Real-time trajectory prediction system and method for unmanned aerial vehicle

Abstract

The invention discloses an unmanned aerial vehicle real-time trajectory prediction system and method, and relates to the technical field of unmanned aerial vehicle trajectory prediction, and the system comprises a rotor rotating speed monitoring module, a flight attitude monitoring module, an unmanned aerial vehicle state judgment module, a control information acquisition module, a flight attitude comparison module, an attitude calibration module, and a trajectory prediction module. The rotor rotating speed monitoring module is used for monitoring the rotating speed of each rotor of the unmanned aerial vehicle to obtain the corresponding rotor rotating speed; according to the real-time trajectory prediction system and method for the unmanned aerial vehicle, whether the unmanned aerial vehicle and the flight attitude thereof are abnormal or not is judged according to the relationship among the rotor rotating speed, the flight attitude data and the control information of the unmanned aerial vehicle, and the rotor rotating speed and the flight attitude of the unmanned aerial vehicle can be adjusted when the unmanned aerial vehicle or the flight attitude thereof is abnormal; the influence of interference factors on the flight attitude is reduced, and the trajectory prediction difficulty is reduced, so that the trajectory prediction precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle trajectory prediction, and in particular to a system and method for real-time unmanned aerial vehicle trajectory prediction. Background Art

[0002] With the development of science and technology, the application fields of drones are becoming more and more extensive, such as agriculture: used for precision agriculture, monitoring crop growth, fertilization and spraying pesticides; surveying and geographic information: conducting topographic surveying, land management, urban planning, etc.; logistics and transportation: used for express delivery, medical supplies transportation, etc. to improve logistics efficiency; environmental monitoring: monitoring environmental changes such as air quality, water pollution, and forest fires; disaster relief: providing real-time monitoring and search and rescue support in natural disasters; building and construction: monitoring construction sites, progress inspections and safety assessments; entertainment and photography: used for aerial photography, videography and entertainment activities to create unique visual effects; military: used for reconnaissance, surveillance and attacking targets.

[0003] In order to ensure the quality of the drone when performing the task, it is necessary to predict the flight trajectory of the drone to ensure that the trajectory of the drone is on the predetermined trajectory and can successfully complete the task. The prior art with the publication number CN116627154A discloses a method for guiding the landing of a drone based on posture prediction and trajectory optimization and a drone. The specific steps of the method in the close-range landing process are as follows: when the landing angle of the drone reaches the landing threshold during the medium-distance tracking process, it enters the pre-landing state; obtains the motion state of the current target landing object, and predicts the motion state of the target landing object through the Kalman filter algorithm; compares the predicted motion process of the target landing object with the set yaw motion angle to determine whether the motion state of the target landing object is stable and whether the drone enters the landing state; after entering the landing state, the landing angle is kept unchanged for landing, and the initial landing trajectory is constructed according to the current posture and dynamic structure; the initial landing curve is optimized by the Bezier curve to find the optimal landing trajectory; the drone lands on the target landing object based on the optimal landing trajectory. The stability and accuracy of the multi-rotor drone in visually guided landing are improved.

[0004] However, when a drone is performing a mission, one or more rotors may be damaged, causing the lift provided by the damaged rotor to change under the same circumstances, thereby changing the flight trajectory of the drone. In addition, the change in the trajectory of the drone caused by the damage to the drone will also affect the prediction accuracy of the trajectory prediction algorithm and the accuracy of the drone trajectory control. Summary of the invention

[0005] The purpose of the present invention is to provide a real-time trajectory prediction system and method for unmanned aerial vehicles to solve the above-mentioned deficiencies in the prior art.

[0006] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: a real-time trajectory prediction system for unmanned aerial vehicles, comprising a rotor speed monitoring module, a flight attitude monitoring module, a UAV state judgment module, a control information acquisition module, a flight attitude comparison module, an attitude calibration module, and a trajectory prediction module; The rotor speed monitoring module is used to monitor the rotation speed of each rotor of the drone respectively to obtain the corresponding rotor speed, wherein the rotation speed of each rotor can be monitored by setting a speed sensor on each rotor of the drone; The flight attitude monitoring module is used to monitor the flight attitude of the drone and obtain the flight attitude data of the drone, wherein the flight attitude data includes pitch angle, yaw angle, roll angle, speed, acceleration, height, position, and attitude angular velocity, wherein the pitch angle is the angle at which the drone tilts forward and backward, affecting its up and down rise; the yaw angle is the angle at which the drone rotates left and right, affecting its heading; the roll angle is the angle at which the drone tilts sideways, affecting its lateral movement; the frontal angular velocity of the drone can be measured by a gyroscope to calculate the pitch angle, yaw angle, and roll angle; the speed includes horizontal speed and vertical speed, indicating the flight speed of the drone; the acceleration includes acceleration data in all directions, which is helpful to evaluate the dynamic changes of the drone, and the acceleration can be measured by acceleration, and the speed and position of the drone can be inferred by the change in acceleration; the height is the height information relative to the ground, usually obtained by a pressure sensor or GPS; the position is GPS positioning data, including longitude and latitude and relative altitude; the attitude angular velocity represents the angular velocity of pitch, yaw, and roll, which is helpful to dynamically adjust the flight attitude; The drone state judgment module is used to judge whether the drone flight attitude matches the rotor speed of each rotor based on the rotor speed and flight attitude data of each rotor; The control information acquisition module is used to acquire control information for controlling the drone; The flight attitude comparison module is used to determine whether the flight attitude of the UAV matches the control information based on the comparison between the flight attitude data and the control information; The attitude calibration module is used to re-match the flight attitude with the rotation speed of each rotor when the judgment of the flight attitude comparison module is no, so that the flight attitude matches the control information; The trajectory prediction module is used to predict the flight trajectory of the UAV based on the flight attitude data to obtain predicted trajectory coordinates after the attitude calibration module rematches the flight attitude with the rotation speed of each rotor. Predicting the flight trajectory based on the flight attitude data is common knowledge in the prior art and is directly applied here without any changes. Therefore, it is not specifically described in this technical solution, and will not cause any trouble to this technical solution based on the field. For example, a recurrent neural network (RNN) and a long short-term memory network (LSTM) can be used.

[0007] Furthermore, the drone state judgment module is used to judge whether the drone flight attitude matches the rotor speed of each rotor based on the rotor speed and flight attitude data of each rotor, including the following steps: Obtain the historical flight attitude data of the UAV when it is flying without external interference and the corresponding rotor speed of each rotor; Based on the historical flight attitude data and the corresponding rotor speeds of each rotor, regression analysis is performed to obtain the attitude speed relationship formula between the flight attitude data and each rotor speed, and the rotor attitude relationship of the flight attitude change corresponding to the change of each rotor speed. For example, the UAV has four rotors, namely rotor No. 1, rotor No. 2, rotor No. 3 and rotor No. 4 at the four corners of the rectangle, wherein rotor No. 1 is diagonally arranged with rotor No. 3, and rotor No. 2 is diagonally arranged with rotor No. 4. The rotor attitude relationship includes synchronously reducing the speeds of rotor No. 1 and rotor No. 3, and the UAV rotates around the central vertical line toward the transfer direction of No. 2; Obtain the current rotor speed and flight attitude data of the UAV, substitute the rotor speed into the attitude speed relationship formula, and obtain theoretical flight attitude data; Determining whether the theoretical flight attitude data matches the current flight attitude data; If so, the flight attitude is normal; if not, the flight attitude is the first abnormality, indicating that there are abnormal reasons interfering with the current flight, where the abnormal reasons may include rotor damage affecting rotor lift, fuselage affecting air resistance, encountering external wind influence, etc.

[0008] Further, the flight attitude comparison module is used to determine whether the flight attitude of the UAV matches the control information based on the comparison between the flight attitude data and the control information, including the following steps: Acquire historical flight attitude data and corresponding control information, and associate the flight attitude data with the corresponding control information; Get current flight attitude data and control information; Determine whether the current flight attitude data and control information data are related; If yes, the flight attitude is normal; if no, the flight attitude is the second abnormality.

[0009] Further, the attitude calibration module is used to re-match the flight attitude with each rotor speed when the flight attitude comparison module determines that the flight attitude is negative, so that the flight attitude matches the control information, including the following steps: Determine whether the flight attitude is the second abnormality; If yes, then obtain the flight attitude data associated with the current control information to obtain the target flight attitude data; Calculating the difference between the current flight attitude data and the target flight attitude data to obtain the flight attitude data to be adjusted; Obtaining a required flight attitude change based on the flight attitude data to be adjusted, and obtaining a required rotor speed change based on the rotor attitude relationship; Adjusting the corresponding rotor speed based on the required rotor speed change until the current flight attitude data is the same as the target flight attitude data; Substituting the flight attitude data at this time into the attitude speed relationship formula, the corresponding rotor speed is obtained, which is marked as the theoretical rotor speed; The actual rotor speed is obtained by obtaining the rotation speed of each rotor at this time, and the difference between the actual rotor speed and the corresponding theoretical rotor speed is calculated to obtain the corrected rotor speed. The corrected rotor speed is then associated with the flight attitude data at this time to obtain the corrected rotor speed flight attitude relationship.

[0010] Furthermore, the system also includes a flight attitude abnormality cause judgment module, the flight attitude abnormality cause judgment module is used to mark the abnormal cause corresponding to each rotor correction speed flight attitude relationship, based on the rotor correction speed flight attitude relationship and the corresponding abnormal cause, train the machine learning model to obtain the abnormal cause prediction model, the abnormal cause prediction model is used to output the predicted abnormal cause based on the input rotor correction speed flight attitude relationship, including the following steps: Get the rotor correction speed flight attitude relationship of the set continuous duration; Set multiple exception reason labels corresponding to the exception reasons; Associating the rotor corrected speed flight attitude relationship of each set continuous duration with the corresponding abnormal reason label; Based on the rotor-corrected speed-flight attitude relationship of each set continuous duration and the corresponding abnormal cause label, a machine learning model is trained to obtain an abnormal cause prediction model. The present invention does not limit the specific machine learning model, and can be selected according to the needs during training, such as support vector machine (SVM), decision tree, random forest, convolutional neural network (CNN) and recurrent neural network (RNN), etc.; and during training, the rotor-corrected speed and its corresponding flight attitude data in the rotor-corrected speed-flight attitude relationship of the set continuous duration are used as input, and after extracting the features of the input data, the features of the input data are output in the hidden layer, and the features of the output data are associated with the corresponding abnormal cause label, and then the abnormal cause label is used as the output of the model; The most recent rotor correction speed flight attitude relationship with a long continuous time is obtained, and the abnormal cause prediction model is input to obtain a predicted abnormal cause label.

[0011] Furthermore, the abnormal cause label includes an internal abnormality label and an external force abnormality label. The abnormal cause corresponding to the internal abnormality label is the abnormality caused by damage to the drone itself, such as rotor damage, fuselage damage, etc. The abnormal cause corresponding to the external force abnormality label is the abnormality caused by external force influence, such as wind influence.

[0012] Furthermore, the flight attitude abnormality cause judgment module is also used to remind the user when the output abnormality cause is a senior level abnormality label. Specifically, the user can be reminded by popping up a window or sending a message.

[0013] A method for predicting the real-time trajectory of a UAV comprises the following steps: Monitor and obtain the flight attitude data, rotor speed and control information of the UAV; Regression analysis is performed on the flight attitude data and the rotor speeds of the UAV without external interference to obtain the attitude speed relationship formula between the flight attitude data and the rotor speeds, and the rotor attitude relationship of the flight attitude change corresponding to the change of the rotor speed; Matching flight attitude data with control information; Determine whether the control information matches the current flight attitude data: If not, then according to the rotor attitude relationship of the flight attitude change corresponding to each rotor speed change, the corresponding rotor speed is adjusted until the control information matches the current flight attitude data; Based on the matched flight attitude data, the flight trajectory of the UAV is predicted and the predicted trajectory coordinates are obtained.

[0014] 1. Compared with the prior art, the real-time trajectory prediction system and method of a UAV provided by the present invention can determine whether the UAV and its flight attitude are abnormal based on the relationship between the UAV rotor speed, flight attitude data and control information by setting a rotor speed monitoring module, a flight attitude monitoring module, a UAV state judgment module, a control information acquisition module and a flight attitude comparison module.

[0015] 2. Compared with the prior art, the real-time trajectory prediction system and method for a UAV provided by the present invention can adjust the rotor speed and flight attitude of the UAV or its flight attitude when it is abnormal by setting an attitude calibration module and a trajectory prediction module, thereby reducing the impact of interference factors on the flight attitude and reducing the difficulty of trajectory prediction, thereby improving the trajectory prediction accuracy.

[0016] 3. Compared with the prior art, the real-time trajectory prediction system and method of a UAV provided by the present invention can also predict the cause of the abnormal flight attitude of the UAV by setting a flight attitude abnormality cause judgment module, by analyzing the rotor speed and flight attitude data of the UAV before and after flight attitude correction, and remind the user when the abnormal cause is self-damage, so as to maintain the UAV in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0018] Figure 1 A system structure block diagram provided for an embodiment of the present invention; Figure 2 A diagram of the method steps provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0020] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0021] Example embodiments will be described more fully below with reference to the accompanying drawings, but the example embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. On the contrary, the purpose of providing these embodiments is to make the present disclosure thorough and complete and to enable those skilled in the art to fully understand the scope of the present disclosure.

[0022] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.

[0023] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0024] See also Figure 1 - Figure, a UAV real-time trajectory prediction system, including a rotor speed monitoring module, a flight attitude monitoring module, a UAV state judgment module, a control information acquisition module, a flight attitude comparison module, an attitude calibration module, and a trajectory prediction module; The rotor speed monitoring module is used to monitor the rotation speed of each rotor of the drone respectively to obtain the corresponding rotor speed, wherein the rotation speed of each rotor can be monitored by setting a speed sensor on each rotor of the drone; The flight attitude monitoring module is used to monitor the flight attitude of the drone and obtain the flight attitude data of the drone, wherein the flight attitude data includes pitch angle, yaw angle, roll angle, speed, acceleration, altitude, position, and attitude angular velocity, wherein the pitch angle is the angle at which the drone tilts forward and backward, affecting its up and down rise; the yaw angle is the angle at which the drone rotates left and right, affecting its heading; the roll angle is the angle at which the drone tilts sideways, affecting its lateral movement; the angular velocity of the drone can be measured by a gyroscope to calculate the pitch angle, yaw angle, and roll angle; the speed includes horizontal speed and vertical speed, indicating the flight speed of the drone; the acceleration includes acceleration data in all directions, which helps to evaluate the dynamic changes of the drone, and the acceleration can be measured by acceleration, and the speed and position of the drone can be inferred by the change in acceleration; the altitude is the height information relative to the ground, usually obtained by a pressure sensor or GPS; the position is GPS positioning data, including longitude and latitude and relative altitude; the attitude angular velocity represents the angular velocity of pitch, yaw, and roll, which helps to dynamically adjust the flight attitude; The UAV state judgment module is used to judge whether the UAV flight attitude matches the rotor speed of each rotor based on the rotor speed and flight attitude data of each rotor, including the following steps: A1. Obtain historical flight attitude data of the UAV when it is flying without external interference and the corresponding rotor speed of each rotor; A2. Based on the historical flight attitude data and the corresponding rotor speeds of each rotor, regression analysis is performed to obtain the attitude speed relationship formula between the flight attitude data and each rotor speed, and the rotor attitude relationship of the flight attitude change corresponding to the change of each rotor speed. For example, the UAV has four rotors, namely rotor No. 1, rotor No. 2, rotor No. 3 and rotor No. 4 at the four corners of the rectangle, wherein rotor No. 1 is diagonally arranged with rotor No. 3, and rotor No. 2 is diagonally arranged with rotor No. 4. The rotor attitude relationship includes synchronously reducing the speeds of rotor No. 1 and rotor No. 3, and the UAV rotates around the central vertical line toward the transfer direction of No. 2; A3. Obtain the current rotor speed and flight attitude data of the UAV, substitute the rotor speed into the attitude speed relationship formula, and obtain the theoretical flight attitude data; A4. Determine whether the theoretical flight attitude data matches the current flight attitude data; A5. If yes, the flight attitude is normal; if not, the flight attitude is abnormal, indicating that the current flight is disturbed by abnormal reasons, where the abnormal reasons may include rotor damage affecting rotor lift, fuselage affecting air resistance, and encountering external wind influence, etc.

[0025] The control information acquisition module is used to obtain control information for controlling the drone; The flight attitude comparison module is used to determine whether the flight attitude of the drone matches the control information based on the comparison between the flight attitude data and the control information, including the following steps: B1. Acquire historical flight attitude data and corresponding control information, and associate the flight attitude data with the corresponding control information; B2. Obtain current flight attitude data and control information; B3, determine whether the current flight attitude data and control information data are related; B4. If yes, the flight attitude is normal; if no, the flight attitude is the second abnormality.

[0026] The attitude calibration module is used to re-match the flight attitude with the rotation speed of each rotor when the flight attitude comparison module determines that the flight attitude matches the control information, including the following steps: C1. Determine whether the flight attitude is the second abnormality; C2. If yes, obtain the flight attitude data associated with the current control information to obtain the target flight attitude data; C3. Calculate the difference between the current flight attitude data and the target flight attitude data to obtain the flight attitude data to be adjusted; C4. Obtaining the required flight attitude change based on the flight attitude data to be adjusted, and obtaining the required rotor speed change according to the rotor attitude relationship; C5. adjusting the corresponding rotor speed based on the required rotor speed change until the current flight attitude data is the same as the target flight attitude data; C6. Substitute the flight attitude data at this time into the attitude speed relationship formula to obtain the corresponding rotor speed, which is marked as the theoretical rotor speed; C7. Obtain the actual rotor speed at each rotor speed at this time, calculate the difference between the actual rotor speed and the corresponding theoretical rotor speed, obtain the rotor correction speed, and associate the rotor correction speed with the flight attitude data at this time to obtain the rotor correction speed flight attitude relationship The trajectory prediction module is used to predict the flight trajectory of the UAV based on the flight attitude data after the attitude calibration module rematches the flight attitude with the rotation speed of each rotor, and obtain the predicted trajectory coordinates. The flight trajectory prediction based on the flight attitude data is common knowledge in the prior art and is directly applied here without any changes. Therefore, it is not described in detail in this technical solution, and it will not cause any trouble to this technical solution based on the field. For example, recurrent neural networks (RNN) and long short-term memory networks (LSTM) can be used.

[0027] The system also includes a flight attitude abnormality cause judgment module, which is used to mark the abnormal causes corresponding to the rotor corrected speed flight attitude relationship, and train the machine learning model based on the rotor corrected speed flight attitude relationship and the corresponding abnormal causes to obtain an abnormal cause prediction model. The abnormal cause prediction model is used to output the predicted abnormal cause based on the input rotor corrected speed flight attitude relationship, and includes the following steps: D1. Obtain the rotor correction speed and flight attitude relationship of the set continuous time; D2. Set multiple abnormal cause labels corresponding to the abnormal causes. Furthermore, the abnormal cause labels include self-abnormal labels and external force abnormal labels. The abnormal causes corresponding to the self-abnormal labels are abnormalities caused by damage to the drone itself, such as rotor damage, fuselage damage, etc. The abnormal causes corresponding to the external force abnormal labels are abnormalities caused by external force influences, such as wind influences.

[0028] D3, associating the rotor correction speed flight attitude relationship of each set continuous time with the corresponding abnormal reason label; D4. Based on the rotor corrected speed-flight attitude relationship for each set continuous duration and the corresponding abnormal cause label, a machine learning model is trained to obtain an abnormal cause prediction model. The present invention does not limit the specific machine learning model, which can be selected according to the needs during training, such as support vector machine (SVM), decision tree, random forest, convolutional neural network (CNN) and recurrent neural network (RNN), etc.; and during training, the rotor corrected speed and its corresponding flight attitude data in the rotor corrected speed-flight attitude relationship for each set continuous duration are used as input, and after extracting the features of the input data, the features of the input data are output in the hidden layer, and the features of the output data are associated with the corresponding abnormal cause label, and then the abnormal cause label is used as the output of the model.

[0029] D5. Obtain the most recent rotor correction speed flight attitude relationship with the longest continuous time, input it into the abnormal cause prediction model, and obtain the predicted abnormal cause label; D6. The flight attitude abnormality cause judgment module is also used to remind the user when the output abnormality cause is a senior abnormality label. Specifically, the user can be reminded by popping up a window or sending a message.

[0030] A method for predicting the real-time trajectory of a UAV comprises the following steps: Monitor and obtain the flight attitude data, rotor speed and control information of the UAV; Regression analysis is performed on the flight attitude data and the rotor speeds of the UAV without external interference to obtain the attitude speed relationship formula between the flight attitude data and the rotor speeds, and the rotor attitude relationship of the flight attitude change corresponding to the change of the rotor speed; Matching flight attitude data with control information; Determine whether the control information matches the current flight attitude data: If not, then according to the rotor attitude relationship of the flight attitude change corresponding to each rotor speed change, the corresponding rotor speed is adjusted until the control information matches the current flight attitude data; Based on the matched flight attitude data, the flight trajectory of the UAV is predicted and the predicted trajectory coordinates are obtained.

[0031] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A real-time trajectory prediction system for unmanned aerial vehicles, characterized by: It includes rotor speed monitoring module, flight attitude monitoring module, UAV state judgment module, control information acquisition module, flight attitude comparison module, attitude calibration module, and trajectory prediction module; The rotor speed monitoring module is used to monitor the rotation speed of each rotor of the UAV respectively to obtain the corresponding rotor speed; The flight attitude monitoring module is used to monitor the flight attitude of the UAV and obtain the flight attitude data of the UAV; The drone state judgment module is used to judge whether the drone flight attitude matches the rotor speed of each rotor based on the rotor speed and flight attitude data of each rotor; The control information acquisition module is used to acquire control information for controlling the drone; The flight attitude comparison module is used to determine whether the flight attitude of the UAV matches the control information based on the comparison between the flight attitude data and the control information; The attitude calibration module is used to re-match the flight attitude with the rotation speed of each rotor when the judgment of the flight attitude comparison module is no, so that the flight attitude matches the control information; The trajectory prediction module is used to predict the flight trajectory of the UAV based on the flight attitude data after the attitude calibration module re-matches the flight attitude with the rotation speed of each rotor to obtain the predicted trajectory coordinates.

2. The real-time trajectory prediction system for unmanned aerial vehicles according to claim 1, characterized in that: The drone state judgment module is used to judge whether the drone flight attitude matches the rotor speed of each rotor based on the rotor speed and flight attitude data of each rotor, and includes the following steps: Obtain the historical flight attitude data of the UAV when it is flying without external interference and the corresponding rotor speed of each rotor; Based on the historical flight attitude data and the corresponding rotor speeds of each rotor, regression analysis is performed to obtain the attitude speed relationship formula between the flight attitude data and each rotor speed, and the rotor attitude relationship between the flight attitude change corresponding to the change of each rotor speed; Obtain the current rotor speed and flight attitude data of the UAV, substitute the rotor speed into the attitude speed relationship formula, and obtain theoretical flight attitude data; Determining whether the theoretical flight attitude data matches the current flight attitude data; If yes, the flight attitude is normal; if no, the flight attitude is abnormal.

3. The real-time trajectory prediction system of a UAV according to claim 2, characterized in that: The flight attitude comparison module is used to determine whether the flight attitude of the UAV matches the control information based on the comparison between the flight attitude data and the control information, and includes the following steps: Acquire historical flight attitude data and corresponding control information, and associate the flight attitude data with the corresponding control information; Get current flight attitude data and control information; Determine whether the current flight attitude data and control information data are related; If yes, the flight attitude is normal; if no, the flight attitude is the second abnormality.

4. The real-time trajectory prediction system for unmanned aerial vehicles according to claim 3, characterized in that: The attitude calibration module is used to re-match the flight attitude with the rotation speed of each rotor when the flight attitude comparison module determines that the flight attitude is negative, so that the flight attitude matches the control information, and includes the following steps: Determine whether the flight attitude is the second abnormality; If yes, then obtain the flight attitude data associated with the current control information to obtain the target flight attitude data; Calculating the difference between the current flight attitude data and the target flight attitude data to obtain the flight attitude data to be adjusted; Obtaining a required flight attitude change based on the flight attitude data to be adjusted, and obtaining a required rotor speed change based on the rotor attitude relationship; Adjusting the corresponding rotor speed based on the required rotor speed change until the current flight attitude data is the same as the target flight attitude data; Substituting the flight attitude data at this time into the attitude speed relationship formula, the corresponding rotor speed is obtained, which is marked as the theoretical rotor speed; The actual rotor speed is obtained by obtaining the rotation speed of each rotor at this time, and the difference between the actual rotor speed and the corresponding theoretical rotor speed is calculated to obtain the corrected rotor speed. The corrected rotor speed is then associated with the flight attitude data at this time to obtain the corrected rotor speed flight attitude relationship.

5. The real-time trajectory prediction system for unmanned aerial vehicles according to claim 4, characterized in that: The system further includes a flight attitude abnormality cause judgment module, the flight attitude abnormality cause judgment module is used to mark the abnormal cause corresponding to each rotor corrected speed flight attitude relationship, based on the rotor corrected speed flight attitude relationship and the corresponding abnormal cause, train a machine learning model to obtain an abnormal cause prediction model, the abnormal cause prediction model is used to output the predicted abnormal cause based on the input rotor corrected speed flight attitude relationship, including the following steps: Get the rotor correction speed flight attitude relationship of the set continuous duration; Set multiple exception reason labels corresponding to the exception reasons; Associating the rotor corrected speed flight attitude relationship of each set continuous duration with the corresponding abnormal reason label; Based on the relationship between the rotor corrected speed and flight attitude for each set continuous duration and the corresponding abnormal cause label, a machine learning model is trained to obtain an abnormal cause prediction model; The most recent rotor correction speed flight attitude relationship with a long continuous time is obtained, and the abnormal cause prediction model is input to obtain a predicted abnormal cause label.

6. The real-time trajectory prediction system for unmanned aerial vehicles according to claim 5, characterized in that: The abnormal cause label includes an own abnormal label and an external force abnormal label. The abnormal cause corresponding to the own abnormal label is the abnormality caused by the damage of the drone itself, and the abnormal cause corresponding to the external force abnormal label is the abnormality caused by the influence of external force.

7. The real-time trajectory prediction system for unmanned aerial vehicles according to claim 6, characterized in that: The flight attitude abnormality cause judgment module is also used to issue a reminder to the user when the output abnormality cause is a senior level abnormality label.

8. A method for predicting the real-time trajectory of an unmanned aerial vehicle, applicable to a real-time trajectory prediction system for an unmanned aerial vehicle according to any one of claims 1 to 8, characterized in that: The following steps are involved: Monitor and obtain the UAV's flight attitude data, rotor speed and control information; Regression analysis is performed on the flight attitude data and the rotor speeds of the UAV without external interference to obtain the attitude speed relationship formula between the flight attitude data and the rotor speeds, and the rotor attitude relationship of the flight attitude change corresponding to the change of the rotor speed; Matching flight attitude data with control information; Determine whether the control information matches the current flight attitude data: If not, then according to the rotor attitude relationship of the flight attitude change corresponding to the rotor speed change, the corresponding rotor speed is adjusted until the control information matches the current flight attitude data; Based on the matched flight attitude data, the flight trajectory of the UAV is predicted and the predicted trajectory coordinates are obtained.

Images