection signals, and the IMU can provide attitude estimation. Therefore, the weight value of the IMU will be higher than that of the lidar, and the role of vision is relatively small. In complex areas, the weights of lidar and vision will increase relatively, because the error of IMU may be more serious in complex terrain, so its weight value will be smaller.

[0052] S150: Correct the first position according to the positioning method and the weight combination to obtain a second position.

[0053] Specifically, the first position refers to the position of the drone that is initially calculated based on sensor data, which may not be accurate enough due to sensor errors or environmental influences. The second position is the new position obtained after position correction. The correction process is based on the measurement results of each sensor and the weighted average of the data of each sensor according to the weight combination. In this way, the second position obtained will be more accurate than the first position because it combines the data of multiple sensors and optimizes the position estimate through weights.

[0054] In a possible implementation, the first position is corrected according to the positioning mode and the weight combination to obtain the second position, which specifically includes: ; in, For the second position, is the position of the drone measured by lidar positioning. is the position of the drone measured by IMU positioning. is the position of the drone measured by visual positioning. is the weight value corresponding to the laser radar positioning method, is the weight value corresponding to the IMU positioning method, is the weight value corresponding to the visual positioning method, is a weighted combination.

[0055] Specifically, by weighted averaging the measurement results of different sensors and combining the influence of the environment type, the position estimation of the drone is made more accurate. By adjusting the weights of different sensors, the position correction can be optimized in different environments, thereby reducing errors and improving positioning accuracy. This method can effectively deal with the performance differences of different sensors in different environments, enhancing the reliability and flexibility of the system.

[0056] S160: Positioning the UAV based on the second position.

[0057] Specifically, the position of the drone is measured by lidar, IMU, and visual cameras. The initial position is corrected according to the environment type and the weight combination of different sensors to obtain a more accurate position. After obtaining the second position, the control system uses this precise position as a reference to guide the flight, navigation, and mission execution of the drone. Positioning based on the second position means that the drone's subsequent flight decisions, navigation paths, and mission execution will all rely on the corrected and more accurate position data. This is to ensure that the drone can maintain stable and accurate flight in dynamic or complex environments.

[0058] In a possible implementation, the take-off position of the UAV is obtained; the mileage of the UAV's odometer is corrected according to the take-off position and the second position to obtain target mileage data; and the target mileage data is sent to the user device.

[0059] Specifically, when the drone takes off, it records its take-off position, which is the coordinate when the drone is first started. The take-off position is regarded as the reference point for the flight. During the flight, after multiple sensor data processing and position corrections, the drone will get a more accurate "second position". This position can be regarded as a corrected position of the take-off position, eliminating the error. Based on the take-off position and the second position, the control system can calculate the actual mileage of the drone's flight. The mileage correction is to ensure that the statistical data of the flight mileage is more accurate under the influence of multiple sensor measurement errors or environmental factors. For example, if the coordinate difference between the take-off position and the second position is large, such as an offset during the flight, the control system will correct the original mileage data based on this difference to make it more consistent with the actual flight path. Through the above corrections, the target mileage data is calculated, which represents the accurate flight distance of the drone from the take-off position to the corrected second position. Finally, the corrected mileage data will be sent to the user device, that is, the mobile phone, computer or monitoring system used by the drone operator. The user device can obtain the actual flight distance of the drone through this data for mission planning, flight log recording or further analysis.

[0060] In one possible implementation, refer to Figure 2 , Figure 2 Another flow chart of a position correction method applied to a UAV laser SLAM system provided in an embodiment of the present application includes steps S210 to S230, and the above steps are as follows: S210, obtaining a target map and a historical positioning error distribution from the UAV laser SLAM system, and comparing them with a second position to obtain a comparison result; S220, if it is determined that the comparison result indicates that the UAV positioning error is greater than the error threshold, then correct the weight combination and generate a target weight configuration; S230, reposition the UAV according to the target weight configuration.

[0061] Specifically, the laser SLAM system of the drone generates a target map, which is a real-time map of the current environment, and a historical positioning error distribution, which is the error pattern of the drone's positioning in history. The historical positioning error distribution records the positioning errors of the drone in different environments or conditions. These data can help evaluate the current positioning accuracy and possible sources of error. The second position is the current positioning of the drone obtained through position correction. By comparing the second position with the target map and the historical positioning error distribution, the system can analyze the accuracy of the current positioning. For example, if the current second position is far from the position on the target map, or the error exceeds the normal range, that is, the error threshold, the system will conclude that "the positioning error is large". If the comparison result shows that the current positioning error of the drone is greater than the set error threshold, that is, the positioning accuracy is not enough, the system will correct the weights of the lidar, IMU, and visual sensors in positioning. Weight correction means that the system will adjust the degree of influence of different sensors in the positioning process according to their performance and reliability. For example, the weight of the lidar may be increased to improve positioning accuracy, or the weight of the visual sensor may be reduced because its accuracy is poor under certain conditions. The corrected weight combination, that is, the target weight configuration, reflects how to balance the influence of each sensor in different environments and conditions. These weight configurations will be used for subsequent positioning corrections. Once the weight combination is corrected, the system will reposition the drone according to the new weight configuration. In this way, the positioning accuracy of the drone will be improved, especially in complex environments or when the error is large.

[0062] This application also provides a position correction device for a UAV laser SLAM system, referring to Figure 3 , Figure 3A schematic diagram of a module of a position correction device applied to a UAV laser SLAM system provided in an embodiment of the present application. The position correction device is a UAV controller, which includes an acquisition module 31 and a processing module 32, wherein the acquisition module 31 acquires initial positioning data and environmental data sent by a sensor device located on the UAV; the processing module 32 determines the first position of the UAV according to the initial positioning data; the processing module 32 determines the environment type according to the environmental data, and the environment type includes densely built areas, open areas, and complex areas; the processing module 32 determines the weight combination of the positioning method for the UAV based on the environment type, and the positioning method includes a laser radar positioning method, an IMU positioning method, and a visual positioning method, and one positioning method corresponds to one weight value; the processing module 32 corrects the first position according to the positioning method and the weight combination to obtain a second position; the processing module 32 locates the UAV based on the second position.

[0063] In a possible implementation, the environment type is determined according to the environmental data, specifically including: the processing module 32 determines multiple feature points of the environment in which the drone is located according to the environmental data; the processing module 32 inputs the multiple feature points into the neural network, calculates the feature scores corresponding to the respective feature points, and obtains the total feature value by statistics; if the processing module 32 determines that the total feature value is within the first preset interval, the environment type is determined to be a densely built-up area; if the processing module 32 determines that the total feature value is within the second preset interval, the environment type is determined to be an open area; if the processing module 32 determines that the total feature value is within the third preset interval, the environment type is determined to be a complex area, wherein the first preset interval, the second preset interval and the third preset interval are three continuous intervals, and the first preset interval, the second preset interval and the third preset interval are all left-closed and right-open intervals.

[0064] In one possible implementation, the processing module 32 determines a weight combination of positioning methods for the drone based on the environment type, specifically including: if the processing module 32 determines that the environment type is a densely built-up area, then the weight value corresponding to the lidar positioning method is determined to be greater than the weight value corresponding to the IMU positioning method, and greater than the weight value corresponding to the visual positioning method; if the processing module 32 determines that the environment type is an open area, then the weight value corresponding to the IMU positioning method is determined to be greater than the weight value corresponding to the lidar positioning method, and greater than the weight value corresponding to the visual positioning method; if the processing module 32 determines that the environment type is a complex area, then the weight value corresponding to the IMU positioning method is determined to be less than the weight value corresponding to the lidar positioning method, and less than the weight value corresponding to the visual positioning method.

[0065] In a possible implementation manner, the processing module 32 performs position correction on the first position according to the positioning mode and the weight combination to obtain the second position, specifically including: ; in, For the second position, is the position of the drone measured by lidar positioning. is the position of the drone measured by IMU positioning. is the position of the drone measured by visual positioning. is the weight value corresponding to the laser radar positioning method, is the weight value corresponding to the IMU positioning method, is the weight value corresponding to the visual positioning method, is a weighted combination.

[0066] In a possible implementation, the acquisition module 31 acquires the take-off position of the UAV; the processing module 32 performs mileage correction on the odometer of the UAV according to the take-off position and the second position to obtain target mileage data; the processing module 32 sends the target mileage data to the user device.

[0067] In one possible implementation, the processing module 32 obtains the target map and historical positioning error distribution from the UAV laser SLAM system, compares them with the second position, and obtains a comparison result; if the processing module 32 determines that the comparison result indicates that the UAV positioning error is greater than the error threshold, the weight combination is corrected to generate a target weight configuration; the processing module 32 repositions the UAV according to the target weight configuration.

[0068] In one possible implementation, the sensor device includes a lidar, an IMU, and a visual camera.

[0069] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0070] The present application also provides an electronic device, referring to Figure 4 , Figure 4 The electronic device may include: at least one processor 41 , at least one network interface 44 , a user interface 43 , a memory 45 , and at least one communication bus 42 .

[0071] The communication bus 42 is used to realize the connection and communication between these components.

[0072] The user interface 43 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 43 may also include a standard wired interface and a wireless interface.

[0073] The network interface 44 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0074] Among them, the processor 41 may include one or more processing cores. The processor 41 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 45, and calling data stored in the memory 45. Optionally, the processor 41 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 41 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 41, and it can be implemented separately through a chip.

[0075] Among them, the memory 45 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 45 includes a non-transitory computer-readable storage medium. The memory 45 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 45 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 45 may also be optionally at least one storage device located away from the aforementioned processor 41. As Figure 4As shown, the memory 45 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program for a position correction method applied to a UAV laser SLAM system.

[0076] exist Figure 4 In the electronic device shown, the user interface 43 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 41 can be used to call an application program stored in the memory 45 for a position correction method applied to a drone laser SLAM system. When executed by one or more processors, the electronic device executes one or more methods in the above-mentioned embodiments.

[0077] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0078] The present application also provides a computer-readable storage medium, which stores instructions. When executed by one or more processors, the electronic device executes one or more of the methods described in the above embodiments.

[0079] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0080] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0081] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0082] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0083] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0084] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, it will be easy for those skilled in the art to think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A position correction method applied to a UAV laser SLAM system, characterized in that: The method comprises: Obtaining initial positioning data and environmental data sent by sensor equipment located on the drone; Determining a first position of the drone based on the initial positioning data; Determining an environment type according to the environment data, wherein the environment type includes a densely built area, an open area, and a complex area; Based on the environment type, determine a weight combination of positioning methods for the drone, where the positioning methods include a laser radar positioning method, an IMU positioning method, and a visual positioning method, and one positioning method corresponds to one weight value; According to the positioning method and the weight combination, the first position is corrected to obtain a second position; Based on the second position, the drone is positioned.

2. The position correction method applied to the unmanned aerial vehicle laser SLAM system according to claim 1 is characterized in that: Determining the environment type according to the environment data specifically includes: Determine, based on the environmental data, a plurality of characteristic points of the environment in which the drone is located; Inputting the plurality of feature points into the neural network, calculating the feature scores corresponding to the respective feature points, and obtaining the total feature value by statistics; If it is determined that the total characteristic value is within the first preset interval, the environment type is determined to be the densely built-up area; If it is determined that the total characteristic value is within the second preset interval, then determining that the environment type is the open area; If it is determined that the total feature value is within a third preset interval, the environment type is determined to be the complex area, wherein the first preset interval, the second preset interval and the third preset interval are three continuous intervals, and the first preset interval, the second preset interval and the third preset interval are all left-closed and right-open intervals.

3. The position correction method applied to the unmanned aerial vehicle laser SLAM system according to claim 1, characterized in that: The determining, based on the environment type, a weight combination of the positioning methods for the drone specifically includes: If it is determined that the environment type is the densely built-up area, determining that the weight value corresponding to the laser radar positioning method is greater than the weight value corresponding to the IMU positioning method, and greater than the weight value corresponding to the visual positioning method; If it is determined that the environment type is the open area, determine that the weight value corresponding to the IMU positioning method is greater than the weight value corresponding to the laser radar positioning method, and greater than the weight value corresponding to the visual positioning method; If it is determined that the environment type is the complex area, it is determined that the weight value corresponding to the IMU positioning method is smaller than the weight value corresponding to the lidar positioning method, and smaller than the weight value corresponding to the visual positioning method.

4. The position correction method applied to the unmanned aerial vehicle laser SLAM system according to claim 1, characterized in that: The performing position correction on the first position according to the positioning mode and the weight combination to obtain the second position specifically includes: ； in, For the second position, is the position of the drone measured by lidar positioning. is the position of the drone measured by IMU positioning. is the position of the drone measured by visual positioning. is the weight value corresponding to the laser radar positioning method, is the weight value corresponding to the IMU positioning method, is the weight value corresponding to the visual positioning method, is a weighted combination.

5. The position correction method applied to the unmanned aerial vehicle laser SLAM system according to claim 1, characterized in that: The method further comprises: Obtaining the take-off position of the UAV; According to the take-off position and the second position, the mileage of the odometer of the UAV is corrected to obtain target mileage data; The target mileage data is sent to a user device.

6. The position correction method for the unmanned aerial vehicle laser SLAM system according to claim 1, characterized in that: The method further comprises: Obtain a target map and a historical positioning error distribution from the UAV laser SLAM system, and compare them with the second position to obtain a comparison result; If it is determined that the comparison result indicates that the positioning error of the UAV is greater than the error threshold, the weight combination is corrected to generate a target weight configuration; The UAV is repositioned according to the target weight configuration.

7. The position correction method applied to the unmanned aerial vehicle laser SLAM system according to claim 1, characterized in that: The sensor equipment includes a lidar, an IMU and a visual camera.

8. A position correction device applied to a UAV laser SLAM system, characterized in that: The position correction device comprises an acquisition module (31) and a processing module (32), wherein: The acquisition module (31) is used to acquire initial positioning data and environmental data sent by a sensor device located on the drone; The processing module (32) is used to determine the first position of the drone based on the initial positioning data; The processing module (32) is further used to determine the environment type according to the environmental data, wherein the environment type includes a densely built area, an open area, and a complex area; The processing module (32) is further used to determine a weight combination of positioning methods for the drone based on the environment type, wherein the positioning methods include a laser radar positioning method, an IMU positioning method, and a visual positioning method, and each positioning method corresponds to one weight value; The processing module (32) is further used to correct the first position according to the positioning method and the weight combination to obtain a second position; The processing module (32) is further used to locate the drone based on the second position.

9. An electronic device, characterized in that: The electronic device comprises a processor (41), a memory (45), a user interface (43) and a network interface (44), wherein the memory (45) is used to store instructions, the user interface (43) and the network interface (44) are both used to communicate with other devices, and the processor (41) is used to execute the instructions stored in the memory (45) so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is performed.

Images