Unmanned aerial vehicle real-time trajectory prediction method and system based on low-altitude economy

Abstract

The invention relates to the technical field of data processing, in particular to an unmanned aerial vehicle real-time trajectory prediction method and system based on low-altitude economy. The method comprises the following steps: collecting each position point on an inspection path and data of an unmanned aerial vehicle on each position point in a historical inspection process; obtaining each historical position point corresponding to each position point in the inspection path and the attitude similarity of the historical position point; based on the attitude similarity degree, acquiring the attitude complexity degree of each position point in the inspection path; based on the attitude complexity, the distance between each group of flight data vectors corresponding to each position point in the inspection path is self-adapted; and based on the distance, clustering the flight data vector corresponding to each position point in the inspection path to obtain a plurality of clusters of each position point in the inspection path, and predicting the flight state of the next position point of the unmanned aerial vehicle according to the clusters, thereby improving the accuracy of flight state prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a real-time trajectory prediction method and system for unmanned aerial vehicles based on low-altitude economy. Background Art

[0002] With the rapid development of the low-altitude economy, the large-scale application of unmanned aerial vehicle (UAV) inspection is becoming increasingly popular. During the inspection operation, although there is a pre-set fixed route plan for UAV inspection, due to environmental factors, the actual flight trajectory is difficult to exactly match the planned trajectory. Therefore, during the actual flight of the UAV, it is necessary to predict its trajectory to make the actual flight trajectory of the UAV closer to the planned trajectory.

[0003] In related technologies, for example, a Chinese patent application document with the authorization announcement number CN111461292B discloses a real-time trajectory prediction method for UAVs, including: acquiring data; preprocessing the data; generating a data set of each variable of the UAV trajectory; predicting the UAV state using a Markov model; adding a BN layer according to the state of the UAV, and establishing a UAV trajectory prediction model based on an LSTM network; predicting the longitude, latitude and altitude of the UAV.

[0004] In the above related technologies, the UAV state is classified only based on the speed of the UAV, without considering environmental change factors. For example, when the UAV is in two states with the same speed but very different environmental conditions, one is flying with the wind and the other is flying against the wind. Although the speeds are the same, their actual motion states (such as power consumption, flight stability, etc.) are significantly different. Therefore, the one-sidedness of this UAV state classification method will cause the predicted next position state of the Markov model not to be well applied to trajectory prediction, and further cause the subsequent trajectory prediction result to deviate from the actual flight trajectory. Summary of the Invention

[0005] In order to solve the technical problem that when classifying the UAV state only based on the speed of the UAV, the environmental conditions are not considered, the present invention provides a real-time trajectory prediction method and system for UAVs based on low-altitude economy.

[0006] In the first aspect, the present invention provides a real-time trajectory prediction method for UAVs based on low-altitude economy, adopting the following technical solution: A real-time trajectory prediction method for UAVs based on low-altitude economy includes the steps of: Collecting each position point on the inspection path and the data of the UAV at each position point during each historical inspection; obtaining the attitude feature vector of the UAV at each position point during each historical inspection according to the data of the UAV at each position point during each historical inspection; Obtain each historical position point corresponding to each position point in the inspection path, as well as the first neighboring position point and the second neighboring position point of each historical position point; based on the attitude feature vectors of the historical position point and the first and second neighboring position points of the historical position point, obtain the attitude similarity degree of each historical position point corresponding to each position point in the inspection path; based on the attitude similarity degree, obtain the attitude complexity of each position point in the inspection path; Obtain each group of flight data vectors corresponding to each position point in each inspection path; based on the attitude complexity, obtain the distance between each group of flight data vectors corresponding to each position point in the inspection path; based on the distance between each group of flight data vectors, cluster the flight data vectors corresponding to each position point in the inspection path to obtain several clustering clusters for each position point in the inspection path; according to the several clustering clusters for each position point in the inspection path, predict the flight state of the UAV during the actual inspection process, and then predict the trajectory of the UAV.

[0007] The innovation of the present invention is that first, the attitude feature vector of the UAV at each position point during the historical inspection process is obtained. Based on the attitude feature vector, the attitude complexity of each position point in the inspection path is obtained, which reflects the number of types of flight states existing at each position point. Then, considering the environmental change factors, each group of flight data vectors corresponding to each position point in each inspection path is obtained; based on the attitude complexity, the distance between each group of flight data vectors corresponding to each position point in the inspection path is adjusted so that the distance between the flight data vectors with more flight states as the fulcrum is larger, in order to obtain clustering clusters of multiple flight states subsequently. Finally, based on the distance metric, the flight data vectors are clustered to obtain several clustering clusters for each position point in the inspection path, improving the self - adaptability and robustness of the UAV state division at each position point, and subsequently improving the accuracy of predicting the flight state and trajectory based on the clustering clusters.

[0008] Preferably, the obtaining of the attitude feature vector of the UAV at each position point during each historical inspection process includes: ; In the formula, represents the attitude feature vector of the UAV at the j - th position point during the i - th historical inspection process; represents the pitch angle of the UAV at the j - th position point during the i - th historical inspection process; represents the roll angle of the UAV at the j - th position point during the i - th historical inspection process; represents the yaw angle of the UAV at the j - th position point during the i - th historical inspection process; represents the gray information entropy of the image collected by the UAV at the j-th position point during the i-th historical inspection process.

[0009] It is convenient to obtain the attitude complexity of each position point in the inspection path according to the attitude feature vector.

[0010] Preferably, the obtaining of each historical position point corresponding to each position point in the inspection path and the first neighboring position point and the second neighboring position point of each historical position point includes: Obtain the distance between the k-th position point in the inspection path and each position point in any historical inspection process, and record the position point in the historical inspection process corresponding to the minimum distance as a historical position point corresponding to the k-th position point in the inspection path. Record the adjacent position points before and after the historical position point in its historical inspection process as the first neighboring position point and the second neighboring position point of the historical position point corresponding to the k-th position point in the inspection path.

[0011] Preferably, the obtaining of the attitude similarity degree of each historical position point corresponding to each position point in the inspection path includes:

[0012] In the formula, represents the attitude similarity degree of the h-th historical position point corresponding to the k-th position point in the inspection path; represents the attitude feature vector of the h-th historical position point corresponding to the k-th position point in the inspection path; represents the attitude feature vector of the first neighboring position point of the h-th historical position point corresponding to the k-th position point in the inspection path; represents the attitude feature vector of the second neighboring position point of the h-th historical position point corresponding to the k-th position point in the inspection path; represents the normalized cosine similarity function.

[0013] Preferably, the obtaining of the attitude complexity of each position point in the inspection path includes: Take the standard deviation of the attitude similarity degrees of all historical position points corresponding to each position point in the inspection path as the attitude complexity of each position point in the inspection path.

[0014] The attitude complexity reflects the number of types of flight states of each position point in the inspection path.

[0015] Preferably, the obtaining of each group of flight data vectors corresponding to each position point in each inspection path includes: Fly the drone to the longitude, latitude, horizontal velocity, vertical velocity, x-axis acceleration, y-axis acceleration, z-axis acceleration, pitch angle, roll angle, yaw angle, and altitude at each historical position point corresponding to each position point on the inspection path, and use them as each flight data vector corresponding to each position point on the inspection path; Combine the flight data vectors corresponding to each position point on the inspection path in pairs to obtain each group of flight data vectors corresponding to each position point on the inspection path.

[0016] Preferably, obtaining the distance between each group of flight data vectors corresponding to each position point on the inspection path includes:

[0017] In the formula, represents the distance between the o-th group of flight data vectors corresponding to the k-th position point on the inspection path; represents the attitude complexity of the k-th position point on the inspection path; represents the number of elements in each flight data vector corresponding to each position point on the inspection path; and represents the value of the m-th element of the first flight data vector and the second flight data vector in the o-th group of flight data vectors corresponding to the k-th position point on the inspection path; represents the information entropy of the m-th element in all flight data vectors corresponding to the k-th position point on the inspection path.

[0018] Enable more types of flight states to be formed at position points with greater attitude complexity.

[0019] Preferably, obtaining several clustering clusters for each position point on the inspection path includes: Preset the neighborhood radius ε and the minimum number of points minpts within the neighborhood radius. Based on the distance between each group of historical position points corresponding to each position point on the inspection path, use DBSCAN clustering to cluster the flight data vectors corresponding to each position point on the inspection path to obtain several clustering clusters for each position point on the inspection path.

[0020] Preferably, predicting the flight state of the drone during actual inspection according to several clustering clusters of each position point on the inspection path includes: During the actual inspection of the drone, obtain the flight data vector of the current position point of the drone; obtain the distance between the current position point of the drone and each position point on the inspection path, and record the position point on the inspection path corresponding to the minimum value of the distance as the target position point; According to the flight data vector of the current position point of the UAV, the KNN algorithm with K = 5 is used in the clustering cluster of the target position point to obtain the clustering cluster to which the flight data vector of the current position point of the UAV belongs; the clustering cluster to which the flight data vector of the current position point of the UAV belongs is input into the Markov model to predict the flight state of the next position point of the UAV.

[0021] The environmental change factors are considered, and the accuracy of flight state prediction is improved.

[0022] In a second aspect, the present invention provides a real-time trajectory prediction system for a UAV based on the low-altitude economy, and adopts the following technical solutions: A real-time trajectory prediction system for a UAV based on the low-altitude economy includes: a processor and a memory, and the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned real-time trajectory prediction method for a UAV based on the low-altitude economy is implemented.

[0023] By adopting the above technical solutions, the above-mentioned real-time trajectory prediction method for a UAV based on the low-altitude economy is generated into a computer program and stored in the memory, so as to be loaded and executed by the processor, and thus a terminal device is manufactured according to the memory and the processor, which is convenient to use.

[0024] The present invention has the following technical effects: First, according to the differences in the historical inspection data at the position points, the attitude complexity of each position point in the inspection path is obtained, which reflects the number of types of flight states existing at each position point; then, each group of flight data vectors corresponding to each position point in each inspection path is obtained, which represents the flight state of the UAV, and based on the attitude complexity, the distance between each group of flight data vectors corresponding to each position point in the inspection path is adjusted, so that the distance between the flight data vectors at the position points with more flight states is larger, in order to obtain more clustering clusters, improving the self-adaptability and robustness of the UAV state division at each position point, and subsequently predicting the flight state and trajectory during the actual inspection based on the clustering clusters. Description of the Drawings

[0025] Figure 1 It is a flowchart of the method in an embodiment of a real-time trajectory prediction method for a UAV based on the low-altitude economy of the present invention. Detailed Embodiments

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments.

[0027] The embodiments of the present invention disclose a real-time trajectory prediction method for a UAV based on the low-altitude economy, referring toFigure 1 , including steps S1 - S4: S1: Collect each position point on the inspection path and collect the data of the drone at each position point during historical inspections.

[0028] In an embodiment of the present invention, each position point on the inspection path is obtained, and the longitude, latitude, horizontal velocity, vertical velocity, acceleration in the x - axis direction, acceleration in the y - axis direction, acceleration in the z - axis direction, pitch angle, roll angle, yaw angle, and altitude of the drone at each position point during each historical inspection are obtained, as well as the images collected at each position.

[0029] S2: According to the data of the drone at each position point during historical inspections, obtain the attitude feature vector of the drone at each position point during each historical inspection; obtain each historical position point corresponding to each position point on the inspection path; obtain the attitude similarity degree of each historical position point corresponding to each position point on the inspection path; based on the attitude similarity degree, obtain the attitude complexity degree of each position point on the inspection path.

[0030] It should be noted that generally, the drone should fly stably during the inspection process. When the flight environment (wind speed, air current, or obstacles) changes, it is usually necessary to adjust the power system of the drone to maintain the position stability of the drone. Otherwise, it will deviate from the planned trajectory. The adjustment of the drone's power system will cause a certain change in the attitude angle of the drone. Also, since the camera carried by the drone is usually fixed on the fuselage and is strictly bound to the fuselage attitude, when the fuselage attitude angle changes, the camera will tilt and rotate synchronously with the fuselage, resulting in a change in the shooting angle, that is, a change in the information entropy of the images collected by the drone. Therefore, an attitude feature vector of the drone during the inspection process is constructed through the pitch angle, roll angle, and yaw angle of the drone during the inspection process and the information entropy of the collected images, which is used to describe the attitude characteristics of the drone during the inspection process.

[0031] In an embodiment of the present invention, obtain the attitude feature vector of the drone at each position point during each historical inspection:

[0032] where represents the attitude feature vector of the drone at the j - th position point during the i - th historical inspection; represents the pitch angle of the drone at the j - th position point during the i - th historical inspection; represents the roll angle of the drone at the j - th position point during the i - th historical inspection; represents the yaw angle of the drone at the j - th position point during the i - th historical inspection; represents the entropy of the grayscale information of the image collected by the drone at the j-th position point during the i-th historical inspection process; It should be noted that when the inspection path of the drone is specified and the environment does not change significantly, it is theoretically considered that the flight states of the drone at the same position on the path during all historical inspection processes should be similar; however, if the drone is interfered by the environment during a certain inspection process, resulting in the flight state deviating from the flight state of the drone when it flew to this position during the historical inspection process when the drone flies to this position on the inspection path, resulting in abnormal fluctuations in the flight state; for any position point on the inspection path, if the difference in the similarity degree of the attitudes of all historical inspection processes when passing through this position point and the attitudes of its adjacent position points before and after is greater, it means that the drone may pass through this position point on the inspection path in different attitudes, then the attitude complexity of this position point on the inspection path is higher.

[0033] In the embodiment of the present invention, the distance between the k-th position point on the inspection path and each position point in any historical inspection process is obtained, and the position point in the historical inspection process corresponding to the minimum distance is recorded as a historical position point corresponding to the k-th position point on the inspection path, and the adjacent position points before and after the historical position point in its historical inspection process are recorded as the first neighboring position point and the second neighboring position point of the historical position point corresponding to the k-th position point on the inspection path; Similarly, each historical position point corresponding to each position point on the inspection path is obtained, and the first neighboring position point and the second neighboring position point of each historical position point corresponding to each position point on the inspection path are obtained.

[0034] Obtain the similarity degree of the attitudes of each historical position point corresponding to each position point on the inspection path:

[0035] In the formula, represents the similarity degree of the attitude of the h-th historical position point corresponding to the k-th position point on the inspection path; represents the attitude feature vector of the h-th historical position point corresponding to the k-th position point on the inspection path; represents the first neighboring position point of the h-th historical position point corresponding to the k-th position point on the inspection path posture features vector; represents the attitude feature vector of the second neighboring position point of the h-th historical position point corresponding to the k-th position point on the inspection path; represents the normalized cosine similarity function. It should be noted that is equal to the cosine similarity between A and B plus 1 divided by 2; The larger the value, the more similar the attitude of the drone when it flies to the k-th position point in the inspection path during the historical inspection process is to the attitudes when it flies to the position points before and after the k-th position point. The larger the value, the more likely the drone passes through the k-th position point and its adjacent position points in the inspection path with the same attitude during the historical inspection process.

[0036] It should be noted that if the drone passes through any position point in the inspection path and its adjacent position points with the same attitude during all historical inspection processes, then the attitude complexity of this position point in the inspection path is smaller. On the contrary, if the drone passes through any position point in the inspection path and its adjacent position points with different attitudes during all historical inspection processes, then the attitude complexity of this position point in the inspection path is larger.

[0037] The standard deviation of the attitude similarity degrees of all historical position points corresponding to each position point in the inspection path is used as the attitude complexity of each position point in the inspection path.

[0038] S3: Obtain the flight data vectors corresponding to each position point in the inspection path. According to the attitude complexity of each position point in the inspection path and the flight data vectors, obtain the distances between each group of flight data vectors corresponding to each position point in the inspection path; based on the distances, cluster the flight data vectors corresponding to each position point in the inspection path to obtain several clustering clusters for each position point in the inspection path.

[0039] It should be noted that since the flight states of the drone are different when it flies to different position points in the inspection path, if the DBSCAN clustering algorithm is directly used to cluster the states of the drone at all positions on the inspection path, the unique flight states of different position points will be confused, resulting in the clustering results being unable to accurately reflect the actual flight characteristics of each positioning point, making the subsequent state prediction of the drone inaccurate. Therefore, in the present invention, the DBSCAN clustering algorithm is used to cluster the flight states at all historical position points corresponding to each position point in the inspection path; However, when clustering the flight states at all historical position points corresponding to each position point in the inspection path using a fixed distance threshold, if there are more types of flight states at all historical position points corresponding to any position point in the inspection path, then using a fixed threshold may cause historical position points originally belonging to different flight states to be merged into the same cluster, affecting the subsequent prediction of flight states. Therefore, according to the attitude complexity of each position point in the inspection path, the present invention dynamically adjusts the distance measurement method between the historical position points corresponding to each position point in the inspection path, so that the actual distance between historical position points of different flight states exceeds the fixed distance threshold, avoiding the merging of clusters that should be independent, thereby making the clustering result more conform to the actual number of states. That is, the greater the attitude complexity of any position point in the inspection path, the more types of flight states exist at all historical position points corresponding to this position point in the inspection path. At this time, the distance between the historical position points corresponding to each position point in the inspection path is increased.

[0040] In the embodiment of the present invention, the longitude, latitude, horizontal velocity, vertical velocity, x-axis acceleration, y-axis acceleration, z-axis acceleration, pitch angle, roll angle, yaw angle, and altitude of the drone flying to each historical position point corresponding to each position point in the inspection path are used as each flight data vector corresponding to each position point in the inspection path.

[0041] The flight data vectors corresponding to each position point in the inspection path are combined in pairs to obtain each group of flight data vectors corresponding to each position point in the inspection path.

[0042] Obtain the distance between each group of flight data vectors corresponding to each position point in the inspection path:

[0043] In the formula, represents the distance between the o-th group of flight data vectors corresponding to the k-th position point in the inspection path; represents the attitude complexity of the k-th position point in the inspection path; represents the number of elements in each flight data vector corresponding to each position point in the inspection path; represents the value of the m-th element of the first flight data vector in the o-th group of flight data vectors corresponding to the k-th position point in the inspection path; represents the value of the m-th element of the second flight data vector in the o-th group of flight data vectors corresponding to the k-th position point in the inspection path; represents the information entropy of the m-th element in all flight data vectors corresponding to the k-th position point in the inspection path; It represents the difference of the m-th element in the o-th group of flight data vectors corresponding to the k-th position point in the inspection path. The larger its value, the greater the distance between the o-th group of flight data vectors; The larger the value of, it indicates that the fluctuation amplitude of the m-th element when the drone flies to the k-th position point in the inspection path during all historical inspections is relatively large. Therefore When the value of is larger, the higher the credibility that the distance between the o-th group of flight data vectors is larger; The larger the value of, it represents that there are more types of flight states when the drone flies to the k-th position point in the inspection path during all historical inspections, indicating that the actual position points corresponding to when the drone flies to the k-th position point in the inspection path are more dispersed in spatial distribution. At this time, the distance between the o-th group of flight data vectors corresponding to the k-th position point in the inspection path is larger, making the distance more likely to exceed the distance threshold preset by the DBSCAN clustering algorithm, thereby forming more types of flight states at the k-th position point in the inspection path.

[0044] The preset neighborhood radius ε = 5, and the minimum number of points minpts within the neighborhood radius = 15. Based on the distances between each group of historical position points corresponding to each position point in the inspection path, DBSCAN clustering is used to cluster the flight data vectors corresponding to each position point in the inspection path, obtaining several clustering clusters for each position point in the inspection path.

[0045] S4: Obtain the flight data vector of the current position point of the drone during the actual inspection process. According to the several clustering clusters of each position point in the inspection path and the flight data vector of the current position point of the drone, after predicting the flight state of the next position point of the drone, predict the trajectory of the drone.

[0046] In the embodiment of the present invention, during the actual inspection of the drone, obtain the flight data vector of the current position point of the drone; Obtain the distances between the current position point of the drone and each position point in the inspection path, and record the position point in the inspection path corresponding to the minimum value of the distances as the target position point. According to the flight data vector of the current position point of the drone, use the KNN algorithm with K = 5 in the clustering cluster of the target position point to obtain the clustering cluster to which the flight data vector of the current position point of the drone belongs; Input the clustering cluster to which the flight data vector of the current position point of the drone belongs into the Markov model to predict the flight state of the next position point of the drone, and finally predict the trajectory of the drone according to the content in the Chinese patent application document with the authorization announcement number CN111461292B.

[0047] The above are all preferred embodiments of the present invention, and the protection scope of the present invention is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present invention shall be covered within the protection scope of the present invention.

Claims

1. A real-time trajectory prediction method for drones based on low-altitude economy, characterized in that, Including: Collecting each position point on the inspection path and the data of the UAV at each position point during each historical inspection; Obtaining the attitude feature vector of the UAV at each position point during each historical inspection according to the data of the UAV at each position point during each historical inspection; Obtaining each historical position point corresponding to each position point on the inspection path and the first neighboring position point and the second neighboring position point of each historical position point; Obtaining the attitude similarity degree of each historical position point corresponding to each position point on the inspection path based on the attitude feature vectors of the historical position point and the first neighboring position point and the second neighboring position point of the historical position point; obtaining the attitude complexity of each position point on the inspection path based on the attitude similarity degree; Obtaining each group of flight data vectors corresponding to each position point on each inspection path; obtaining the distance between each group of flight data vectors corresponding to each position point on the inspection path based on the attitude complexity; Clustering the flight data vectors corresponding to each position point on the inspection path based on the distance between each group of flight data vectors, obtaining several clustering clusters for each position point on the inspection path; predicting the flight state of the UAV during the actual inspection according to the several clustering clusters for each position point on the inspection path, and further predicting the trajectory of the UAV.

2. The real-time trajectory prediction method of an unmanned aerial vehicle based on low-altitude economy according to claim 1, characterized in that, The obtaining of the attitude feature vector of the UAV at each position point during each historical inspection includes: ； Wherein, represents the attitude feature vector of the UAV at the j-th position point during the i-th historical inspection; represents the pitch angle of the UAV at the j-th position point during the i-th historical inspection; represents the roll angle of the UAV at the j-th position point during the i-th historical inspection; represents the yaw angle of the UAV at the j-th position point during the i-th historical inspection; represents the gray information entropy of the image collected by the UAV at the j-th position point during the i-th historical inspection.

3. A real-time trajectory prediction method for drones based on low-altitude economy according to claim 1, characterized in that, The obtaining of each historical position point corresponding to each position point on the inspection path and the first neighboring position point and the second neighboring position point of each historical position point includes: Obtaining the distance between the k-th position point on the inspection path and each position point during any historical inspection, and recording the position point during the historical inspection corresponding to the minimum distance as a historical position point corresponding to the k-th position point on the inspection path, and recording the adjacent position points before and after the historical position point during its historical inspection as the first neighboring position point and the second neighboring position point of the historical position point corresponding to the k-th position point on the inspection path.

4. A real-time trajectory prediction method for drones based on low-altitude economy according to claim 1, characterized in that, The obtaining of the attitude similarity degree of each historical position point corresponding to each position point on the inspection path includes: In the formula, represents the attitude similarity degree of the h-th historical position point corresponding to the k-th position point in the inspection path; represents the attitude feature vector of the h-th historical position point corresponding to the k-th position point in the inspection path; represents the attitude feature vector of the first neighboring position point of the h-th historical position point corresponding to the k-th position point in the inspection path; represents the attitude feature vector of the second neighboring position point of the h-th historical position point corresponding to the k-th position point in the inspection path; represents the normalized cosine similarity function.

5. A real-time trajectory prediction method for drones based on low-altitude economy according to claim 1, characterized in that, The obtaining of the attitude complexity of each position point on the inspection path includes: Taking the standard deviation of the attitude similarity degrees of all historical position points corresponding to each position point on the inspection path as the attitude complexity of each position point on the inspection path.

6. A real-time trajectory prediction method for drones based on low-altitude economy according to claim 1, characterized in that, The obtaining of each group of flight data vectors corresponding to each position point on each inspection path includes: Taking the longitude, latitude, horizontal velocity, vertical velocity, x-axis acceleration, y-axis acceleration, z-axis acceleration, pitch angle, roll angle, yaw angle and altitude of the UAV when flying to each historical position point corresponding to each position point on the inspection path as each flight data vector corresponding to each position point on the inspection path; Combining the flight data vectors corresponding to each position point on the inspection path in pairs to obtain each group of flight data vectors corresponding to each position point on the inspection path.

7. A real-time trajectory prediction method for drones based on low-altitude economy according to claim 1, characterized in that, The obtaining of the distance between each group of flight data vectors corresponding to each position point on the inspection path includes: In the formula, represents the distance between the o-th group of flight data vectors corresponding to the k-th position point in the inspection path; represents the complexity of the attitude at the k-th position point in the inspection path; represents the number of elements in each flight data vector corresponding to each position point in the inspection path; and represents the value of the m-th element of the first flight data vector and the second flight data vector in the o-th group of flight data vectors corresponding to the k-th position point in the inspection path; represents the information entropy of the m-th element in all flight data vectors corresponding to the k-th position point in the inspection path.

8. A real-time trajectory prediction method for drones based on low-altitude economy according to claim 1, characterized in that Obtaining several clustering clusters for each position point in the inspection path includes: Presetting a neighborhood radius ε and the minimum number of points minpts within the neighborhood radius. Based on the distances between each group of historical position points corresponding to each position point in the inspection path, use DBSCAN clustering to cluster the flight data vectors corresponding to each position point in the inspection path, and obtain several clustering clusters for each position point in the inspection path.

9. A real-time trajectory prediction method for drones based on low-altitude economy according to claim 1, characterized in that Predicting the flight state of the UAV during actual inspection according to several clustering clusters for each position point in the inspection path includes: During the actual inspection of the UAV, obtain the flight data vector of the current position point of the UAV; obtain the distances between the current position point of the UAV and each position point in the inspection path, and record the position point in the inspection path corresponding to the minimum distance as the target position point; According to the flight data vector of the current position point of the UAV, use the KNN algorithm with K = 5 in the clustering cluster of the target position point to obtain the clustering cluster to which the flight data vector of the current position point of the UAV belongs; input the clustering cluster to which the flight data vector of the current position point of the UAV belongs into the Markov model to predict the flight state of the next position point of the UAV.

10. A real-time trajectory prediction system for drones based on low-altitude economy, characterized in that, Including: A processor and a memory, where the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a real-time trajectory prediction method for an unmanned aerial vehicle based on the low-altitude economy according to any one of claims 1-9 is implemented.

Images