CN116451126A

CN116451126A - Method and device for determining helicopter operation track and pesticide application amount

Info

Publication number: CN116451126A
Application number: CN202310695502.XA
Authority: CN
Inventors: 赵春江; 徐刚; 陈立平; 张瑞瑞; 王维佳; 伊铜川; 孔祥宁; 杨恒毅
Original assignee: Intelligent Equipment Technology Research Center of Beijing Academy of Agricultural and Forestry Sciences
Current assignee: Intelligent Equipment Technology Research Center of Beijing Academy of Agricultural and Forestry Sciences
Priority date: 2023-06-13
Filing date: 2023-06-13
Publication date: 2023-07-18
Anticipated expiration: 2043-06-13
Also published as: CN116451126B

Abstract

The invention provides a method and a device for determining the operation track and the pesticide application amount of a helicopter, which belong to the technical field of agriculture and forestry information and comprise the following steps: acquiring operation data of a helicopter; determining a job status based on the job data to determine a job trajectory based on the job status; and determining the pesticide application frame times, the spraying flow and the spraying time based on the operation track so as to obtain the pesticide application amount. According to the invention, the flight state and the spraying state of the helicopter are determined by acquiring the flight state data and the spraying state data of the helicopter, and the operation track is determined according to the flight state and the spraying state of the helicopter, so that all the pesticide application frames of the helicopter in the operation period and the spraying flow and the spraying time of each pesticide application frame are accurately determined, and finally the pesticide application amount of each pesticide application frame is determined, thereby accurately identifying the operation state and the track of the helicopter, conveniently and accurately distinguishing the pesticide application frames, improving the pesticide application amount measurement precision, and realizing automatic pesticide application amount measurement without on-site monitoring and recording of staff.

Description

Method and device for determining helicopter operation track and pesticide application amount

Technical Field

The invention relates to the technical field of agriculture and forestry information, in particular to a method and a device for determining the operation track and the pesticide application amount of a helicopter.

Background

In the case of large-area aerial application in mountainous areas, it is common practice in the industry to apply the medicaments aeronautically by means of helicopters.

At present, in the traditional helicopter pesticide spraying operation process, monitoring equipment is arranged on the helicopter, and the spraying state of the helicopter is judged by monitoring a flow threshold value or a pressure threshold value, so that the pesticide spraying amount of the helicopter is determined.

However, in the traditional method for measuring the application amount, because the mountain terrain environment is complex, the altitude drop is large, the condition of flow and pressure fluctuation easily occurs in the operation process, so that the spraying state is difficult to accurately identify, and the working personnel are required to monitor and record the application place of the helicopter on site, so that the defects of low application amount measurement precision and incapability of full-course automatic monitoring exist.

Accordingly, there is a need in the art for a method of determining the trajectory and dosage of a helicopter that overcomes the above-described deficiencies in the prior art.

Disclosure of Invention

The invention provides a method and a device for determining the operation track and the pesticide application amount of a helicopter, which are used for solving the defects that the pesticide application amount measurement precision is low and the whole-course automatic monitoring cannot be realized in the prior art.

In a first aspect, the invention provides a method of determining a helicopter operational trajectory and a dosage comprising: acquiring operation data of the helicopter in one operation period, wherein the operation data comprise flight state data and spraying state data; determining an operation state of the helicopter at each sampling time in the operation period based on the operation data, so as to determine an operation track of the helicopter in the operation period based on the operation state; the working state comprises a flight state and a spraying state; determining all pesticide application frames of the helicopter in the working period, and the spraying flow and the spraying duration in each pesticide application frame based on the working track of the helicopter; and acquiring the application amount of the helicopter in each application rack according to the spraying flow and the spraying time in each application rack.

According to the method for determining the operation track and the application amount of the helicopter provided by the invention, the operation state of the helicopter at each sampling moment in the operation period is determined based on the operation data, and the method specifically comprises the following steps: inputting the flight state data of each sampling time in the working period to a flight state identification model group so as to determine the flight state of the helicopter at each sampling time in the working period; inputting the spraying state data of each sampling time in the working period to a spraying state identification model group so as to determine the spraying state of the helicopter at each sampling time in the working period; the flight state recognition model set is obtained by training based on a first training sample set, wherein the first training sample set comprises a plurality of flight state data samples and flight state labels corresponding to each flight state data sample; the spray state recognition model group is obtained by training based on a second training sample set, and the second training sample set comprises a plurality of spray state data samples and spray state labels corresponding to each spray state data sample.

According to the method for determining the helicopter operation track and the application amount provided by the invention, the flight state identification model group comprises at least one of the following plurality of flight state identification models: a shutdown state recognition model for discriminating whether the helicopter is in a shutdown state, a cruise state recognition model for discriminating whether the helicopter is in a cruise state, a hover state recognition model for discriminating whether the helicopter is in a hover state, a climb state recognition model for discriminating whether the helicopter is in a climb state, a descent state recognition model for discriminating whether the helicopter is in a descent state, a vertical ascent state recognition model for discriminating whether the helicopter is in a vertical ascent state, and a vertical descent state recognition model for discriminating whether the helicopter is in a vertical descent state; the step of inputting the flight state data of each sampling time in the working period into a flight state identification model group to determine the flight state of the helicopter at each sampling time in the working period specifically comprises the following steps: respectively inputting the flight state data of each sampling moment into each flight state identification model, and acquiring the confidence coefficient of the flight state identification result output by each flight state identification model; determining the maximum value of all the confidence coefficients as a first characteristic value; if the first characteristic value is larger than a first preset threshold value, determining a flight state identification result corresponding to the first characteristic value as a flight state of the helicopter at the sampling moment; otherwise, determining the flight state of the helicopter at the sampling moment as other flight states.

According to the method for determining the helicopter operation track and the pesticide application amount provided by the invention, the flight state data comprise the flying ground speed, the altitude from the ground, the altitude increment from the ground and the triaxial acceleration.

According to the method for determining the helicopter operation track and the application amount provided by the invention, the spraying state identification model group comprises at least one of the following multiple spraying state identification models: a stopped state recognition model for judging whether the helicopter is in a spraying stopped state, a spraying state recognition model for judging whether the helicopter is in a spraying state, a starting state recognition model for judging whether the helicopter is in a spraying starting state, a closed state recognition model for judging whether the helicopter is in a spraying closed state, a blocking state recognition model for judging whether the helicopter is in a spraying blocking state, a leaking state recognition model for judging whether the helicopter is in a spraying leaking state and a testing state recognition model for judging whether the helicopter is in a spraying testing state; the step of inputting the spraying state data of each sampling time in the working period into a spraying state identification model group so as to determine the spraying state of the helicopter at each sampling time in the working period, specifically comprising the following steps: the spraying state data at each sampling moment are respectively input into each spraying state identification model, and the confidence coefficient of the spraying state identification result output by each spraying state identification model is obtained; determining the maximum value of all the confidence coefficients as a second characteristic value; if the second characteristic value is larger than a second preset threshold value, determining a spraying state identification result corresponding to the second characteristic value as a spraying state of the helicopter at the sampling moment; otherwise, determining that the spraying state of the helicopter at the sampling moment is other spraying states.

According to the method for determining the helicopter operation track and the application rate, the spraying state data comprise the flying ground speed, the altitude, the spraying flow, the flow trend, the spraying pressure and the pressure trend and the geometric average value of the triaxial acceleration.

According to the method for determining the operation track and the pesticide application amount of the helicopter, after the spraying flow and the spraying time length in each pesticide application frame are obtained so as to obtain the pesticide application amount of the helicopter in each pesticide application frame, the method further comprises the following steps: if the absolute value of the deviation between the drug application amount of the helicopter in any drug application rack and the capacity of a drug box of the helicopter is determined, and the absolute value of the deviation is larger than a third preset threshold value, marking any drug application rack as a rack to be calibrated; taking the flight state data of the rack to be calibrated as a training sample, acquiring a real flight state corresponding to the flight state data of the rack to be calibrated as a tag, and training the flight state identification model group again; taking the spraying state data of the rack to be calibrated as a training sample, acquiring a real spraying state corresponding to the spraying state data of the rack to be calibrated as a label, and training the spraying state identification model group again.

The method for determining the helicopter operation track and the pesticide application amount provided by the invention further comprises the following steps: acquiring triaxial acceleration of the helicopter at any sampling moment in the working period; based on the triaxial acceleration, acquiring the flying ground speed and the flying direction of the helicopter at any sampling moment; and determining the longitude and latitude information of the helicopter at any sampling moment according to the flying ground speed and the flying direction of the helicopter at any sampling moment and the longitude and latitude information of the helicopter at the last sampling moment.

In a second aspect, the present invention also provides a device for determining the working track and the dosage of a helicopter, comprising: the data acquisition unit is used for acquiring operation data of the helicopter in one operation period, wherein the operation data comprise flight state data and spraying state data; a track determining unit, configured to determine, based on the operation data, an operation state of the helicopter at each sampling time in the operation period, so as to determine, based on the operation state, an operation track of the helicopter in the operation period; the working state comprises a flight state and a spraying state; the system comprises an overhead determining unit, a control unit and a control unit, wherein the overhead determining unit is used for determining all pesticide application overhead of the helicopter in the working period, and the spraying flow and the spraying duration in each pesticide application overhead based on the working track of the helicopter; the medicine amount acquisition unit is used for acquiring the medicine amount of the helicopter in each medicine application frame according to the spraying flow and the spraying time in each medicine application frame.

In a third aspect, the invention provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method of determining a helicopter operational trajectory and dosage as described in any of the foregoing when the program is executed.

In a fourth aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of a method of determining a helicopter work track and a dosage as described in any of the above.

In a fifth aspect, the invention also provides a computer program product comprising a computer program which, when executed by a processor, carries out the steps of a method of determining a helicopter working track and a dosage as described in any of the above.

According to the method and the device for determining the operation track and the pesticide application amount of the helicopter, the flight state and the spraying state of the helicopter are determined by acquiring the flight state data and the spraying state data of the helicopter, the operation track is determined according to the flight state and the spraying state of the helicopter, and then all pesticide application frames of the helicopter in an operation period and the spraying flow and the spraying time of each pesticide application frame are accurately determined, and finally the pesticide application amount of each pesticide application frame is determined, so that the operation state and the track of the helicopter are accurately identified, the pesticide application frame is accurately distinguished, the pesticide application amount measurement precision is improved, and the on-site monitoring record of workers is not needed, so that the automatic pesticide application amount measurement is realized.

Drawings

In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a method for determining helicopter operational trajectories and drug delivery rates provided by the present invention;

FIG. 2 is a second flow chart of the method of determining the trajectory and dosage of a helicopter provided by the present invention;

FIG. 3 is a schematic diagram of the apparatus for determining the trajectory and dosage of a helicopter provided by the present invention;

fig. 4 is a schematic structural diagram of an electronic device provided by the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

It should be noted that in the description of embodiments of the present invention, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. The orientation or positional relationship indicated by the terms "upper", "lower", etc. are based on the orientation or positional relationship shown in the drawings, are merely for convenience of description and to simplify the description, and are not indicative or implying that the apparatus or elements in question must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as limiting the present invention. Unless specifically stated or limited otherwise, the terms "mounted," "connected," and "coupled" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.

The terms "first," "second," and the like in this application are used for distinguishing between similar objects and not for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged, as appropriate, such that embodiments of the present application may be implemented in sequences other than those illustrated or described herein, and that the objects identified by "first," "second," etc. are generally of a type and not limited to the number of objects, e.g., the first object may be one or more.

The method and apparatus for determining the track and dosage of a helicopter according to the present invention are described below with reference to fig. 1-4.

FIG. 1 is a schematic flow chart of a method for determining helicopter operational trajectories and drug delivery rates according to the present invention, as shown in FIG. 1, including, but not limited to, the steps of:

step 101: and acquiring operation data of the helicopter in one operation period, wherein the operation data comprise flight state data and spraying state data.

Specifically, the on-board terminal is arranged on the helicopter, so that the on-board terminal can acquire the operation data of the helicopter according to a data acquisition frequency in the process of the pesticide application operation of the helicopter. The data acquisition frequency can be preset according to the use requirement of a specific scene, for example, preset to be 1Hz or 10Hz.

The operation data may include flight status data and spraying status data of the helicopter, for example, when the operation data of the helicopter in one operation period is obtained through an onboard terminal, various operation data such as a positioning status of a global navigation satellite system (Global Navigation Satellite System, GNSS), universal time (Universal Time Coordinated, UTC), longitude information, latitude information, ground speed, flight direction, altitude, ground altitude, spraying flow, spraying pressure, and triaxial acceleration may be collected.

Optionally, after the on-board terminal on the helicopter acquires the operation data, all the operation data can be packaged and sent to the remote server side, so that the server can uniformly process the operation data of the helicopter in one operation period.

Step 102: determining an operation state of the helicopter at each sampling moment in the operation period based on the operation data, so as to determine an operation track of the helicopter in the operation period based on the operation state; the operational state includes a flight state and a spraying state.

Specifically, after the server receives the operation data of the helicopter, the operation state of the helicopter at each sampling time in the operation period can be determined based on the operation data. Each sampling time corresponds to the data acquisition frequency of the operation data acquired by the airborne terminal.

Further, the server can determine the working state of the helicopter by matching the working data with a state parameter table which is set through experiments in advance; the working state of the helicopter can also be identified by inputting working data into a pre-trained state identification model.

The flight state of the helicopter at each sampling moment can be determined based on the flight state data of the helicopter at each sampling moment in one working period; the spraying state of the helicopter at each sampling moment can also be determined based on the spraying state data of the helicopter at each sampling moment in one working period.

Further, after the working state of each sampling time of the helicopter is determined, the working track of the helicopter in the working period can be determined according to time information, positioning information (such as longitude information and latitude information) of the helicopter and altitude information.

Step 103: based on the working track of the helicopter, determining all the pesticide application frames of the helicopter in the working period, and the spraying flow and the spraying time length in each pesticide application frame.

Specifically, after the operation track of the helicopter in the operation period is determined, all the pesticide application frames of the helicopter in the operation period, and the spraying flow and the spraying duration in each pesticide application frame are determined based on the operation track of the helicopter.

As an alternative embodiment, all working data between two adjacent shutdown states of the helicopter may be identified as one cycle based on the sampling times at which all flight states are determined as shutdown states, and further, all pesticide application cycles of the helicopter within a working cycle are determined based on the spraying state of the helicopter at each sampling time under each cycle.

For example, based on the operation data of the helicopter in one operation period, determining that the total five flight states of the helicopter are determined to be in a stop state, further determining that the helicopter has four frames in the operation period, further acquiring a spraying state mark of the helicopter at each sampling moment under each frame, and if the frames do not comprise the state marks of spraying states such as spraying start, spraying closing and the like, determining that the frames are transition or test frames of the helicopter and not applying the pesticide; if the rack includes a status flag of the spraying status such as spraying start, spraying off, etc., the rack can be determined to be the medicine applying rack.

Further, the spraying flow rate and the spraying time length of each application frame can be determined according to the data such as the spraying flow rate and the spraying pressure at each sampling time in each application frame.

Optionally, after determining that all the pesticide application frames of the helicopter are in the working period, the method can automatically identify the stopping and pesticide adding place of the helicopter further based on positioning information, longitude information and latitude information of the helicopter in a stopping state before and after the pesticide application frames, so that the aviation pesticide application working process of the helicopter is automatically monitored.

Step 104: and acquiring the pesticide application amount of the helicopter in each pesticide application rack according to the spraying flow and the spraying time in each pesticide application rack.

Specifically, after determining all the pesticide application frames of the helicopter in one working period and the spraying flow and the spraying time length of each pesticide application frame, the pesticide application amount of the helicopter in each pesticide application frame can be rapidly measured based on the spraying flow and the spraying time length of each pesticide application frame, so that the pesticide application amount is automatically measured.

The existing helicopter pesticide application amount measuring method is characterized in that monitoring equipment is installed on a helicopter, and a spraying state is judged by simply using a flow threshold value or a pressure threshold value, but often the working state of the helicopter cannot be accurately identified and distinguished due to the complex working environment of a mountain area, and the spraying state of the helicopter is difficult to accurately identify when the flow and the pressure fluctuate due to the fact that the types of different helicopters and the spraying equipment are large in difference.

In addition, because the helicopter can vertically take off and land, the airport and the dosing point are more flexibly arranged, and when the dosing place is identified through simple positions and speeds in the prior art, the on-site monitoring and recording of staff is needed, and the whole-process automatic monitoring is difficult to realize. Therefore, the existing helicopter pesticide application amount measurement method has the defects of low pesticide application amount measurement precision and incapability of full-course automatic monitoring.

Compared with the prior art, the invention can accurately identify the flight state and the spraying state of the helicopter at each sampling moment in the working period by comprehensively analyzing the working data of the helicopter in one working period, further accurately identify and acquire all the pesticide application times according to the flight state and the spraying state at each sampling moment, and determine the pesticide application amount of each pesticide application time based on the spraying flow and the spraying time in each pesticide application time, thereby effectively improving the measuring precision of the pesticide application amount of the helicopter, and realizing automatic pesticide application measurement without on-site monitoring and recording of staff.

According to the method for determining the operation track and the pesticide application amount of the helicopter, the flight state and the spraying state of the helicopter are determined by acquiring the flight state data and the spraying state data of the helicopter, the operation track is determined according to the flight state and the spraying state of the helicopter, all pesticide application frames of the helicopter in an operation period and the spraying flow and the spraying time of each pesticide application frame are further accurately determined, the pesticide application amount of each pesticide application frame is finally determined, the operation state and the track of the helicopter are accurately identified, so that the pesticide application frame is accurately distinguished, the pesticide application amount measurement precision is improved, and the on-site monitoring record of workers is not needed, so that the automatic pesticide application amount measurement is realized.

Based on the content of the foregoing embodiment, as an alternative embodiment, determining, based on the operation data, an operation state of the helicopter at each sampling time in the operation cycle specifically includes:

and inputting the flight state data of each sampling time in the working period into the flight state identification model group so as to determine the flight state of the helicopter at each sampling time in the working period.

And inputting the spraying state data of each sampling time in the working period into a spraying state identification model group so as to determine the spraying state of the helicopter at each sampling time in the working period.

The flight state recognition model set is obtained by training based on a first training sample set, wherein the first training sample set comprises a plurality of flight state data samples and flight state labels corresponding to each flight state data sample.

The spray state recognition model set is trained based on a second training sample set, wherein the second training sample set comprises a plurality of spray state data samples and spray state labels corresponding to each spray state data sample.

Specifically, the working state of each sampling time of the helicopter is determined based on the working data of the helicopter, and the flight state of each sampling time can be determined by inputting the flight state data of each sampling time in the working period into the flight state identification model group. The flight state recognition model group comprises a plurality of flight state recognition models.

Further, the spraying state data of each sampling time in the working period can be input into the spraying state identification model group so as to determine the spraying state of each sampling time. The spray state recognition model group comprises a plurality of spray state recognition models.

Optionally, the set of flight state recognition models is trained based on a first training sample set, where the first training sample set includes a plurality of flight state data samples and a flight state label corresponding to each of the flight state data samples for pre-training the plurality of flight state recognition models.

The specific training process may be to input each flight state data sample and the flight state label to the flight state recognition model at the same time, and adjust model parameters in the flight state recognition model according to each output result of the flight state recognition model, so as to finally complete the pre-training process of the flight state recognition model.

Alternatively, after the flight state recognition model reaches the preset pre-training times, the pre-training process of the flight state recognition model is considered to be completed; and when the training output result of the flight state recognition model converges, the pre-training process of the flight state recognition model can be considered to be completed.

Correspondingly, the spraying state recognition model set is obtained by training based on a second training sample set, the second training sample set comprises a plurality of spraying state data samples and spraying state labels corresponding to each spraying state data sample, the spraying state recognition model set is used for pre-training the spraying state recognition models, and a specific pre-training process can correspond to pre-training of the flight state recognition model set and is not repeated herein.

Alternatively, the flight state recognition model set and the spray state recognition model set may be formed by models trained in historical experiments, wherein model parameters are regarded as default values.

According to the method for determining the helicopter operation track and the pesticide application amount, the operation data of each sampling moment in the operation period is input into the pre-trained flight state identification model group and the spraying state identification model group, so that the operation state of the helicopter at each sampling moment can be accurately identified, and the pesticide application frame can be conveniently distinguished according to the operation state of the helicopter.

Based on the content of the above embodiments, as an alternative embodiment, the set of flight state recognition models includes at least one of the following plurality of flight state recognition models: a shutdown state recognition model for discriminating whether the helicopter is in a shutdown state, a cruise state recognition model for discriminating whether the helicopter is in a cruise state, a hover state recognition model for discriminating whether the helicopter is in a hover state, a climb state recognition model for discriminating whether the helicopter is in a climb state, a descent state recognition model for discriminating whether the helicopter is in a descent state, a vertical ascent state recognition model for discriminating whether the helicopter is in a vertical ascent state, and a vertical descent state recognition model for discriminating whether the helicopter is in a vertical descent state.

The flight state data of each sampling time in the working period is input into a flight state identification model group to determine the flight state of the helicopter at each sampling time in the working period, and the method specifically comprises the following steps:

and respectively inputting the flight state data of each sampling moment into each flight state recognition model, and acquiring the confidence coefficient of the flight state recognition result output by each flight state recognition model.

The maximum value of all the confidence values is determined as the first feature value.

If the first characteristic value is larger than a first preset threshold value, determining a flight state recognition result corresponding to the first characteristic value as a flight state of the helicopter at the sampling moment.

Otherwise, determining the flight state of the helicopter at the sampling moment as other flight states.

Specifically, the flight state recognition model set includes at least one of a plurality of flight state recognition models, specifically, a shutdown state recognition model for judging whether the helicopter is in a shutdown state, a cruise state recognition model for judging whether the helicopter is in a cruise state, a hover state recognition model for judging whether the helicopter is in a hover state, a climb state recognition model for judging whether the helicopter is in a climb state, a descent state recognition model for judging whether the helicopter is in a descent state, a vertical ascent state recognition model for judging whether the helicopter is in a vertical ascent state, and a vertical descent state recognition model for judging whether the helicopter is in a vertical descent state.

Wherein, the shutdown state represents that the helicopter is static on the apron, the cruise state represents that the helicopter flies horizontally, the hover state represents that the helicopter hovers in the air, the climbing state represents that the helicopter climbs forward, the descending state represents that the helicopter descends forward, the vertical ascending state represents that the helicopter ascends vertically, and the vertical descending state represents that the helicopter descends vertically.

Further, the flight state data of each sampling moment are respectively input into each flight state recognition model, and the confidence coefficient of the flight state recognition result output by each flight state recognition model is obtained. Each of the flight state recognition models may be pre-trained by using a support vector machine (Support Vector Machine, SVM) as an initial model, for example, the shutdown state recognition model may be obtained by training a support vector machine through a plurality of flight state data samples and whether the corresponding shutdown state is a label.

And when the confidence coefficient of the flight state recognition result output by each flight state recognition model is obtained, determining the maximum value in all the confidence coefficients, and taking the maximum value in the confidence coefficients as a first characteristic value. The confidence degree of the flight state recognition results output by each flight state recognition model can be output in a (0.85,0.15) form or can be directly output in a 0.85 form, and the confidence degree of the flight state recognition results output by the current flight state recognition model is represented by 0.85, and the higher the confidence degree is, the closer the flight state of the helicopter at the current sampling moment is to the flight state judged by the flight state recognition model.

Further, after the first characteristic value is obtained, comparing the magnitude relation between the first characteristic value and a first preset threshold value, and if the first characteristic value is larger than the first preset threshold value, determining a flight state identification result corresponding to the first characteristic value as the flight state of the helicopter at the current sampling moment; otherwise, determining the flight state of the helicopter at the current sampling moment as other flight states. The first preset threshold may be preset according to the usage requirement of a specific scene, for example, may be preset to 0.9; other flight states refer to flight states that do not satisfy the respective flight state identification models described above for discrimination, but rather are other specific flight states such as those of ground transportation.

As an alternative embodiment, the flight status data includes flight ground speed, altitude from the ground, altitude delta from the ground, and triaxial acceleration.

Specifically, the flight state data at each sampling time may include a flight ground speed, an altitude, a ground altitude increment, and a triaxial acceleration, where the triaxial acceleration includes an X-axis acceleration, a Y-axis acceleration, and a Z-axis acceleration, and may further include a flight state identified at the last sampling time, and input as a latest state to each flight state identification model. The height increment from the ground is the difference between the current height from the ground and the height from the ground at the last sampling moment.

Optionally, after the flight status data of each sampling time is acquired, the flight status input data may be constructed in the form of the following vectors:

。

as an alternative embodiment, since the latest state is used to represent the last (i.e. last sampling time) flight state of the helicopter, and mainly includes one of seven flight states, the integer values 1-7 may be used to refer to one of the flight states, and the latest state input data is used to construct the flight state input data.

The weight information of each flight state data may be further acquired, and the weight may be assigned to each flight state data by the following vector form:

。

further, the data processing calculation may be performed on the flight status input data and the weight information by the following formula (1):

（1）

wherein, the liquid crystal display device comprises a liquid crystal display device,inputting data processing results of data vectors for flight status, < >>Inputting a data vector for a flight state; />A weight information vector of data is input for the flight state; "/>"means a transpose operation.

Further, the data processing result of the input data vector is processed by the following formula (2) to obtain the confidence level of the flight state recognition result outputted by the flight state recognition model:

（2）

Wherein, the liquid crystal display device comprises a liquid crystal display device,identifying the result confidence level for the flight state; />As a function of the maximum argument.

For example, after the flight state data at one sampling time is respectively input to each flight state recognition model, the maximum value in the confidence coefficient about the flight state recognition result output by each acquired flight state recognition model is 0.95, the corresponding flight state recognition model is a shutdown state recognition model for judging whether the helicopter is in a shutdown state, the first preset threshold value is 0.9, the first characteristic value is determined to be 0.95, the first characteristic value is compared with the first preset threshold value, the first characteristic value is determined to be larger than the first preset threshold value, the flight state recognition result corresponding to the first characteristic value, namely the flight state of the current helicopter is recognized as the shutdown state, and the shutdown state is determined to be the flight state of the helicopter at the current sampling time.

According to the method for determining the helicopter operation track and the application amount, various flight state data are comprehensively analyzed and are respectively input into the flight state identification models, so that the flight state change process of the helicopter in the operation process can be finely defined, the flight state of each sampling moment can be effectively marked, the accuracy of the flight state classification is effectively improved, and the application period of the helicopter can be conveniently determined later.

Based on the foregoing, as an alternative embodiment, the spray condition recognition model set includes at least one of the following spray condition recognition models: a stopped state recognition model for judging whether the helicopter is in a spraying stopped state, a spraying state recognition model for judging whether the helicopter is in a spraying state, a starting state recognition model for judging whether the helicopter is in a spraying starting state, a closed state recognition model for judging whether the helicopter is in a spraying closed state, a blocking state recognition model for judging whether the helicopter is in a spraying blocking state, a leaking state recognition model for judging whether the helicopter is in a spraying leaking state and a testing state recognition model for judging whether the helicopter is in a spraying testing state.

The spraying state data of each sampling time in the working period is input into a spraying state identification model group to determine the spraying state of the helicopter at each sampling time in the working period, and the method specifically comprises the following steps:

and respectively inputting the spraying state data of each sampling moment into each spraying state identification model, and obtaining the confidence coefficient of the spraying state identification result output by each spraying state identification model.

The maximum value of all the confidence values is determined as the second characteristic value.

If the second characteristic value is larger than a second preset threshold value, determining a spraying state identification result corresponding to the second characteristic value as a spraying state of the helicopter at the sampling moment.

Otherwise, determining the spraying state of the helicopter at the sampling moment as other spraying states.

Specifically, the spraying state recognition model set comprises at least one of a plurality of spraying state recognition models, specifically a stopping state recognition model for judging whether the helicopter is in a spraying stopping state, a spraying state recognition model for judging whether the helicopter is in a spraying state, a starting state recognition model for judging whether the helicopter is in a spraying starting state, a closing state recognition model for judging whether the helicopter is in a spraying closing state, a blocking state recognition model for judging whether the helicopter is in a spraying blocking state, a leaking state recognition model for judging whether the helicopter is in a spraying leaking state and a testing state recognition model for judging whether the helicopter is in a spraying testing state.

The spraying stopping state represents a stable state of stopping spraying of the helicopter spraying system, the spraying state represents a stable state of normal spraying of the helicopter spraying system, the starting state represents a transient process from stopping to spraying of the helicopter spraying system, the closing state represents a transient process from spraying to stopping of the helicopter spraying system, the spraying blocking state represents an error state of blocking a spray head of a helicopter part, the spraying liquid leakage state represents an error state of leaking liquid of the spray head of the helicopter part, and the spraying test state represents a spraying test state of the helicopter when the ground is stationary.

Further, the spraying state data of each sampling time are respectively input into each spraying state identification model, and the confidence coefficient of the spraying state identification result output by each spraying state identification model is obtained. Each spray state recognition model may be obtained by training a support vector machine as an initial model in advance, for example, the stop state recognition model may be obtained by training a support vector machine through a plurality of spray state data samples and whether the corresponding spray stop state is a label.

And when the confidence coefficient of the spray state recognition result output by each spray state recognition model is obtained, determining the maximum value in all the confidence coefficients, and taking the maximum value in the confidence coefficients as a second characteristic value. The confidence of the spray state recognition results output by each spray state recognition model can be output in a (0.85,0.15) form or can be directly output in a 0.85 form, and the confidence of the spray state recognition results output by the current spray state recognition model is represented by 0.85, and the higher the confidence is, the closer the spray state of the helicopter at the current sampling moment is to the spray state judged by the spray state recognition model.

Further, after the second characteristic value is obtained, comparing the magnitude relation between the second characteristic value and a second preset threshold value, and if the second characteristic value is larger than the second preset threshold value, determining a spraying state identification result corresponding to the second characteristic value as a spraying state of the helicopter at the current sampling moment; otherwise, determining the spraying state of the helicopter at the current sampling moment as other spraying states. The second preset threshold may be preset according to the usage requirement of the specific scene, for example, may be preset to 0.9; other spray conditions refer to spray conditions that do not satisfy the respective spray condition recognition models described above for discrimination, but are other specific spray conditions, such as spray head maintenance spray conditions.

As an alternative embodiment, the spray status data includes the ground speed, altitude, spray flow, flow trend, spray pressure and pressure trend, and geometric mean of triaxial acceleration.

Specifically, the spraying state data at each sampling time may include a geometric average value of the ground speed, the altitude, the spraying flow, the flow trend, the spraying pressure and the pressure trend, and the triaxial acceleration, and may further include a spraying state identified at the last sampling time as a latest state, and the flight state at the current sampling time are input to each spraying state identification model together. Wherein, the flow trend is the flow difference between the current spraying flow and the spraying flow at the last sampling moment; the pressure trend is the pressure difference between the current spray pressure and the spray pressure at the last sampling instant.

Optionally, after the spraying state data of each sampling time is acquired, the spraying state input data may be constructed in the form of the following vectors:

。

as an alternative embodiment, since the latest state is a spraying state used for characterizing the latest (i.e. last sampling time) of the helicopter, and mainly includes one of seven spraying states, each of the seven spraying states may be referred to by an integer value of 1-7, and the latest state is used as input data corresponding to the latest state to construct the spraying state input data.

The weight information of each spraying state data may be further acquired, and the weight may be assigned to each spraying state data by the following vector form:

。

further, the data processing calculation can be performed on the spraying state input data and the weight information by the following formula (3):

（3）

wherein, the liquid crystal display device comprises a liquid crystal display device,inputting the data processing result of the data vector for the spraying state, < >>Inputting a data vector for a spraying state; />A weight information vector for inputting data for the spraying state; "/>"means a transpose operation.

Further, the data processing result of the spray state input data vector is processed by the following formula (4) to obtain the confidence level of the spray state recognition result outputted by the spray state recognition model:

（4）

Wherein, the liquid crystal display device comprises a liquid crystal display device,identifying the result confidence level for the spraying state; />As a function of the maximum argument.

For example, after the spraying state data at one sampling time is respectively input to each spraying state identification model, the maximum value in the confidence coefficient about the spraying state identification result output by each obtained spraying state identification model is 0.92, the corresponding spraying state identification model is a stopping state identification model for judging whether the helicopter is in a spraying stopping state, the second preset threshold value is 0.9, the second characteristic value is 0.92 and the second preset threshold value is 0.9, the second characteristic value is determined to be larger than the second preset threshold value, the spraying state identification result corresponding to the second characteristic value, namely, the spraying state of the current helicopter is identified as the stopping state, and the stopping state is determined to be the spraying state of the helicopter at the current sampling time.

According to the method for determining the operation track and the application amount of the helicopter, provided by the invention, the plurality of spraying state data are comprehensively analyzed and respectively input into the plurality of spraying state identification models, so that the plurality of spraying states of the helicopter in the operation process are finely defined, the spraying state change process in the helicopter operation process can be described in detail, the spraying state at each sampling moment is effectively marked, the accuracy of spraying state classification is effectively improved, and the application times of the helicopter are conveniently determined subsequently.

Based on the foregoing embodiment, as an alternative embodiment, after obtaining the dose of the helicopter in each application rack according to the spraying flow rate and the spraying duration in each application rack, the method further includes:

if the absolute value of the deviation between the drug application amount of the helicopter in any drug application rack and the capacity of the drug box of the helicopter is determined, and the absolute value of the deviation is larger than a third preset threshold value, marking any drug application rack as a rack to be calibrated.

Taking the flight state data of the rack to be calibrated as a training sample, acquiring the real flight state corresponding to the flight state data of the rack to be calibrated as a tag, and training the flight state recognition model group again.

Taking the spraying state data of the rack to be calibrated as a training sample, acquiring the real spraying state corresponding to the spraying state data of the rack to be calibrated as a label, and training the spraying state identification model group again.

Specifically, after the medicine application amount of the helicopter in each medicine application rack is obtained, the absolute value of the deviation between the medicine application amount of any medicine application rack and the medicine box capacity of the helicopter can be compared, and when the absolute value of the deviation is larger than a third preset threshold value, any medicine application rack is marked as the rack to be calibrated. The third preset threshold may be preset according to the usage requirement of a specific scene, for example, may be preset to be 5%.

In the process of the pesticide application operation of the helicopter, when the absolute value of the deviation between the pesticide application amount in any pesticide application rack and the pesticide box capacity of the helicopter is larger than a third preset threshold value, the deviation is often caused by the following two reasons: 1. the pesticide box is not filled with pesticide, and then the pesticide application operation is started; 2. the error of the recognition results of the flight state recognition model group and the spraying state recognition model group is larger.

Therefore, any medicine application rack with the mark deviation absolute value larger than the third preset threshold value is the rack to be calibrated, and the medicine adding condition of the helicopter can be confirmed through manual analysis so as to determine whether the medicine application rack works for full medicine boxes.

Further, all the flight state data of the rack to be calibrated can be obtained again to serve as training samples, and the real flight state corresponding to the flight state data of the rack to be calibrated can be obtained to serve as a label as the identification result of the flight state identification model group is wrong, so that each flight state identification model in the flight state identification model group is trained again.

The method comprises the steps that various flight state data and corresponding real flight states can be obtained through experiments in advance to serve as a matching table, and the real flight states are used as labels; the real flight state of the rack to be calibrated can be manually marked as a label.

Optionally, when the flight state data of the rack to be calibrated is obtained as a training sample and the real flight state corresponding to the flight state data of the rack to be calibrated is obtained as a label, the flight state identification model group is trained again, the updating training of the flight state identification model group can be completed through a sequence minimum optimization algorithm (Sequential minimal optimization, SMO).

The SMO algorithm can be summarized as the following steps:

1. iterating for each parameter vector in the dataset for training;

2. if the parameter vector can be optimized, then randomly selecting another vector while optimizing both vectors;

3. if not traversing all vectors, returning to the step 1;

4. otherwise, the iteration number is +1;

5. and (3) returning to the step (1) when the iteration times are smaller than the preset maximum iteration times.

Further, all spraying state data of the rack to be calibrated can be correspondingly obtained again to serve as training samples, and the real spraying state corresponding to the spraying state data of the rack to be calibrated can be obtained to serve as a label as the identification result of the spraying state identification model group is wrong, so that each spraying state identification model in the spraying state identification model group is trained again. Wherein, the updating training of the spraying state recognition model group can be completed based on the SMO algorithm.

Therefore, the method and the device for determining the helicopter operation track and the pesticide application rate can dynamically update the flight state identification model group and the spraying state identification model group by acquiring the deviation absolute value of the pesticide application rate of each pesticide application frame and the pesticide box capacity of the helicopter, further determining the training sample and the label for updating the flight state identification model group and the spraying state identification model group, effectively improving the identification accuracy of the helicopter operation state, further being beneficial to improving the identification accuracy of the subsequent helicopter pesticide application frame, improving the pesticide application rate measurement accuracy of the helicopter, and greatly improving the applicability of the method for determining the helicopter operation track and the pesticide application rate.

According to the method for determining the helicopter operation track and the pesticide application amount, provided by the invention, the deviation absolute value of the pesticide application amount of each pesticide application frame and the pesticide box capacity of the helicopter is obtained, so that the training sample and the label for updating the training flight state identification model group and the spraying state identification model group are further determined, the dynamic updating of the flight state identification model group and the spraying state identification model group is realized, the identification accuracy of the helicopter operation state can be effectively improved, the identification accuracy of the subsequent pesticide application frame of the helicopter is further improved, the accuracy of the helicopter pesticide application amount measurement is improved, and the applicability of the method for determining the helicopter operation track and the pesticide application amount provided by the invention is greatly improved.

Based on the content of the above embodiment, as an alternative embodiment, further includes:

and acquiring the triaxial acceleration of the helicopter at any sampling time in the working period.

Based on the triaxial acceleration, the flying ground speed and the flying direction of the helicopter at any sampling moment are obtained.

And determining the longitude and latitude information of the helicopter at any sampling moment according to the flying ground speed and the flying direction of the helicopter at any sampling moment and the longitude and latitude information of the helicopter at the last sampling moment.

Specifically, in the aviation drug application process of the helicopter, the situation that GNSS signals are lost easily caused by the complex terrain environment in mountain areas and the change of the flight attitude of the helicopter occurs, so that the positioning information of the helicopter is lost.

Therefore, the method and the device have the advantages that the problem of GNSS failure caused by the terrain and the flight attitude is solved by acquiring the triaxial acceleration of the helicopter and based on the inertial data.

The acquisition of longitude and latitude information of the helicopter at any sampling moment can be realized by the following specific steps:

1. acquiring the triaxial acceleration of the helicopter at any sampling time in the working period, and measuring the triaxial acceleration of the helicopter at any sampling time by using an MEMS (Micro Electro Mechanical Systems, MEMS) accelerometer;

2. After the triaxial acceleration of any sampling moment of the helicopter is obtained, carrying out data processing on the triaxial acceleration, extracting acceleration amplitude values and direction information in three directions, and determining the flying ground speed and the flying direction of the helicopter at any sampling moment based on the flying ground speed and the flying direction of the last sampling moment;

3. according to the obtained flying ground speed and flying direction of the helicopter at any sampling time and the longitude and latitude information of the helicopter at the last sampling time, the actual geographic position information, namely the longitude and latitude information, of the helicopter at any sampling time can be rapidly measured.

According to the method for determining the operation track and the application amount of the helicopter, the triaxial acceleration of the helicopter at any sampling time is obtained, the longitude and latitude information of the helicopter at the current sampling time is determined according to the longitude and latitude information of the helicopter at the last sampling time, so that the problem of GNSS failure caused by the terrain and the flight attitude is solved through inertial data, and the reliability of flight state data of the helicopter is improved.

Finally, in order to more clearly illustrate the method for determining the operation track and the application rate of the helicopter provided by the invention, the following embodiments are specifically provided:

Fig. 2 is a second flow chart of a method for determining a helicopter operation track and a pesticide application rate according to the present invention, as shown in fig. 2, including:

1. firstly, acquiring operation data of a helicopter in an operation period through an airborne terminal installed on the helicopter;

2. all the job data are packaged and sent to a server side, and the server can establish a plurality of data structures based on the received job data, for example:

a. the helicopter parameter table is used for recording helicopter model, medicine box capacity, spray head number, effective breadth, atomization pressure, rated flow and the like;

b. the helicopter operation data sheet is used for recording UTC time, longitude, latitude, flying ground speed, flying direction, altitude, ground height, spraying flow, spraying pressure, triaxial acceleration and the like;

c. the take-off and landing point record table is used for recording longitude and latitude information of each take-off and landing point of the helicopter and a stopping range, wherein the default range is 100 meters and can be adjusted according to requirements;

d. the system comprises a rack number recording table, a rack number recording table and a rack number recording table, wherein the rack number recording table is used for recording the starting time, the ending time, the medicine adding amount of a medicine box, the average operation speed, the average operation height, the spraying rate, the yes/no medicine box filling operation, the rack number type and the like of each rack number;

e. A flight status recognition model set and a spray status recognition model set.

3. If the GNSS positioning failure problem exists in the operation data, the latitude and longitude information of the helicopter is lost, and the latitude and longitude information of the helicopter is repositioned and acquired through the triaxial acceleration;

4. the operation data of each sampling moment is input into a flight state identification model group and a spraying state identification model group, the operation state of each sampling moment of the helicopter is determined, the corresponding operation state is marked, and the operation track of the helicopter is determined based on the operation state of each sampling moment;

5. determining all pesticide application frames of the helicopter based on the operation track of the helicopter, wherein each time of flight state is a place of shutdown, updating the place of shutdown into a take-off and landing point record table, marking the place of shutdown as a pesticide adding point, marking the frame as a pesticide application frame if state marks such as spraying, starting and closing exist in the frames between two adjacent shutdown states, recording the frame in the frame record table, and marking the frame as a transition/test frame if the state marks such as spraying, starting and closing are not involved;

6. after all the pesticide application frames of the helicopter are determined, determining the pesticide application amount of each pesticide application frame according to the spraying flow and the spraying time length of each pesticide application frame;

7. Further, comparing the medicine application amount of each medicine application rack with the medicine box capacity of the helicopter, if the absolute value of the deviation is larger than a third preset threshold value, marking the rack as the rack to be calibrated, and manually analyzing the actual condition of the rack to be calibrated again;

8. and updating the training flight state recognition model group and the spraying state recognition model group according to the operation data of the to-be-calibrated frame and the real operation state, so that the recognition accuracy of the row state recognition model group and the spraying state recognition model group is improved.

Fig. 3 is a schematic structural diagram of the device for determining the operation track and the drug application rate of the helicopter, and as shown in fig. 3, the device mainly comprises: a data acquisition unit 31, a trajectory determination unit 32, an order determination unit 33, and a dose acquisition unit 34, wherein:

a data acquisition unit 31 for acquiring operation data of the helicopter in one operation cycle, the operation data including flight status data and spraying status data.

A trajectory determining unit 32 for determining an operation state of the helicopter at each sampling time in the operation period based on the operation data, to determine an operation trajectory of the helicopter in the operation period based on the operation state; the operational conditions include a flight condition and a spray condition.

An installment determining unit 33 is configured to determine, based on a working track of the helicopter, all application installments of the helicopter in the working cycle, and a spraying flow rate and a spraying time period in each application installment.

And the medicine amount acquisition unit 34 is used for acquiring the medicine amount of the helicopter in each medicine application frame according to the spraying flow and the spraying time in each medicine application frame.

It should be noted that, when the device for determining the operation track and the application rate of the helicopter provided by the embodiment of the invention is specifically operated, the method for determining the operation track and the application rate of the helicopter described in any one of the above embodiments may be executed, and the description of this embodiment is omitted.

According to the device for determining the operation track and the pesticide application amount of the helicopter, the flight state and the spraying state of the helicopter are determined by acquiring the flight state data and the spraying state data of the helicopter, the operation track is determined according to the flight state and the spraying state of the helicopter, all pesticide application frames of the helicopter in an operation period and the spraying flow and the spraying duration of each pesticide application frame are further accurately determined, the pesticide application amount of each pesticide application frame is finally determined, the operation state and the track of the helicopter are accurately identified, so that the pesticide application frame is accurately distinguished, the pesticide application amount measurement precision is improved, and the on-site monitoring record of workers is not needed, so that the automatic pesticide application amount measurement is realized.

Fig. 4 is a schematic structural diagram of an electronic device according to the present invention, as shown in fig. 4, the electronic device may include: processor 410, communication interface (Communications Interface) 420, memory 430 and communication bus 440, wherein processor 410, communication interface 420 and memory 430 communicate with each other via communication bus 440. Processor 410 may invoke logic instructions in memory 430 to perform a method of determining a helicopter work trajectory and a dosage, the method comprising: acquiring operation data of the helicopter in one operation period, wherein the operation data comprise flight state data and spraying state data; determining an operation state of the helicopter at each sampling time in the operation period based on the operation data, so as to determine an operation track of the helicopter in the operation period based on the operation state; the working state comprises a flight state and a spraying state; determining all pesticide application frames of the helicopter in the working period, and the spraying flow and the spraying duration in each pesticide application frame based on the working track of the helicopter; and acquiring the application amount of the helicopter in each application rack according to the spraying flow and the spraying time in each application rack.

Further, the logic instructions in the memory 430 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, are capable of performing the method of determining a helicopter work trajectory and a drug application provided by the above embodiments, the method comprising: acquiring operation data of the helicopter in one operation period, wherein the operation data comprise flight state data and spraying state data; determining an operation state of the helicopter at each sampling time in the operation period based on the operation data, so as to determine an operation track of the helicopter in the operation period based on the operation state; the working state comprises a flight state and a spraying state; determining all pesticide application frames of the helicopter in the working period, and the spraying flow and the spraying duration in each pesticide application frame based on the working track of the helicopter; and acquiring the application amount of the helicopter in each application rack according to the spraying flow and the spraying time in each application rack.

In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the method of determining a helicopter work trajectory and a drug delivery rate provided by the above embodiments, the method comprising: acquiring operation data of the helicopter in one operation period, wherein the operation data comprise flight state data and spraying state data; determining an operation state of the helicopter at each sampling time in the operation period based on the operation data, so as to determine an operation track of the helicopter in the operation period based on the operation state; the working state comprises a flight state and a spraying state; determining all pesticide application frames of the helicopter in the working period, and the spraying flow and the spraying duration in each pesticide application frame based on the working track of the helicopter; and acquiring the application amount of the helicopter in each application rack according to the spraying flow and the spraying time in each application rack.

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of determining a helicopter operational trajectory and a dosage comprising:

acquiring operation data of the helicopter in one operation period, wherein the operation data comprise flight state data and spraying state data;

determining an operation state of the helicopter at each sampling time in the operation period based on the operation data, so as to determine an operation track of the helicopter in the operation period based on the operation state; the working state comprises a flight state and a spraying state;

determining all pesticide application frames of the helicopter in the working period, and the spraying flow and the spraying duration in each pesticide application frame based on the working track of the helicopter;

and acquiring the application amount of the helicopter in each application rack according to the spraying flow and the spraying time in each application rack.

2. The method for determining the working track and the drug application rate of the helicopter according to claim 1, wherein the determining the working state of the helicopter at each sampling time in the working period based on the working data specifically comprises:

inputting the flight state data of each sampling time in the working period to a flight state identification model group so as to determine the flight state of the helicopter at each sampling time in the working period;

Inputting the spraying state data of each sampling time in the working period to a spraying state identification model group so as to determine the spraying state of the helicopter at each sampling time in the working period;

the flight state recognition model set is obtained by training based on a first training sample set, wherein the first training sample set comprises a plurality of flight state data samples and flight state labels corresponding to each flight state data sample;

the spray state recognition model group is obtained by training based on a second training sample set, and the second training sample set comprises a plurality of spray state data samples and spray state labels corresponding to each spray state data sample.

3. The method of determining a helicopter operational trajectory and dosage of claim 2, wherein the set of flight state identification models comprises at least one of the following plurality of flight state identification models: a shutdown state recognition model for discriminating whether the helicopter is in a shutdown state, a cruise state recognition model for discriminating whether the helicopter is in a cruise state, a hover state recognition model for discriminating whether the helicopter is in a hover state, a climb state recognition model for discriminating whether the helicopter is in a climb state, a descent state recognition model for discriminating whether the helicopter is in a descent state, a vertical ascent state recognition model for discriminating whether the helicopter is in a vertical ascent state, and a vertical descent state recognition model for discriminating whether the helicopter is in a vertical descent state;

The step of inputting the flight state data of each sampling time in the working period into a flight state identification model group to determine the flight state of the helicopter at each sampling time in the working period specifically comprises the following steps:

respectively inputting the flight state data of each sampling moment into each flight state identification model, and acquiring the confidence coefficient of the flight state identification result output by each flight state identification model;

determining the maximum value of all the confidence coefficients as a first characteristic value;

if the first characteristic value is larger than a first preset threshold value, determining a flight state identification result corresponding to the first characteristic value as a flight state of the helicopter at the sampling moment;

4. A method of determining a helicopter operational trajectory and dosage as claimed in claim 3, wherein said flight status data comprises ground speed, altitude from the ground delta and triaxial acceleration.

5. The method of determining a helicopter operational trajectory and dosage of claim 2 wherein said set of spray condition identification models comprises at least one of the following plurality of spray condition identification models: a stopped state recognition model for judging whether the helicopter is in a spraying stopped state, a spraying state recognition model for judging whether the helicopter is in a spraying state, a starting state recognition model for judging whether the helicopter is in a spraying starting state, a closed state recognition model for judging whether the helicopter is in a spraying closed state, a blocking state recognition model for judging whether the helicopter is in a spraying blocking state, a leaking state recognition model for judging whether the helicopter is in a spraying leaking state and a testing state recognition model for judging whether the helicopter is in a spraying testing state;

The step of inputting the spraying state data of each sampling time in the working period into a spraying state identification model group so as to determine the spraying state of the helicopter at each sampling time in the working period, specifically comprising the following steps:

the spraying state data at each sampling moment are respectively input into each spraying state identification model, and the confidence coefficient of the spraying state identification result output by each spraying state identification model is obtained;

determining the maximum value of all the confidence coefficients as a second characteristic value;

if the second characteristic value is larger than a second preset threshold value, determining a spraying state identification result corresponding to the second characteristic value as a spraying state of the helicopter at the sampling moment;

otherwise, determining that the spraying state of the helicopter at the sampling moment is other spraying states.

6. The method of determining helicopter operational trajectories and drug delivery rates of claim 5, wherein the spray status data includes geometric mean of ground speed, altitude, spray flow, flow trend, spray pressure and pressure trend, and tri-axial acceleration.

7. The method of determining a trajectory and dosage of a helicopter according to claim 2, further comprising, after obtaining the dosage of said helicopter in each dosage frame based on said spray flow rate and said spray duration in each dosage frame:

If the absolute value of the deviation between the drug application amount of the helicopter in any drug application rack and the capacity of a drug box of the helicopter is determined, and the absolute value of the deviation is larger than a third preset threshold value, marking any drug application rack as a rack to be calibrated;

taking the flight state data of the rack to be calibrated as a training sample, acquiring a real flight state corresponding to the flight state data of the rack to be calibrated as a tag, and training the flight state identification model group again;

taking the spraying state data of the rack to be calibrated as a training sample, acquiring a real spraying state corresponding to the spraying state data of the rack to be calibrated as a label, and training the spraying state identification model group again.

8. The method of determining a helicopter operational trajectory and dosage as claimed in any one of claims 1-7, further comprising:

acquiring triaxial acceleration of the helicopter at any sampling moment in the working period;

based on the triaxial acceleration, acquiring the flying ground speed and the flying direction of the helicopter at any sampling moment;

9. A device for determining the trajectory and dosage of a helicopter, comprising:

the data acquisition unit is used for acquiring operation data of the helicopter in one operation period, wherein the operation data comprise flight state data and spraying state data;

a track determining unit, configured to determine, based on the operation data, an operation state of the helicopter at each sampling time in the operation period, so as to determine, based on the operation state, an operation track of the helicopter in the operation period; the working state comprises a flight state and a spraying state;

the system comprises an overhead determining unit, a control unit and a control unit, wherein the overhead determining unit is used for determining all pesticide application overhead of the helicopter in the working period, and the spraying flow and the spraying duration in each pesticide application overhead based on the working track of the helicopter;

the medicine amount acquisition unit is used for acquiring the medicine amount of the helicopter in each medicine application frame according to the spraying flow and the spraying time in each medicine application frame.

10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor, when executing the computer program, performs the steps of the method of determining a helicopter operational trajectory and dosage as claimed in any one of claims 1 to 8.

11. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the steps of the method of determining a helicopter work track and a dosage as claimed in any of claims 1 to 8.

12. A computer program product comprising a computer program which, when executed by a processor, implements the steps of the method of determining a helicopter working track and dosage as claimed in any of claims 1 to 8.