A surveying and mapping unmanned aerial vehicle and surveying and mapping method for reducing surveying and mapping errors

Abstract

The present invention relates to the technical field of unmanned aerial vehicle surveying and mapping, and in particular, to a surveying and mapping unmanned aerial vehicle and a surveying and mapping method for reducing surveying and mapping errors. The present invention utilizes a radar device and a camera of an unmanned aerial vehicle to monitor the surrounding environment in real time, determines the position of obstacles through data fusion technology and image recognition algorithm, establishes an obstacle avoidance decision model according to a target surveying and mapping task, introduces a simulated annealing algorithm to optimize the model and formulate an obstacle avoidance strategy, and finally feeds back the obstacle avoidance strategy to a control center, which performs route planning and controls the attitude adjustment speed of the unmanned aerial vehicle in real time to obtain a final route. The present invention can more accurately detect obstacles, formulate reasonable obstacle avoidance strategies, improve surveying and mapping accuracy, reduce surveying and mapping errors, and ensure the flight safety of the unmanned aerial vehicle and the accuracy of surveying and mapping data through the fusion of multiple sensors and optimization algorithms.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle surveying and mapping, and in particular to a surveying and mapping unmanned aerial vehicle and a surveying and mapping method for reducing surveying and mapping errors. Background Art

[0002] UAV surveying and mapping is an important technology. During the UAV surveying and mapping process, the interference of obstacles is one of the important factors leading to errors. Traditional surveying and mapping methods are difficult to accurately detect and identify obstacles around the UAV, especially in complex environments. This may cause the UAV to collide with obstacles, thus affecting the accuracy of surveying and mapping. In addition, the imperfection of obstacle avoidance decision models and strategies will also increase surveying and mapping errors. If the obstacle avoidance path length and the minimum distance to the obstacle cannot be reasonably balanced, the UAV may take unnecessary detours or risk approaching obstacles, thereby affecting the accuracy of surveying and mapping. In order to solve this technical problem, we provide a surveying and mapping UAV and surveying and mapping method that can reduce surveying and mapping errors. Summary of the invention

[0003] The purpose of the present invention is to provide a surveying and mapping UAV and a surveying and mapping method for reducing surveying and mapping errors, so as to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, one of the objects of the present invention is to provide a surveying and mapping UAV and a surveying and mapping method for reducing surveying and mapping errors, comprising the following steps:

[0005] S1. Use the radar device and camera of the target UAV to monitor the surrounding environment of the target UAV in real time, obtain the suspected obstacle image, and calculate the distance of the suspected obstacle through data fusion technology. Then substitute the distance of the suspected obstacle into the image recognition algorithm and judge whether there is an obstacle in the suspected obstacle image. If there is, determine the location coordinates of the obstacle;

[0006] S2. Establish an obstacle avoidance decision model based on the target mapping task and obstacle position coordinates, and introduce a simulated annealing algorithm to optimize the obstacle avoidance decision model and formulate an obstacle avoidance strategy;

[0007] S3. Feedback the obstacle avoidance strategy to the control center of the target UAV. The control center of the target UAV plans the route of the target UAV according to the obstacle avoidance strategy, and controls the attitude adjustment speed of the target UAV in real time to obtain the final route of the target UAV.

[0008] As a further improvement of the technical solution, the method of using the radar device and camera of the target drone to monitor the surrounding environment of the target drone in real time in S1 to obtain a real-time image of the surrounding environment of the target drone specifically includes:

[0009] The radar device of the target UAV is a laser radar and a millimeter-wave radar, and the camera is a binocular vision camera. The laser radar is installed at the bottom center of the target UAV to detect suspected obstacles in front of and below the target UAV. The millimeter-wave radar is installed on both sides of the target UAV to monitor suspected obstacles on the sides of the target UAV. The binocular vision camera is installed in front of the target UAV to provide images of suspected obstacles. The laser radar emits a laser beam to the surroundings at a fixed frequency, receives the reflected signal, and records the distance of the suspected obstacle. The millimeter-wave radar continuously scans the surrounding environment to obtain the distance of the suspected obstacle. The binocular vision camera continuously shoots the picture in front of the target UAV and transmits the image of the suspected obstacle to the processing system of the target UAV.

[0010] As a further improvement of the technical solution, the method of calculating the distance of the suspected obstacle by using the data fusion technology in S1 specifically includes:

[0011] The time when the laser radar and millimeter-wave radar detect suspected obstacles is recorded, and the distance data of the laser radar and millimeter-wave radar detecting suspected obstacles at the same time are counted, which are recorded as and ,Will and Substitute them into the formula to calculate the average distance between the two, and use the average distance between the two as the suspected obstacle distance for preliminary judgment. The suspected obstacle distance is the straight-line distance from the target UAV to the suspected obstacle. The specific calculation formula is:

[0012]

[0013] in, It is expressed as the suspected obstacle distance, and then an image recognition algorithm is used to detect the suspected obstacle outline in the image and determine whether there is an obstacle.

[0014] As a further improvement of the technical solution, the method of substituting the distance of the suspected obstacle into the image recognition algorithm and judging whether there is an obstacle in the suspected obstacle image in S1 specifically includes:

[0015] Use image recognition algorithms to detect the outline of suspected obstacles in the suspected obstacle image. If a suspected obstacle outline is detected, compare its position in the suspected obstacle image with the distance of the suspected obstacle detected by the laser radar and millimeter-wave radar. If a match is successful, it is determined that there is an obstacle at that position.

[0016] A three-dimensional coordinate system is established with the position of the target drone as the origin, and the distance between the target drone and the suspected obstacle is calculated. Get the 3D coordinates of suspected obstacles , the position of the suspected obstacle contour is determined by the image coordinates To express it, the conversion relationship between image coordinates and three-dimensional coordinates is established through the intrinsic parameter matrix and extrinsic parameter matrix of the binocular vision camera. The specific method is:

[0017] Assume that the internal parameter matrix of the binocular vision camera is ,in and is the focal length, and is the image center coordinate, and the external parameter matrix is ​​composed of the rotation matrix and translation vectors It represents the position and direction of the binocular vision camera in the world coordinate system through the formula The contour image coordinates of the suspected obstacle are converted into three-dimensional coordinates, that is, ,in is the scale factor, Expressed as the three-dimensional coordinates of the suspected obstacle outline, calculate the distance between the three-dimensional coordinates of the suspected obstacle outline and the three-dimensional coordinates of the suspected obstacle detected by the lidar and millimeter-wave radar , and set a distance threshold ,like , then the match is considered successful, and it is determined that there is an obstacle at that location.

[0018] As a further improvement of the technical solution, the method of establishing the obstacle avoidance decision model according to the target mapping task and the obstacle position coordinates in S2 specifically includes:

[0019] The target mapping task includes the target mapping area, key mapping targets and mapping time requirements. Suppose the target mapping area is a side length of The square area with coordinate range is to , the key mapping target location is , the time requirement is Complete the surveying and mapping task within hours;

[0020] Define the decision variables of the obstacle avoidance decision model, which include the horizontal flight angle, vertical flight angle, flight speed and altitude adjustment of the target UAV. The horizontal flight angle of the target UAV is , exist The vertical angle is , exist The flight speed is set between and The height adjustment amount is set to , exist and and set the objective function to ,in is the obstacle avoidance path length, which is obtained by integrating the points on the flight trajectory of the target UAV over a period of time. is the minimum distance between the target UAV and the obstacle. The minimum distance is obtained by calculating the position of the UAV and the position of the obstacle, that is, the obstacle avoidance decision model is established.

[0021] As a further improvement of the technical solution, the method S2 introduces a simulated annealing algorithm to optimize the obstacle avoidance decision model and formulate an obstacle avoidance strategy, which specifically includes the following steps:

[0022] S2.1. Combine the horizontal flight angle, vertical flight angle, flight speed and altitude adjustment of the target drone and encode them to obtain a real vector ;

[0023] S2.2. Check whether the position of the target UAV during flight is always within the target mapping area. If it is beyond the range, a target mapping area constraint penalty item is given. The target mapping area constraint penalty item is recorded as , based on the flight speed and path length, estimate the time required to complete the mapping task ,like , then a time requirement penalty item is given, and the time requirement penalty item is recorded as ;

[0024] S2.3, set the fitness function, the expression of the fitness function is ,in , , as well as is a weight coefficient, and the fitness function is used to evaluate the advantages and disadvantages of different combinations of flight directions, speeds, and altitude adjustments;

[0025] S2.4. Randomly generate an initial vector , the initial vector Set the initial horizontal flight angle, vertical flight angle, flight speed, and altitude adjustment for the target drone, and set the initial temperature , Temperature attenuation coefficient , termination temperature And the new vector , calculate the new vector and real vector The fitness difference , ,like If it is less than 0, a new vector is used. ,like If greater than 0, then the probability Accept the new solution, the probability ,in is a constant;

[0026] S2.5, repeat the above steps until the initial temperature drops to the termination temperature, the termination temperature is expressed as , the solution corresponding to the termination temperature is the optimal solution. The optimal solution is analyzed to obtain the final obstacle avoidance strategy, which is the horizontal flight angle, vertical flight angle, flight speed and altitude adjustment of the target UAV when the termination temperature is reached.

[0027] As a further improvement of the technical solution, in S3, the obstacle avoidance strategy is fed back to the control center of the target UAV, and the control center of the target UAV performs route planning for the target UAV according to the obstacle avoidance strategy, specifically including:

[0028] Assume that the horizontal flight angle of the target UAV in the obstacle avoidance strategy is , vertical flight angle is , the flight speed is and height adjustment , the horizontal flight angle and vertical flight angle of the target UAV are extracted from the obstacle avoidance strategy and defined as the adjusted target UAV flight angle , the coordinates of the target drone are set as , the coordinates of the key surveying and mapping targets are set as ;

[0029] According to the adjusted target drone flight angle , determine an initial flight direction vector, the initial flight direction vector is recorded as ,in, , , , starting from the initial position, it gradually advances along the flight direction vector, generates a series of intermediate points, forms a preliminary route, and determines the position of the target drone based on the flight speed and time interval, that is, at time After that, the position coordinates of the target UAV are expressed as .

[0030] As a further improvement of the technical solution, the method of performing real-time control on the attitude adjustment speed of the target UAV in S3 to obtain the final route of the target UAV specifically includes:

[0031] Obtain the horizontal attitude angle of the current target drone through the attitude sensor of the target drone and vertical attitude angle , the horizontal flight angle of the target drone in the obstacle avoidance strategy Horizontal attitude angle with the current target drone Compare and get the horizontal angle difference Similarly, the vertical angle difference is obtained ;

[0032] According to the time requirements for completing surveying and mapping tasks Set expected posture adjustment time , and using the angle difference and the expected adjustment time, calculate the angle change rate in the horizontal and vertical directions, the angle change rate in the horizontal direction is The vertical angle change rate is , assuming that the attitude adjustment speed of the target UAV is linearly related to the flight speed, height adjustment amount and angle change rate of the target UAV, the expression of the linear relationship is:

[0033]

[0034] in, Adjust the speed for the target drone’s attitude, , , as well as is the weight coefficient. During the flight of the target UAV, the position and attitude changes of the target UAV are continuously monitored. If a new obstacle is found, the new obstacle image is fed back to the obstacle avoidance decision model, and the obstacle avoidance strategy is recalculated. According to the new obstacle avoidance strategy and the real-time flight status, the route is adjusted and optimized in real time. After continuous adjustment and optimization, when the UAV reaches the target position, the final route is determined.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] A surveying and mapping UAV and surveying and mapping method for reducing surveying and mapping errors can more accurately detect the position and distance of obstacles and formulate more reasonable obstacle avoidance strategies through the fusion of multiple sensors and an optimized obstacle avoidance decision model, thereby reducing the risk of collision between the UAV and obstacles and improving the accuracy of surveying and mapping. Accurate route planning and real-time attitude adjustment speed control enable the UAV to fly more stably, reduce surveying and mapping errors caused by unstable flight or unreasonable routes, and ensure the accuracy of surveying and mapping data. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is the overall workflow diagram of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.

[0039] Example 1

[0040] See also Figure 1 As shown, this embodiment provides a surveying and mapping UAV and a surveying and mapping method for reducing surveying and mapping errors, comprising the following steps:

[0041] S1. Use the radar device and camera of the target UAV to monitor the surrounding environment of the target UAV in real time, obtain the suspected obstacle image, and calculate the distance of the suspected obstacle through data fusion technology. Then substitute the distance of the suspected obstacle into the image recognition algorithm and judge whether there is an obstacle in the suspected obstacle image. If there is, determine the location coordinates of the obstacle;

[0042] In S1, the radar device and camera of the target UAV are used to monitor the surrounding environment of the target UAV in real time, and the method of obtaining a real-time image of the surrounding environment of the target UAV specifically includes:

[0043] The radar device of the target drone is a laser radar and a millimeter wave radar, and the camera is a binocular vision camera. The laser radar is installed at the bottom center of the target drone to detect suspected obstacles in front of and below the target drone. The millimeter wave radar is installed on both sides of the target drone to monitor suspected obstacles on the side of the target drone. The binocular vision camera is installed in front of the target drone to provide images of suspected obstacles. The laser radar emits laser beams to the surroundings at a fixed frequency, receives the reflected signal, and records the distance of the suspected obstacle. The millimeter wave radar continuously scans the surrounding environment to obtain the distance of the suspected obstacle. The binocular vision camera continuously shoots the front of the target drone and transmits the suspected obstacle image to the processing system of the target drone, and obtains the real-time image of the surrounding environment of the target drone in real time, providing an accurate information source for subsequent data fusion and obstacle judgment. In this way, obstacles can be discovered more timely, the obstacle avoidance ability of the drone can be improved, and the risk of collision can be reduced, thereby ensuring the smooth progress of the surveying and mapping tasks and improving the accuracy and safety of surveying and mapping.

[0044] The method of calculating the distance of the suspected obstacle by data fusion technology in S1 specifically includes:

[0045] LiDAR and millimeter-wave radar each have their own advantages and disadvantages when detecting obstacles. Data fusion technology can comprehensively utilize their advantages to improve the accuracy of distance calculation of suspected obstacles. The time when LiDAR and millimeter-wave radar detect suspected obstacles is recorded, and the distance data of LiDAR and millimeter-wave radar detecting suspected obstacles at the same time are counted, which are recorded as and ,Will and Substitute the two into the formula to calculate the average distance of the two, and use the average distance of the two as the preliminary judgment of the suspected obstacle distance, make full use of the measurement data of the two radars, reduce the error and uncertainty of single radar measurement, and calculate the average distance to get a more reliable preliminary judgment value of the suspected obstacle distance. The suspected obstacle distance is the straight-line distance from the target UAV to the suspected obstacle. The specific calculation formula is:

[0046]

[0047] in, It is expressed as the distance of suspected obstacles, and then uses image recognition algorithms to detect the outline of suspected obstacles in the image and determine whether there are obstacles, providing a more accurate data basis for subsequent image recognition and obstacle judgment. This can improve the accuracy of obstacle detection, reduce misjudgment and missed judgment, thereby enhancing the obstacle avoidance ability of drones, reducing surveying and mapping errors, and improving the reliability and safety of surveying and mapping.

[0048] The method of substituting the distance of the suspected obstacle into the image recognition algorithm in S1 and judging whether there is an obstacle in the suspected obstacle image specifically includes:

[0049] In order to more accurately determine whether there is really an obstacle in the suspected obstacle image. By combining the image recognition algorithm with the detection data of the laser radar and millimeter-wave radar, the accuracy and reliability of obstacle judgment can be improved. The image recognition algorithm is used to detect the suspected obstacle outline in the suspected obstacle image. If the suspected obstacle outline is detected, its position in the suspected obstacle image is compared with the suspected obstacle distance detected by the laser radar and millimeter-wave radar. If the match is successful, it is determined that there is an obstacle at that position. The image information and the distance information measured by the radar are comprehensively utilized, giving full play to the advantages of different sensors and reducing the limitations of a single sensor.

[0050] By comparing the position of the suspected obstacle outline with the distance detected by the radar, and converting and matching the three-dimensional coordinates, the existence of the obstacle can be determined more accurately, reducing the probability of misjudgment. The three-dimensional coordinate system is established with the position of the target drone as the origin, and the distance between the target drone and the suspected obstacle is calculated. Get the 3D coordinates of suspected obstacles , the position of the suspected obstacle contour is determined by the image coordinates To express it, the conversion relationship between image coordinates and three-dimensional coordinates is established through the intrinsic parameter matrix and extrinsic parameter matrix of the binocular vision camera. The specific method is:

[0051] Assume that the internal parameter matrix of the binocular vision camera is ,in and is the focal length, and is the image center coordinate, and the external parameter matrix is ​​composed of the rotation matrix and translation vectors It represents the position and direction of the binocular vision camera in the world coordinate system through the formula The contour image coordinates of the suspected obstacle are converted into three-dimensional coordinates, that is, ,in is the scale factor, Expressed as the three-dimensional coordinates of the suspected obstacle outline, calculate the distance between the three-dimensional coordinates of the suspected obstacle outline and the three-dimensional coordinates of the suspected obstacle detected by the lidar and millimeter-wave radar , and set a distance threshold ,like , the match is considered successful, and it is determined that there is an obstacle at that location. A three-dimensional coordinate system is established and coordinate transformation is performed, which can more accurately determine the position of the obstacle in space, provide more precise information for the obstacle avoidance decision of the UAV, and ensure the accuracy and completeness of the surveying and mapping data.

[0052] S2. Establish an obstacle avoidance decision model based on the target mapping task and obstacle position coordinates, and introduce a simulated annealing algorithm to optimize the obstacle avoidance decision model and formulate an obstacle avoidance strategy;

[0053] The method of establishing the obstacle avoidance decision model in S2 according to the target mapping task and the obstacle position coordinates includes:

[0054] In order to establish a suitable obstacle avoidance decision model according to the specific target mapping task, the UAV can effectively avoid obstacles under the premise of meeting the mapping task requirements. The target mapping task includes the target mapping area, key mapping targets and mapping time requirements. The target mapping area is set to be The square area with coordinate range is to , the key mapping target location is , the time requirement is The surveying task was completed within 1 hour, taking into account specific task factors such as the target surveying area, key surveying targets and surveying time requirements, so that obstacle avoidance decisions are more in line with actual needs;

[0055] Define the decision variables of the obstacle avoidance decision model. The decision variables include the horizontal flight angle, vertical flight angle, flight speed, and altitude adjustment of the target UAV. The horizontal flight angle of the target UAV is , exist Adjust between, the vertical angle is , exist The flight speed is set between and The height adjustment is set to , exist and The objective function is set as ,in is the obstacle avoidance path length, which is obtained by integrating the points on the flight trajectory of the target UAV over a period of time. The minimum distance between the target UAV and the obstacle is obtained by calculating the position of the UAV and the obstacle, that is, the obstacle avoidance decision model is established. By setting the objective function, the obstacle avoidance path length and the minimum distance to the obstacle are used as optimization targets to reduce unnecessary flight distance and time and improve work efficiency.

[0056] S2 introduces a simulated annealing algorithm to optimize the obstacle avoidance decision model and formulate an obstacle avoidance strategy, which specifically includes the following steps:

[0057] The simulated annealing algorithm can search for the optimal or near-optimal solution in a complex solution space, and is suitable for optimizing obstacle avoidance decision models and formulating obstacle avoidance strategies;

[0058] S2.1. Combine the horizontal flight angle, vertical flight angle, flight speed and altitude adjustment of the target drone and encode them to obtain a real vector ;

[0059] To ensure that the target UAV can meet the requirements of the target mapping task during the flight, including mapping within the specified mapping area and completing the task within the specified time. If the UAV exceeds the scope of the target mapping area or fails to complete the task within the specified time, it may cause incomplete or inaccurate mapping data, affecting the quality and efficiency of the mapping.

[0060] S2.2. Check whether the position of the target UAV during flight is always within the target mapping area. If it is beyond the range, a target mapping area constraint penalty item is given. The target mapping area constraint penalty item is recorded as , based on the flight speed and path length, estimate the time required to complete the mapping task ,like , then a time requirement penalty is given, and the time requirement penalty is recorded as ,Through the constraint and punishment mechanism, the stability and controllability of UAV flight are enhanced, and the risk of unexpected situations is reduced;

[0061] S2.3, set the fitness function, the expression of the fitness function is ,in , , as well as is the weight coefficient, and the fitness function is used to evaluate the advantages and disadvantages of different combinations of flight direction, speed and altitude adjustment;

[0062] S2.4. Randomly generate an initial vector , the initial vector Set the initial horizontal flight angle, vertical flight angle, flight speed, and altitude adjustment for the target drone, and set the initial temperature , Temperature attenuation coefficient , termination temperature And the new vector , calculate the new vector and real vector The fitness difference , ,like If it is less than 0, a new vector is used. ,like If greater than 0, then the probability Accept the new solution, probability ,in The simulated annealing algorithm randomly generates an initial vector and continuously updates the vector during the iteration process to find the optimal obstacle avoidance strategy. Setting the initial temperature, temperature attenuation coefficient and termination temperature, as well as the way of accepting new solutions with probability are the core mechanisms of the simulated annealing algorithm, which helps the simulated annealing algorithm avoid falling into the local optimal solution during the search process, while improving the search efficiency and increasing the chance of finding the global optimal solution. By continuously iterating and updating the vector, the simulated annealing algorithm can find a better combination of the target drone's horizontal flight angle, vertical flight angle, flight speed and altitude adjustment, thereby formulating a more effective obstacle avoidance strategy.

[0063] S2.5. Repeat the above steps until the initial temperature drops to the final temperature. The final temperature is expressed as , then the solution corresponding to the termination temperature is the optimal solution. The optimal solution is analyzed to obtain the final obstacle avoidance strategy. The final obstacle avoidance strategy is the horizontal flight angle, vertical flight angle, flight speed and altitude adjustment of the target UAV when the termination temperature is reached.

[0064] S3. Feedback the obstacle avoidance strategy to the control center of the target UAV. The control center of the target UAV plans the route of the target UAV according to the obstacle avoidance strategy, and controls the attitude adjustment speed of the target UAV in real time to obtain the final route of the target UAV.

[0065] In S3, the obstacle avoidance strategy is fed back to the control center of the target UAV. The control center of the target UAV plans the route of the target UAV according to the obstacle avoidance strategy, specifically including:

[0066] Assume that the horizontal flight angle of the target UAV in the obstacle avoidance strategy is , vertical flight angle is , the flight speed is and height adjustment , the horizontal flight angle and vertical flight angle of the target UAV are extracted from the obstacle avoidance strategy and defined as the adjusted target UAV flight angle , the coordinates of the target drone are set as , the coordinates of the key surveying and mapping targets are set as ;

[0067] According to the adjusted target drone flight angle , determine an initial flight direction vector, the initial flight direction vector is recorded as ,in, , , ,By extracting the horizontal and vertical flight angles to determine the flight direction vector, and calculating the position coordinates according to the flight speed and time interval, it is possible to achieve accurate route planning.,Route planning is based on the obstacle avoidance strategy, which fully considers the obstacle avoidance needs of the drone and improves the safety of flight.,Starting from the initial position, it gradually advances along the flight direction vector, generates a series of intermediate points, and forms a preliminary route, ensuring that the drone maintains a stable flight trajectory during flight, reducing unnecessary fluctuations and adjustments, and determining the position of the target drone according to the flight speed and time interval, that is, at time After that, the position coordinates of the target UAV are expressed as .

[0068] In S3, the attitude adjustment speed of the target UAV is controlled in real time to obtain the final route of the target UAV, specifically including:

[0069] In order to control the attitude adjustment speed of the target UAV in real time, so that it can adjust its attitude in time according to the obstacle avoidance strategy and actual flight conditions, ensure that the UAV can fly accurately along the planned route, and respond in time when encountering new obstacles, and re-plan the route to obtain the final optimal route.

[0070] Obtain the horizontal attitude angle of the current target drone through the attitude sensor of the target drone and vertical attitude angle , the horizontal flight angle of the target drone in the obstacle avoidance strategy Horizontal attitude angle with the current target drone Compare and get the horizontal angle difference Similarly, the vertical angle difference is obtained , the attitude sensor can obtain the attitude angle of the drone in real time and compare it with the angle in the obstacle avoidance strategy, so as to find the angle difference in time and make attitude adjustments quickly to improve the mapping accuracy;

[0071] According to the time requirements for completing surveying and mapping tasks Set expected posture adjustment time , and use the angle difference and expected adjustment time to calculate the angle change rate in the horizontal and vertical directions. The angle change rate in the horizontal direction is The vertical angle change rate is , the angle change rate is calculated using the angle difference and the expected adjustment time, and the attitude adjustment speed is determined through a linear relationship, which can achieve precise control of the attitude adjustment. Assuming that the attitude adjustment speed of the target UAV is linearly related to the flight speed, height adjustment amount and angle change rate of the target UAV, the expression of the linear relationship is:

[0072]

[0073] in, Adjust the speed for the target drone’s attitude, , , as well as The weight coefficient is used. During the flight of the target UAV, the position and attitude changes of the target UAV are continuously monitored. If a new obstacle is found, the new obstacle image is fed back to the obstacle avoidance decision model, and the obstacle avoidance strategy is recalculated. According to the new obstacle avoidance strategy and the real-time flight status, the route is adjusted and optimized in real time. After continuous adjustment and optimization, when the UAV reaches the target position, the final route is determined, so that the UAV can adapt to complex and changing environments, and can fly stably and reliably in various situations, thereby improving the reliability of the entire surveying and mapping system.

[0074] The above-mentioned surveying and mapping UAV and surveying and mapping method for reducing surveying and mapping errors use the radar device and camera of the UAV to monitor the surrounding environment in real time, obtain images of suspected obstacles and determine their positions, and establish an obstacle avoidance decision model according to the target surveying and mapping task, introduce a simulated annealing algorithm optimization model to formulate an obstacle avoidance strategy, and finally feed back the obstacle avoidance strategy to the control center. The control center plans the route based on this and controls the attitude adjustment speed of the UAV in real time to obtain the final route. The method includes the specific use of radar and camera, the steps of data fusion and image recognition, the setting of obstacle avoidance decision model, the application of simulated annealing algorithm, and the control of route planning and attitude adjustment, etc., aiming to improve the obstacle avoidance capability and surveying and mapping accuracy of the UAV and reduce surveying and mapping errors through these technical means.

[0075] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A surveying and mapping UAV and surveying and mapping method for reducing surveying and mapping errors, characterized in that: The following steps are involved: S1. Use the radar device and camera of the target UAV to monitor the surrounding environment of the target UAV in real time, obtain the suspected obstacle image, and calculate the distance of the suspected obstacle through data fusion technology. Then substitute the distance of the suspected obstacle into the image recognition algorithm and judge whether there is an obstacle in the suspected obstacle image. If there is, determine the location coordinates of the obstacle; S2. Establish an obstacle avoidance decision model according to the target mapping task and the obstacle position coordinates, and introduce a simulated annealing algorithm to optimize the obstacle avoidance decision model and formulate an obstacle avoidance strategy, wherein the method of optimizing the obstacle avoidance decision model and formulating the obstacle avoidance strategy by the simulated annealing algorithm specifically includes the following steps: S2.

1. Combine the horizontal flight angle, vertical flight angle, flight speed and altitude adjustment of the target UAV and encode them to obtain a real vector W = [θ, φ, v, Δh]; S2.

2. Check whether the position of the target UAV during the flight is always within the target mapping area. If it is beyond the range, a target mapping area constraint penalty item is given, and the target mapping area constraint penalty item is recorded as P1. According to the flight speed and path length, the time T required to complete the mapping task is estimated. If T ≥ t, a time requirement penalty item is given, and the time requirement penalty item is recorded as P2. S2.3, set the fitness function, the fitness function expression is F e =ω1L+ω2D+ω3P1+ω4P2, where ω1, ω2, ω3 and ω4 are weight coefficients, and the fitness function is used to evaluate the advantages and disadvantages of different combinations of flight direction, speed and altitude adjustment; S2.4, randomly generate an initial vector W0, which is a combination of the target drone's initial horizontal flight angle, vertical flight angle, flight speed, and altitude adjustment, and set the initial temperature T0, temperature attenuation coefficient α, and termination temperature T end and a new vector W', calculate the fitness difference ΔF between the new vector W' and the real number vector W, where ΔF = F e (W')-F e (W), if ΔF is less than 0, use the new vector W', if ΔF is greater than 0, accept the new solution with probability I, the probability Where y is a constant; S2.5, repeat the above steps until the initial temperature drops to the termination temperature, the termination temperature is expressed as T end =T0×α, then the solution corresponding to the termination temperature is the optimal solution. The optimal solution is analyzed to obtain the final obstacle avoidance strategy, which is the horizontal flight angle, vertical flight angle, flight speed and altitude adjustment of the target UAV when the termination temperature is reached; S3. Feedback the obstacle avoidance strategy to the control center of the target UAV. The control center of the target UAV plans the route of the target UAV according to the obstacle avoidance strategy, and controls the attitude adjustment speed of the target UAV in real time to obtain the final route of the target UAV.

2. A surveying and mapping UAV and surveying and mapping method for reducing surveying and mapping errors according to claim 1, characterized in that: The method of using the radar device and camera of the target drone to monitor the surrounding environment of the target drone in real time in S1 to obtain a real-time image of the surrounding environment of the target drone specifically includes: The radar device of the target UAV is a laser radar and a millimeter-wave radar, and the camera is a binocular vision camera. The laser radar is installed at the bottom center of the target UAV to detect suspected obstacles in front of and below the target UAV. The millimeter-wave radar is installed on both sides of the target UAV to monitor suspected obstacles on the sides of the target UAV. The binocular vision camera is installed in front of the target UAV to provide images of suspected obstacles. The laser radar emits a laser beam to the surroundings at a fixed frequency, receives the reflected signal, and records the distance of the suspected obstacle. The millimeter-wave radar continuously scans the surrounding environment to obtain the distance of the suspected obstacle. The binocular vision camera continuously shoots the picture in front of the target UAV and transmits the image of the suspected obstacle to the processing system of the target UAV.

3. The surveying and mapping UAV and surveying and mapping method for reducing surveying and mapping errors according to claim 1, characterized in that: The method of calculating the distance of the suspected obstacle by using the data fusion technology in S1 specifically includes: The time when the laser radar and millimeter-wave radar detect suspected obstacles is recorded, and the distance data of the laser radar and millimeter-wave radar detecting suspected obstacles at the same time are counted, which are recorded as d j and d h , d j and d h Substitute them into the formula to calculate the average distance between the two, and use the average distance between the two as the suspected obstacle distance for preliminary judgment. The suspected obstacle distance is the straight-line distance from the target UAV to the suspected obstacle. The specific calculation formula is: Among them, d z It is expressed as the suspected obstacle distance, and then an image recognition algorithm is used to detect the suspected obstacle outline in the image and determine whether there is an obstacle.

4. The surveying and mapping UAV and surveying and mapping method for reducing surveying and mapping errors according to claim 3, characterized in that: The method of substituting the distance of the suspected obstacle into the image recognition algorithm and judging whether there is an obstacle in the suspected obstacle image in S1 specifically includes: Use image recognition algorithms to detect the outline of suspected obstacles in the suspected obstacle image. If a suspected obstacle outline is detected, compare its position in the suspected obstacle image with the distance of the suspected obstacle detected by the laser radar and millimeter-wave radar. If a match is successful, it is determined that there is an obstacle at that position. A three-dimensional coordinate system is established with the position of the target UAV as the origin, and the distance d between the target UAV and the suspected obstacle is calculated. z The three-dimensional coordinates (x1, y1, z1) of the suspected obstacle are obtained. The position of the suspected obstacle outline is represented by the image coordinates (u, v). The conversion relationship between the image coordinates and the three-dimensional coordinates is established through the intrinsic matrix and extrinsic matrix of the binocular vision camera. The specific method is: Assume that the internal parameter matrix of the binocular vision camera is where f x and f y is the focal length, c x and c y is the image center coordinate, the external parameter matrix is ​​composed of the rotation matrix R and the translation vector t, which represents the position and direction of the binocular vision camera in the world coordinate system, through the formula The coordinates of the suspected obstacle contour image are converted into three-dimensional coordinates, i.e. (x2, y2, z2), where s is the scale factor. The (x2, y2, z2) represents the three-dimensional coordinates of the suspected obstacle contour. The distance d between the three-dimensional coordinates of the suspected obstacle contour and the three-dimensional coordinates of the suspected obstacle detected by the laser radar and the millimeter-wave radar is calculated. q , and set a distance threshold T d , if d q ≤T d , then the match is considered successful, and it is determined that there is an obstacle at that location.

5. The surveying and mapping UAV and surveying and mapping method for reducing surveying and mapping errors according to claim 1, characterized in that: The method of establishing the obstacle avoidance decision model according to the target mapping task and the obstacle position coordinates in S2 specifically includes: The target mapping task includes the target mapping area, key mapping targets and mapping time requirements. The target mapping area is a square area with a side length of a, the coordinate range is (0, 0, 0) to (a, a, 0), and the key mapping target position is (x t ,y t , z t ), the time requirement is to complete the surveying and mapping task within t hours; Define the decision variables of the obstacle avoidance decision model, the decision variables include the horizontal flight angle, vertical flight angle, flight speed and altitude adjustment of the target UAV, the horizontal flight angle of the target UAV is θ, and θ is in [-θ max ,θ max ], the vertical flight angle of the target UAV is φ, and φ is adjusted between [-φ max ,φ max ], the flight speed is set at v min and v max The height adjustment amount is set to Δh, and the Δh is -Δh max and Δh max The objective function is set to F=ω1L+ω2D, where L is the obstacle avoidance path length, which is obtained by integrating the points on the flight trajectory of the target UAV over a period of time, and D is the minimum distance between the target UAV and the obstacle, which is obtained by calculating the position of the UAV and the position of the obstacle, thus completing the establishment of the obstacle avoidance decision model.

6. The surveying and mapping UAV and surveying and mapping method for reducing surveying and mapping errors according to claim 1, characterized in that: In S3, the obstacle avoidance strategy is fed back to the control center of the target UAV. The control center of the target UAV performs route planning for the target UAV according to the obstacle avoidance strategy, specifically including: Assume that the horizontal flight angle of the target UAV in the obstacle avoidance strategy is θ1, the vertical flight angle is φ1, the flight speed is v1, and the height adjustment amount is Δh1. The horizontal flight angle and vertical flight angle of the target UAV are extracted from the obstacle avoidance strategy and defined as the adjusted target UAV flight angle λ = (θ1, φ1). The coordinates of the target UAV are set as (x0, y0, z0), and the coordinates of the key mapping target are set as (x t ,y t , z t ); According to the adjusted target UAV flight angle λ = (θ1, φ1), an initial flight direction vector is determined, and the initial flight direction vector is recorded as Among them, d x =sinφ1cosθ1,d y =sinφ1sinθ1,d z = cosφ1, starting from the initial position, it gradually advances along the flight direction vector to generate a series of intermediate points to form a preliminary route, and determines the position of the target UAV according to the flight speed and time interval. That is, after time t, the position coordinates of the target UAV can be expressed as (x0+v×t×d x ,y0+v×t×d y , z0+v×t×d z ).

7. The surveying and mapping UAV and surveying and mapping method for reducing surveying and mapping errors according to claim 1, characterized in that: The method of performing real-time control on the attitude adjustment speed of the target UAV in S3 to obtain the final route of the target UAV specifically includes: The horizontal attitude angle θ2 and vertical attitude angle φ2 of the current target UAV are obtained through the attitude sensor of the target UAV, and the horizontal flight angle θ1 of the target UAV in the obstacle avoidance strategy is compared with the horizontal attitude angle θ2 of the current target UAV to obtain the horizontal angle difference Δθ=θ1-θ2. Similarly, the vertical flight angle difference Δφ=φ1-φ2 ​​is obtained. According to the time requirement t for completing the surveying and mapping task, the expected attitude adjustment time t1 is set, and the angle change rate in the horizontal and vertical directions is calculated using the angle difference and the expected adjustment time. The angle change rate in the horizontal direction is The rate of change of the vertical angle is Assuming that the attitude adjustment speed of the target UAV is linearly related to the flight speed, height adjustment amount and angle change rate of the target UAV, the expression of the linear relationship is: Among them, v ed is the attitude adjustment speed of the target UAV, ω1, ω2, ω3 and ω4 are weight coefficients. During the flight of the target UAV, the position and attitude changes of the target UAV are continuously monitored. If a new obstacle is found, the new obstacle image is fed back to the obstacle avoidance decision model, and the obstacle avoidance strategy is recalculated. According to the new obstacle avoidance strategy and real-time flight status, the route is adjusted and optimized in real time. After continuous adjustment and optimization, when the UAV reaches the target position, the final route is determined.

Images