A flight data acquisition method and system for hydrogen energy unmanned aerial vehicle

Abstract

The present application relates to the technical field of flight data collection, in particular to a flight data collection method and system for hydrogen energy unmanned aerial vehicle. The present application first collects flight data of each hydrogen energy unmanned aerial vehicle at each historical time; then determines all yaw time periods of each hydrogen energy unmanned aerial vehicle within a preset historical time period, and obtains a yaw risk coefficient in each yaw time period and a yaw attention coefficient of each historical yaw time; further, in combination with a time interval between each historical time and the to-be-adjusted time, the present application predicts a future yaw distance of each hydrogen energy unmanned aerial vehicle to determine a collection frequency of flight data of the hydrogen energy unmanned aerial vehicle within a preset future time period and to collect the flight data. The present application comprehensively analyzes flight dynamics, flight space and environment of the hydrogen energy unmanned aerial vehicle during flight, evaluates flight risk, predicts future yaw dynamics to adjust the collection frequency, reduces the risk of failing to timely discover the yaw of the unmanned aerial vehicle, and improves the flight data collection effect of the hydrogen energy unmanned aerial vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of flight data collection, in particular to a flight data collection method and system for hydrogen energy unmanned aerial vehicle. BACKGROUND

[0002] The hydrogen energy unmanned aerial vehicle is an unmanned aerial vehicle using hydrogen fuel cell as power source. Unlike the traditional battery-powered unmanned aerial vehicle, its core advantage is its long flight time, so it is very suitable for large-scale unmanned aerial vehicle light shows and the like. During the flight performance of the hydrogen energy unmanned aerial vehicle, the flight data thereof usually needs to be collected and monitored to ensure that each unmanned aerial vehicle can fly along an accurate preset route and coordinate actions.

[0003] At present, the flight data of the hydrogen energy unmanned aerial vehicle is usually collected at a fixed collection frequency. However, during the flight performance, due to the uncontrollable outdoor environment, the hydrogen energy unmanned aerial vehicle may be disturbed by wind speed and direction during flight, which may cause deviation or collision and other flight accidents. The fixed collection frequency may not be able to capture the real-time position or attitude change of the hydrogen energy unmanned aerial vehicle in time, resulting in poor flight data collection effect, and thus the deviation of the unmanned aerial vehicle cannot be found and intervened in time. SUMMARY

[0004] In order to solve the technical problem of poor flight data collection effect of the existing hydrogen energy unmanned aerial vehicle, the purpose of the present application is to provide a flight data collection method and system for hydrogen energy unmanned aerial vehicle, and the technical solution adopted is as follows:

[0005] A flight data collection method for hydrogen energy unmanned aerial vehicle, the method comprising:

[0006] Collecting flight data of each hydrogen energy unmanned aerial vehicle at each historical moment within a preset historical period at a tuning moment, the tuning moment being the end moment of the preset historical period, the flight data at least including flight position, flight speed, heading, flight environment information and obstacle detection information;

[0007] For each hydrogen energy unmanned aerial vehicle, within the preset historical period, the deviation distance of each historical moment relative to the preset planning position is determined to determine all deviation periods, and the obstacle detection information and flight speed of each historical moment within each deviation period are combined to obtain the deviation risk coefficient of the hydrogen energy unmanned aerial vehicle within each deviation period; at each historical moment within each deviation period, the deviation attention coefficient of the hydrogen energy unmanned aerial vehicle is obtained according to the complexity of the flight planning route of the hydrogen energy unmanned aerial vehicle and the deviation risk coefficient thereof within the deviation period, combined with the heading and flight environment information thereof;

[0008] At the time to be adjusted, for each hydrogen energy unmanned aerial vehicle, a future yaw distance of the hydrogen energy unmanned aerial vehicle in a preset future period is predicted according to the yaw attention coefficient of the hydrogen energy unmanned aerial vehicle at each historical time and a time interval between the historical time and the time to be adjusted, and the acquisition frequency of flight data of the hydrogen energy unmanned aerial vehicle in the preset future period is determined based on the future yaw distance and the flight data is acquired.

[0009] Further, the flight environment information at least includes wind direction and wind speed; and the obstacle detection information at least includes the number of detected obstacles in a preset range centered on the hydrogen energy unmanned aerial vehicle and the spatial detection distance between the hydrogen energy unmanned aerial vehicle and each obstacle.

[0010] Further, the method for obtaining the yaw period comprises:

[0011] For each hydrogen energy unmanned aerial vehicle, a historical time with a yaw distance greater than a preset distance threshold is taken as a yaw time, and a corresponding period of continuous yaw times in a preset historical period is taken as a yaw period.

[0012] Further, the method for obtaining the yaw risk coefficient comprises:

[0013] In each yaw period, a time sequence of the yaw distance is fitted, and all monotonically increasing sub-sections in the yaw period are obtained, a yaw parameter of the yaw period is obtained according to the length of each monotonically increasing sub-section and the variation amplitude of the yaw distance in the monotonically increasing sub-section;

[0014] At each historical time of each yaw period, a collision parameter is obtained according to the spatial detection distance between the hydrogen energy unmanned aerial vehicle and each obstacle and the total number of obstacles, the collision parameter is weighted by the flight speed of the hydrogen energy unmanned aerial vehicle to obtain a risk sub-parameter, and a risk parameter of the yaw period is obtained according to the concentration characteristics of the risk sub-parameters in the yaw period;

[0015] The yaw risk coefficient of the hydrogen energy unmanned aerial vehicle in the corresponding yaw period is obtained by fusing the yaw parameter and the risk parameter.

[0016] Further, the method for obtaining the yaw parameter comprises:

[0017] The length proportion of all monotonically increasing sub-sections in the corresponding time sequence is taken as a first yaw factor, in each monotonically increasing sub-section, the length of the sub-section is taken as a yaw duration weight, and the range of the yaw distance in the sub-section is taken as a yaw amplitude parameter, the yaw duration weight is weighted to the yaw amplitude parameter, and the weighted sum result of all monotonically increasing sub-sections is taken as a second yaw factor, and the first yaw factor and the second yaw factor are fused to obtain the yaw parameter.

[0018] Further, the method for obtaining the yaw attention coefficient comprises:

[0019] At each historical moment in each drift period, a resistance factor is obtained according to the minimum angle between the heading direction of the hydrogen energy unmanned aerial vehicle and the wind direction, the flight resistance coefficient is obtained by weighting the wind speed on the resistance factor, a future flight period of a preset time length is determined along the time sequence direction starting from the historical moment, and a flight complexity coefficient of the hydrogen energy unmanned aerial vehicle is obtained according to the angle change between the planned headings of adjacent moments in the flight planning route of the hydrogen energy unmanned aerial vehicle in the future flight period.

[0020] The flight resistance coefficient of the hydrogen energy unmanned aerial vehicle at each historical moment, the flight complexity coefficient and the drift risk coefficient of the drift period to which the historical moment belongs are fused, and the fusion result is taken as the drift attention coefficient of the hydrogen energy unmanned aerial vehicle.

[0021] Further, the flight complexity coefficient obtaining method comprises:

[0022] The minimum angle between the planned heading of each moment and the planned heading of the adjacent next moment in the flight planning route of the hydrogen energy unmanned aerial vehicle in the future flight period is taken as the steering amplitude parameter of each moment, the range of the steering amplitude parameter in the future flight period is taken as the steering difficulty factor, the negative correlation mapping result of the time interval between the range corresponding moments is taken as the difficulty weight, the steering difficulty factor is weighted by using the difficulty weight, and the weighted result is taken as the flight complexity coefficient.

[0023] Further, the future drift distance obtaining method comprises:

[0024] In a preset historical period, for each hydrogen energy unmanned aerial vehicle, the negative correlation mapping result of the time interval between each historical moment and the to-be-adjusted moment in the drift period is taken as the reference weight of the corresponding historical moment, the corresponding drift attention coefficient is weighted by using the reference weight, and the attention weight of the corresponding historical moment is obtained; wherein the attention weight of the historical moment in the non-drift period is a preset value.

[0025] Based on the weighted ARIMA algorithm and the attention weight, the drift distance of the hydrogen energy unmanned aerial vehicle in the preset historical period is sequentially predicted, and the mean value of the predicted drift distance in the preset future period is taken as the future drift distance.

[0026] Further, the acquisition frequency obtaining method comprises:

[0027] When the future drift distance is greater than a preset distance threshold, the future drift distance is mapped into [0, 1], and a constant 1 is added to the mapped value to obtain an adjustment weight for weighting the preset acquisition frequency to obtain the acquisition frequency.

[0028] The application discloses a flight data acquisition system for a hydrogen energy unmanned aerial vehicle, and the system comprises a memory, a processor and a computer program stored in the memory and executable on the processor.

[0029] The application has the following advantages:

[0030] The application first acquires flight data of each hydrogen energy unmanned aerial vehicle at each historical moment in a preset historical period at a to-be-adjusted moment, so as to provide a data basis for subsequent deviation analysis; in the preset historical period, a deviation distance of each hydrogen energy unmanned aerial vehicle at each historical moment is analyzed, all deviation periods are determined, and a deviation risk coefficient reflecting a possibility or a risk degree of a deviation accident of the hydrogen energy unmanned aerial vehicle in the deviation period is acquired in combination with obstacle detection information and a flight speed at each historical moment in each deviation period; further, at each historical moment in each deviation period, a deviation attention coefficient of the hydrogen energy unmanned aerial vehicle is acquired according to a complex situation of a flight planning route of the hydrogen energy unmanned aerial vehicle and the deviation risk coefficient in the deviation period, in combination with a heading and flight environment information of the hydrogen energy unmanned aerial vehicle, the deviation attention coefficient not only reflects an accident risk degree of the hydrogen energy unmanned aerial vehicle, but also reflects a necessity of focusing on the hydrogen energy unmanned aerial vehicle, so as to prepare for subsequent adjustment of an acquisition frequency; then, at the to-be-adjusted moment, a future deviation distance of each hydrogen energy unmanned aerial vehicle in a preset future period is predicted according to the deviation attention coefficient of each hydrogen energy unmanned aerial vehicle at each historical moment and a time interval between the historical moment and the to-be-adjusted moment, and finally, the acquisition frequency of flight data of the hydrogen energy unmanned aerial vehicle in the preset future period is determined based on the future deviation distance and the flight data is acquired. The application comprehensively analyzes flight dynamics, flight space and environment of the hydrogen energy unmanned aerial vehicle in a flight process, evaluates a flight risk of the hydrogen energy unmanned aerial vehicle, predicts future deviation dynamics of the hydrogen energy unmanned aerial vehicle, adjusts the acquisition frequency, reduces a situation that the deviation risk of the unmanned aerial vehicle cannot be found in time, and improves flight data acquisition effect of the hydrogen energy unmanned aerial vehicle. BRIEF DESCRIPTION OF DRAWINGS

[0031] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0032] Figure 1 A flow chart of a flight data acquisition method for a hydrogen energy unmanned aerial vehicle provided by an embodiment of the present application;

[0033] Figure 2 A flow chart of a deviation risk coefficient acquisition method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0034] To further clarify the technical means and effects taken by the present application to achieve the intended purpose, the following describes in detail the specific implementation, structure, features and effects of a flight data collection method and system for hydrogen energy unmanned aerial vehicles according to the present application, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0036] The specific scheme of the flight data collection method and system for hydrogen energy unmanned aerial vehicles provided by the present application is described in detail below in conjunction with the accompanying drawings.

[0037] Please refer to Figure 1 which shows a flowchart of a flight data collection method for hydrogen energy unmanned aerial vehicles according to one embodiment of the present application, which specifically includes:

[0038] Step S1, collect the flight data of each hydrogen energy unmanned aerial vehicle at each historical time within a preset historical period at a tuning time, the tuning time being the end time of the preset historical period, and the flight data including at least flight position, flight speed, heading, flight environment information and obstacle detection information.

[0039] It should be noted that the implementation scenario targeted by the present application is that when hydrogen energy unmanned aerial vehicles are flying in formation for a performance, their flight data is analyzed in real time and the collection frequency is determined to assist relevant personnel in discovering abnormal flight dynamics and adjusting flight instructions in a timely manner.

[0040] Specifically, the flight performance process of all hydrogen energy unmanned aerial vehicles is divided into several monitoring periods, and the end collection time of each monitoring period is taken as the tuning time. At the tuning time, the flight data of the hydrogen energy unmanned aerial vehicle at each historical time within the preset historical period (the monitoring period) is analyzed, the yawing condition is evaluated, and it is determined whether to adjust the collection frequency. In the case of no yawing, the flight data is collected at a lower preset collection frequency to reduce the energy consumption of the unmanned aerial vehicle, and in the case of yawing, the collection frequency of the flight data is moderately increased to assist in timely discovery and intervention, thereby ensuring flight stability.

[0041] Among them, the flight data is collected at a preset collection frequency such as once per second in the first monitoring period, and the collection frequency of the flight data in the subsequent non-first monitoring period needs to be analyzed and adjusted. The collection frequencies in different non-first monitoring periods may not be consistent.

[0042] In one embodiment of the present application, the flight data of the hydrogen-powered UAV in each monitoring period is collected in real time by using various sensors or modules carried on the hydrogen-powered UAV to evaluate the yaw condition of the hydrogen-powered UAV, so as to prepare for subsequent evaluation and adjustment of the collection frequency of the flight data; the flight data at least includes flight position (referring to spatial coordinates, determined by a GPS module), flight speed (obtained by an inertial measurement unit), heading (obtained by the inertial measurement unit), flight environment information (collected by an anemometer), and obstacle detection information (obtained by a radar module); the collection process of the above data is a prior art and will not be described in detail.

[0043] In one preferred embodiment of the present application, the flight environment information at least includes wind speed and wind direction, which are used to evaluate the flight interference of wind on the hydrogen-powered UAV; the obstacle detection information at least includes the number of obstacles detected within a preset range centered on the hydrogen-powered UAV and the spatial detection distance between the hydrogen-powered UAV and each obstacle, which are used to evaluate whether there is a collision risk; the preset range can be set as the safe flight distance of the hydrogen-powered UAV (obtained from the UAV design specification, which will not be described in detail), and can also be customized by the implementer; the obstacles include other hydrogen-powered UAVs or aerial objects.

[0044] It should be noted that the analysis and adjustment method of the collection frequency of each hydrogen-powered UAV at each to-be-adjusted time is consistent, and only the adjustment of the collection frequency of any hydrogen-powered UAV at any to-be-adjusted time is analyzed and described.

[0045] Specifically, the flight data of the hydrogen-powered UAV at each historical time within a preset historical period such as the last 10 minutes at the to-be-adjusted time is obtained; the to-be-adjusted time and the preset historical period are a monitoring period, and the implementer can also adjust the preset historical period.

[0046] In step S2, for each hydrogen-powered UAV, in the preset historical period, the yaw distance of each historical time relative to the preset planning position is determined to determine all yaw periods, and the yaw risk coefficient of the hydrogen-powered UAV in each yaw period is obtained by combining the obstacle detection information and the flight speed of each historical time in each yaw period; in each historical time in each yaw period, the yaw attention coefficient of the hydrogen-powered UAV is obtained according to the complexity of the flight planning route of the hydrogen-powered UAV and the yaw risk coefficient in the yaw period, combined with the heading and flight environment information.

[0047] In the process of formation flight of the hydrogen energy unmanned aerial vehicle, the flight path, speed and attitude thereof need to be accurately consistent with the planned flight trajectory, so as to ensure flight safety and avoid deviation or collision. When the hydrogen energy unmanned aerial vehicle deviates, it needs to be closely monitored. Based on this, in the preset historical period, the deviation distance of the flight position of each historical moment relative to the preset planned position is determined to determine all deviation periods, so as to prepare for subsequent evaluation of deviation risk.

[0048] Preferably, in an embodiment of the present application, for each hydrogen energy unmanned aerial vehicle, the historical moment with a deviation distance greater than a preset distance threshold is regarded as a deviation moment, and the corresponding period of continuous deviation moments in the preset historical period is regarded as a deviation period.

[0049] The spatial distance (Euclidean distance) between the flight position of the hydrogen energy unmanned aerial vehicle at each historical moment and the preset planned position corresponding to the historical moment is regarded as the deviation distance, the preset distance threshold is set to 0.2 m, and the implementer can also adjust it. The preset planned position of each historical moment is preset during formation flight, and the determination process is not described again.

[0050] Modern hydrogen energy unmanned aerial vehicles are usually equipped with flight control systems. When the unmanned aerial vehicle deviates from the planned flight path, the control system will automatically start the attitude adjustment algorithm to correct the flight. If the deviation distance of the hydrogen energy unmanned aerial vehicle in each deviation period continuously increases, the possibility of an accident will be greater. When the hydrogen energy unmanned aerial vehicle performs actions such as turning or adjusting attitude, deviation may further cause it to collide with surrounding objects. If the flight speed of the hydrogen energy unmanned aerial vehicle is greater, the risk of collision accidents will be greater.

[0051] Based on this, after analyzing the deviation distance to determine the deviation period, the embodiment of the present application further combines the obstacle detection information and flight speed of each historical moment in each deviation period to obtain the deviation risk coefficient of the hydrogen energy unmanned aerial vehicle in each deviation period. By analyzing the flight space and the flight condition of the hydrogen energy unmanned aerial vehicle, the deviation risk coefficient reflects the possibility or risk degree of deviation accidents in the deviation period, and reflects the flight accuracy on the side, so as to prepare for subsequent evaluation of the deviation attention degree to adjust the collection frequency.

[0052] Preferably, in an embodiment of the present application, the method for obtaining the deviation risk coefficient comprises:

[0053] Please refer to Figure 2 which shows a method flowchart for obtaining a deviation risk coefficient provided by an embodiment of the present application, and specifically comprises:

[0054] Step S201, in each yaw period, a time sequence of the yaw distance is fitted, and all the monotonically increasing sub-sections in the time sequence are obtained, according to the length of each monotonically increasing sub-section and the variation range of the yaw distance in the monotonically increasing sub-section, the yaw parameter of the yaw period is obtained.

[0055] Considering that the yaw distance is continuously accumulated and increased, it is indicated that the hydrogen energy unmanned aerial vehicle is continuously yawing and the yaw degree is gradually increased, and considering that the time sequence of the yaw distance is fitted and the monotonically increasing sub-sections are determined, the continuous yawing of the hydrogen energy unmanned aerial vehicle can be evaluated, the greater the length of the monotonically increasing sub-sections, the greater the continuous yawing time, and the greater the variation range of the yaw distance in the monotonically increasing sub-sections, the greater the continuous yawing and the accumulated growth degree, and the greater the yaw parameter.

[0056] Based on this, in one preferred embodiment of the present application, the method for obtaining the yaw parameter comprises:

[0057] The length proportion of all the monotonically increasing sub-sections in the corresponding time sequence is taken as a first yaw factor, in each monotonically increasing sub-section, the length of the sub-section is taken as a yaw duration weight, and the range of the yaw distance in the sub-section is taken as a yaw range parameter, the yaw duration weight is weighted to the yaw range parameter, and the weighted sum result of all the monotonically increasing sub-sections is taken as a second yaw factor, and the first yaw factor and the second yaw factor are fused to obtain the yaw parameter.

[0058] It should be noted that fitting the time sequence and obtaining all the monotonically increasing sub-sections in the time sequence, and obtaining the range have been prior art, and will not be repeated here, the total length of all the monotonically increasing sub-sections is taken as a numerator, and the length of the time sequence is taken as a denominator to obtain the length proportion, that is, the first yaw factor, the length of each monotonically increasing sub-section is multiplied by the range of the yaw distance in the sub-section, the sum of the products corresponding to all the monotonically increasing sub-sections is obtained, and the weighted sum result is obtained, that is, the second yaw factor, and finally the first yaw factor and the second yaw factor are multiplied and fused to obtain the yaw parameter.

[0059] Step S202, at each historical time in each yaw period, according to the spatial detection distance between the hydrogen energy unmanned aerial vehicle and each obstacle and the total number of obstacles, a collision parameter is obtained, the flight speed of the hydrogen energy unmanned aerial vehicle is weighted to the collision parameter to obtain a risk sub-parameter, and according to the centralized characteristics of the risk sub-parameters in the yaw period, a risk parameter of the yaw period is obtained.

[0060] The more the total number of obstacles around the hydrogen energy unmanned aerial vehicle at each historical moment in the yaw period, and the closer the distance between the hydrogen energy unmanned aerial vehicle and each obstacle, the greater the risk of collision with the obstacle in the subsequent flight pose adjustment process. Meanwhile, the greater the flight speed of the hydrogen energy unmanned aerial vehicle, the more any slight yaw change will seriously affect the flight trajectory, thereby increasing the risk of yaw collision. Based on this, the yaw collision risk at all historical moments can be analyzed to determine the risk parameter of the hydrogen energy unmanned aerial vehicle in the yaw period.

[0061] As an example, at each historical moment in each yaw period, a minimum spatial detection distance is selected from the spatial detection distance between the hydrogen energy unmanned aerial vehicle and each obstacle, the reciprocal of the minimum spatial detection distance is adjusted to a negative correlation mapping logic, and then the reciprocal is multiplied and combined with the total number of obstacles to obtain a collision parameter, so that the more the number of obstacles and the closer the distance, the greater the collision parameter. Then, the flight speed of the hydrogen energy unmanned aerial vehicle is multiplied by the collision parameter to obtain a risk sub-parameter corresponding to the historical moment. Finally, the mean value of the risk sub-parameters of all historical moments is taken as the risk parameter of the yaw period.

[0062] Step S203, fusion of yaw parameter and risk parameter, obtaining the yaw risk coefficient of the hydrogen energy unmanned aerial vehicle in the corresponding yaw period.

[0063] It is considered that the yaw parameter reflects the degree of continuous yaw of the hydrogen energy unmanned aerial vehicle, and the risk parameter reflects the possibility of yaw collision, both of which reflect the yaw risk. Based on this, the yaw parameter and the risk parameter are multiplied and fused, and the product is taken as the yaw risk coefficient of the hydrogen energy unmanned aerial vehicle in the corresponding yaw period.

[0064] At this point, the yaw risk coefficient of the hydrogen energy unmanned aerial vehicle in each yaw period is obtained.

[0065] It is considered that at each historical moment in the yaw period, the more complex the future planned flight path of the hydrogen energy unmanned aerial vehicle, the more unstable the current flight state, and the greater the flight uncertainty and risk. At this time, if the hydrogen energy unmanned aerial vehicle is affected by the flight environment such as wind, the flight uncertainty and risk will further increase. It is also considered that the wind direction may cause the hydrogen energy unmanned aerial vehicle to deviate from the planned heading, thereby increasing the flight risk. The greater the flight risk, the more attention should be paid to the flight direction of the hydrogen energy unmanned aerial vehicle in order to timely adjust the unmanned aerial vehicle and reduce the risk of flight accidents.

[0066] Based on this, the embodiment of the present application obtains the deviation attention coefficient of the hydrogen energy unmanned aerial vehicle at each historical moment in each deviation period according to the complex situation of the flight planning route of the hydrogen energy unmanned aerial vehicle and the deviation risk coefficient of the hydrogen energy unmanned aerial vehicle in the deviation period, and combines the heading and flight environment information of the hydrogen energy unmanned aerial vehicle. The deviation attention coefficient not only reflects the accident risk degree of the hydrogen energy unmanned aerial vehicle, but also reflects the necessity of focusing on it, and prepares for subsequent adjustment of the collection frequency.

[0067] Preferably, in an embodiment of the present application, it is considered that the greater the deviation risk coefficient of the deviation period is, the greater the attention degree should be. Meanwhile, at each historical moment in the deviation period, the greater the possibility that the unmanned aerial vehicle deviates from the preset heading due to wind interference is when the wind direction is more perpendicular to the heading of the hydrogen energy unmanned aerial vehicle, and the greater the wind speed is, the greater the influence it receives is, and at this time, it should be paid more attention to. It is also considered that the more complex the flight route planned by the hydrogen energy unmanned aerial vehicle in the future period is, and the more frequent the heading conversion is, the more unstable the flight state is, and it should also be paid more attention to.

[0068] Based on this, the method for obtaining the deviation attention coefficient comprises:

[0069] At each historical moment in each deviation period, a resistance factor is obtained according to the minimum included angle between the heading of the hydrogen energy unmanned aerial vehicle and the wind direction, and a flight resistance coefficient is obtained by weighting the resistance factor with the wind speed. A future flight period of a preset length is determined along the time sequence direction from the historical moment as a starting point, a flight complex coefficient of the hydrogen energy unmanned aerial vehicle is obtained according to the angle change between the planned headings of adjacent moments in the flight planning route of the hydrogen energy unmanned aerial vehicle in the future flight period.

[0070] The flight resistance coefficient, the flight complex coefficient and the deviation risk coefficient of the deviation period to which the historical moment belongs of the hydrogen energy unmanned aerial vehicle at each historical moment are fused, and the fusion result is taken as the deviation attention coefficient of the hydrogen energy unmanned aerial vehicle.

[0071] As an example, first, at each historical moment in each deviation period, the absolute value of the difference between the minimum included angle between the heading and the wind direction and the preset angle 90° (which needs to be normalized to remove the dimension, that is, only the angle value is concerned) is mapped into the exponential function exp(-x) adjustment logic with the natural constant e as the base number. The smaller the angle difference absolute value is, the greater the possibility that the heading is perpendicular to the wind direction is, and the greater the exponential value is. The exponential value is taken as the resistance factor and multiplied by the wind speed to obtain the flight resistance coefficient.

[0072] Then, the preset length is set to 10, and the implementer can also customize it. In the deviation period, 10 collection moments are determined along the time sequence direction from each historical moment as a starting point to obtain the future flight period of the historical moment, and the flight planning route of the hydrogen energy unmanned aerial vehicle in the future flight period is obtained. The flight planning route is preset when the hydrogen energy unmanned aerial vehicle is in formation flight, and the determination process is not described again.

[0073] In a preferred embodiment of the present application, considering that the greater the planned heading change between adjacent time points in the flight planning route within the future flight period, the greater the turning range, the greater the flight difficulty and uncertainty; therefore, the minimum angle between the planned heading of each time point in the flight planning route of the hydrogen energy unmanned aerial vehicle within the future flight period and the planned heading of the adjacent next time point is taken as the turning range parameter of each time point; in the future flight period, the range of the turning range parameter is taken as the turning difficulty factor, and the greater the range, the greater the turning angle; the negative correlation mapping result of the time interval between the corresponding time points is taken as the reciprocal, and the reciprocal is taken as the difficulty weight, and the smaller the time interval, the faster the turning speed of the unmanned aerial vehicle, and the greater the turning difficulty; the turning difficulty factor is weighted by the difficulty weight, and the weighted result is taken as the flight complexity coefficient;

[0074] It should be noted that the minimum angle and the range are obtained by the prior art, and will not be described again; the minimum angle needs to be standardized for operation; the implementer can also use other negative correlation mapping means;

[0075] Finally, at each historical time point, the flight resistance coefficient of the hydrogen energy unmanned aerial vehicle at each historical time point, the flight complexity coefficient and the yaw risk coefficient of the yaw period to which the historical time point belongs are multiplied and fused, and the product is taken as the yaw attention coefficient.

[0076] Step S3, at the to-be-adjusted time point, for each hydrogen energy unmanned aerial vehicle, the future yaw distance of the hydrogen energy unmanned aerial vehicle in the preset future period is predicted according to the yaw attention coefficient of the hydrogen energy unmanned aerial vehicle at each historical time point and the time interval between the historical time point and the to-be-adjusted time point, and the flight data of the hydrogen energy unmanned aerial vehicle in the preset future period is collected based on the future yaw distance.

[0077] It is considered that by predicting the yaw of the hydrogen energy unmanned aerial vehicle in the future period at the to-be-adjusted time point, the necessity of subsequent close attention to the hydrogen energy unmanned aerial vehicle can be evaluated; it is also considered that in the preset historical period at the to-be-adjusted time point, the closer the historical time point to the to-be-adjusted time point and the greater the yaw attention coefficient of the historical time point, the greater the reference or influence of the historical time point on the future yaw analysis of the to-be-adjusted time point, and the greater the attention weight of the historical time point in the prediction;

[0078] Based on this, in the embodiment of the present application, the future yaw distance of the hydrogen energy unmanned aerial vehicle in the preset future period is predicted according to the yaw attention coefficient of the hydrogen energy unmanned aerial vehicle at each historical time point and the time interval between the historical time point and the to-be-adjusted time point at the to-be-adjusted time point, so as to prepare for subsequent determination of the flight data collection frequency of the hydrogen energy unmanned aerial vehicle in the preset future period based on the future yaw distance.

[0079] Preferably, in one embodiment of the present application, considering that the weighted ARIMA algorithm is a time series prediction algorithm, and the weight can adjust the attention degree of each data in the time series, so that the data with higher attention degree has greater impact on the prediction result, thereby improving the prediction effect; and considering that the closer to the to-be-adjusted time and the greater the yaw attention coefficient of the historical time in the yaw period, the greater the reference value provided by the historical time, based on which the attention weight can be determined; and then based on the weighted ARIMA algorithm, the time series prediction is performed to obtain the yaw distance in the preset future period of the to-be-adjusted time;

[0080] Based on this, the method for obtaining the future yaw distance comprises:

[0081] In the preset historical period, for each hydrogen energy unmanned aerial vehicle, the negative correlation mapping result of the time interval between each historical time in the yaw period and the to-be-adjusted time is taken as the reference weight of the corresponding historical time, the corresponding yaw attention coefficient is weighted by using the reference weight, and the attention weight of the corresponding historical time is obtained; wherein the attention weight of the historical time in the non-yaw period is a preset value;

[0082] Based on the weighted ARIMA algorithm and the attention weight, the yaw distance of the hydrogen energy unmanned aerial vehicle in the preset historical period is time series predicted, and the mean value of the yaw distance in the preset future period obtained by the prediction is taken as the future yaw distance.

[0083] As an example, for each historical time in the yaw period, the time interval between the historical time and the to-be-adjusted time is mapped into an exponential function exp(-x) with the natural constant e as the base number, so that the smaller the time interval, the greater the negative correlation mapping value, and the greater the reference weight; and then the reference weight is multiplied by the corresponding yaw attention coefficient to obtain the attention weight;

[0084] The historical time in the non-yaw period also participates in the time series prediction, so the attention weight of the historical time in the non-yaw period is set to a preset value, specifically, the preset value is set to the minimum value of the attention weights of all historical times in the yaw period in the preset historical period, or to any positive number smaller than the minimum value, so that the algorithm pays more attention to the historical times in the yaw period;

[0085] Before prediction, the yaw distance of the hydrogen energy unmanned aerial vehicle at each historical time in the preset historical period of the to-be-adjusted time needs to be obtained, and the yaw distances are sorted according to the collection time sequence to construct a historical time series, and the obtaining process is not repeated here;

[0086] The attention weight of each historical moment in a preset historical period and the historical time sequence are input into the pre-trained weighted ARIMA model for time sequence prediction, and the model automatically outputs the prediction result of the yaw distance of the to-be-adjusted moment in a preset future period. The mean of the yaw distance in the preset future period obtained by prediction is taken as the future yaw distance; wherein the length of the preset future period is consistent with the length of the preset historical period, and the preset future period is also the next adjacent monitoring period of the to-be-adjusted moment.

[0087] It should be noted that the training and application of the weighted ARIMA model are existing technologies and will not be described in detail.

[0088] After obtaining the future yaw distance, the collection frequency of flight data can be further determined and analyzed to assist timely intervention in the flight state of the hydrogen energy unmanned aerial vehicle.

[0089] Preferably, in an embodiment of the present application, considering that the larger the future yaw distance is, the more necessary it is to pay close attention to it, when the future yaw distance is greater than a preset distance threshold, the future yaw distance is mapped into [0, 1], and a constant 1 is added to the mapped value to obtain the collection frequency by weighting the preset collection frequency.

[0090] Wherein, the preset distance threshold is 0.2m, and the implementer can also adjust it; then the future yaw distance is mapped into [0, 1] through the hyperbolic tangent function, when the future yaw distance is larger, the mapped value will be larger, and the adjustment weight will also be larger, so as to increase the preset collection frequency, thereby paying close attention to the flight dynamics of the hydrogen energy unmanned aerial vehicle.

[0091] It should be noted that the collection frequency at the to-be-adjusted moment will be applied to the next monitoring period, and a new round of yaw analysis will be performed in the next monitoring period, and the collection frequency will be re-determined at a new to-be-adjusted moment, which is iterated until the flight performance ends.

[0092] Based on the same inventive concept, the present application also proposes a flight data collection system for a hydrogen energy unmanned aerial vehicle, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the flight data collection method for the hydrogen energy unmanned aerial vehicle described in steps S1-S3 is realized.

[0093] To sum up, the application firstly collects flight data of each hydrogen energy unmanned aerial vehicle at each historical moment in a preset historical period at a to-be-adjusted moment; then determines all yaw periods of each hydrogen energy unmanned aerial vehicle in the preset historical period, and obtains a yaw risk coefficient of the hydrogen energy unmanned aerial vehicle in each yaw period; further obtains a yaw attention coefficient of each hydrogen energy unmanned aerial vehicle at each historical moment in each yaw period, and then, at the to-be-adjusted moment, in combination with a time interval between each historical moment and the to-be-adjusted moment, predicts a future yaw distance of each hydrogen energy unmanned aerial vehicle, and finally determines an acquisition frequency of flight data of the hydrogen energy unmanned aerial vehicle in a preset future period based on the future yaw distance and acquires the flight data. The application comprehensively analyzes flight dynamics, flight space and environment of the hydrogen energy unmanned aerial vehicle in a flight process, evaluates flight risk thereof, predicts future yaw dynamics thereof to adjust the acquisition frequency, adjusts attention to the flight dynamics, reduces the inability to discover yaw risk of the unmanned aerial vehicle in time, and improves flight data acquisition effect of the hydrogen energy unmanned aerial vehicle.

[0094] It should be noted that the above-mentioned sequence of the embodiments of the application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0095] Each embodiment in the specification is described in a progressive manner, and the same and similar parts between each embodiment can be referred to each other. Each embodiment mainly describes the difference from other embodiments.

Claims

1. A flight data acquisition method for a hydrogen energy unmanned aerial vehicle, characterized in that, The method comprises: Collecting flight data of each hydrogen energy unmanned aerial vehicle at each historical moment within a preset historical period at a to-be-adjusted moment, the to-be-adjusted moment being the end moment of the preset historical period, the flight data at least comprising flight position, flight speed, heading, flight environment information and obstacle detection information; For each hydrogen energy unmanned aerial vehicle, within the preset historical period, the yaw distance of each historical moment relative to a preset planning position is used to determine all yaw periods, and the yaw risk coefficient of the hydrogen energy unmanned aerial vehicle in each yaw period is obtained by combining the obstacle detection information and the flight speed of each historical moment within each yaw period; at each historical moment within each yaw period, the yaw attention coefficient of the hydrogen energy unmanned aerial vehicle is obtained according to the complexity of the flight planning route of the hydrogen energy unmanned aerial vehicle and the yaw risk coefficient in the yaw period, in combination with the heading and flight environment information. At the to-be-adjusted moment, for each hydrogen energy unmanned aerial vehicle, the future yaw distance of the hydrogen energy unmanned aerial vehicle in a preset future period is predicted according to the yaw attention coefficient of the hydrogen energy unmanned aerial vehicle at each historical moment and the time interval between the historical moment and the to-be-adjusted moment, and the flight data of the hydrogen energy unmanned aerial vehicle in the preset future period is collected based on the future yaw distance.

2. The flight data collection method for hydrogen energy unmanned aerial vehicles according to claim 1, characterized in that, The flight environment information at least comprises wind direction and wind speed; the obstacle detection information at least comprises the number of obstacles detected within a preset range centered on the hydrogen energy unmanned aerial vehicle and the spatial detection distance between the hydrogen energy unmanned aerial vehicle and each obstacle.

3. The flight data collection method for hydrogen energy unmanned aerial vehicles according to claim 1, characterized in that, The method for obtaining the yaw period comprises: For each hydrogen energy unmanned aerial vehicle, historical moments with a yaw distance greater than a preset distance threshold are taken as yaw moments, and a period corresponding to the continuous yaw moments within the preset historical period is taken as a yaw period.

4. The flight data collection method for hydrogen energy unmanned aerial vehicles according to claim 2, characterized in that, The method for obtaining the yaw risk coefficient comprises: Within each yaw period, a time sequence of the yaw distance is fitted and all monotonically increasing subsegments are obtained, a yaw parameter of the yaw period is obtained according to the length of each monotonically increasing subsegment and the variation amplitude of the yaw distance in the monotonically increasing subsegment; At each historical moment of each yaw period, a collision parameter is obtained according to the spatial detection distance between the hydrogen energy unmanned aerial vehicle and each obstacle and the total number of obstacles, the collision parameter is weighted by the flight speed of the hydrogen energy unmanned aerial vehicle to obtain a risk subparameter, and a risk parameter of the yaw period is obtained according to the concentration characteristics of the risk subparameters in the yaw period; The yaw risk coefficient of the hydrogen energy unmanned aerial vehicle in the corresponding yaw period is obtained by fusing the yaw parameter and the risk parameter.

5. The flight data collection method for hydrogen energy unmanned aerial vehicle according to claim 4, characterized in that, The method for obtaining the yaw parameter comprises: The length proportion of all monotonically increasing subsegments in the corresponding time sequence is taken as a first yaw factor, the length of each monotonically increasing subsegment is taken as a yaw duration weight, and the range of the yaw distance in each monotonically increasing subsegment is taken as a yaw amplitude parameter, the yaw amplitude parameter is weighted by the yaw duration weight, the weighted sum result of all monotonically increasing subsegments is taken as a second yaw factor, and the first yaw factor and the second yaw factor are fused to obtain the yaw parameter.

6. The flight data collection method for hydrogen energy unmanned aerial vehicles according to claim 2, characterized in that, The method for obtaining the yaw attention coefficient comprises: At each historical moment in each drift period, a resistance factor is obtained according to the minimum angle between the heading direction of the hydrogen energy unmanned aerial vehicle and the wind direction, the flight resistance coefficient is obtained by weighting the wind speed on the resistance factor, a future flight period of a preset length is determined along the time sequence direction from the historical moment, and a flight complexity coefficient of the hydrogen energy unmanned aerial vehicle is obtained according to the angle change between the adjacent moments in the flight planning route of the hydrogen energy unmanned aerial vehicle in the future flight period. The flight resistance coefficient, the flight complexity coefficient and the drift risk coefficient of the drift period to which the historical moment belongs are fused, and the fusion result is taken as the drift attention coefficient of the hydrogen energy unmanned aerial vehicle.

7. The flight data collection method for hydrogen energy unmanned aerial vehicles according to claim 6, characterized in that, The flight complexity coefficient is obtained by: The minimum angle between the planning heading direction of the hydrogen energy unmanned aerial vehicle at each moment in the flight planning route in the future flight period and the planning heading direction of the adjacent next moment is taken as the steering amplitude parameter at each moment, the range of the steering amplitude parameter in the future flight period is taken as the steering difficulty factor, the negative correlation mapping result of the time interval between the moments corresponding to the range is taken as the difficulty weight, the steering difficulty factor is weighted by using the difficulty weight, and the weighted result is taken as the flight complexity coefficient.

8. The flight data collection method for hydrogen energy unmanned aerial vehicles according to claim 6, characterized in that, The future drift distance is obtained by: In a preset historical period, for each hydrogen energy unmanned aerial vehicle, the negative correlation mapping result of the time interval between each historical moment and the to-be-adjusted moment in the drift period is taken as the reference weight of the corresponding historical moment, the corresponding drift attention coefficient is weighted by using the reference weight, and the attention weight of the corresponding historical moment is obtained; wherein the attention weight of the historical moment in the non-drift period is a preset value; Based on the weighted ARIMA algorithm and the attention weight, the drift distance of the hydrogen energy unmanned aerial vehicle in the preset historical period is sequentially predicted, and the mean value of the predicted drift distance in the preset future period is taken as the future drift distance.

9. The flight data collection method for hydrogen energy unmanned aerial vehicle according to claim 1, characterized in that, The acquisition frequency is obtained by: When the future drift distance is greater than a preset distance threshold, the future drift distance is mapped into [0, 1], and a constant 1 is added to the mapped value to obtain an adjustment weight for weighting the preset acquisition frequency, thereby obtaining the acquisition frequency.

10. A flight data acquisition system for a hydrogen-powered drone, characterized by, The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the flight data acquisition method for the hydrogen energy unmanned aerial vehicle according to any one of claims 1-9.

Images