Flight data processing method and system

Abstract

The invention relates to the technical field of aviation, in particular to a flight data processing method and system. The flight data processing system comprises a data acquisition module, a flight strategy adjustment module and a flight strategy adjustment model updating module. According to the method, the prediction uncertainty of the airspace state in the space-time analysis is judged through the full-connection operation of random pruning, and the subsequent flight strategy adjustment is guided according to the prediction uncertainty of the airspace state in the space-time analysis, so that the aircraft is in a safer environment when executing a flight task, and the safety of the aircraft is ensured; and uncertain space-time propagation of airspace state prediction is analyzed, subsequent flight strategy adjustment is further guided, and the interpretability of airspace state prediction is enhanced.

Description

Technical Field

[0001] The present invention relates to the field of aviation technology, and in particular to a flight data processing method and system. Background Art

[0002] With the rapid development of aviation technology, aircraft are increasingly used and their tasks are becoming more complex. As a direct reflection of the aircraft's operating status, flight data contains a wealth of flight information and is of great significance to flight safety, efficiency improvement, and airspace management. Effective processing and analysis of flight data, especially airspace status prediction, is a current research hotspot in the aviation field.

[0003] Existing airspace status predictions are generally time series predictions, which do not take into account the uncertainties brought about by changes in spatial and temporal dimensions, especially in complex weather environments, which may lead to safety accidents when aircraft are performing flight missions. Summary of the invention

[0004] The present invention determines the prediction uncertainty of airspace status in spatiotemporal analysis through the fully connected operation of random pruning, and guides the subsequent flight strategy adjustment according to the prediction uncertainty of airspace status in spatiotemporal analysis, so that the aircraft is in a safer environment when performing flight missions, thereby ensuring the safety of the aircraft; and analyzes the spatiotemporal propagation of the uncertainty of airspace status prediction, further guiding the subsequent flight strategy adjustment and enhancing the interpretability of airspace status prediction.

[0005] The present invention provides a flight data processing method, comprising: Acquire the position data and flight data set of the aircraft at the current monitoring time point, the flight data set includes an airspace status data set and an aircraft attitude data set, the position data and the airspace status data set at the current monitoring time point are combined into an airspace status position annotation data set, and the airspace status position annotation data set at the current monitoring time point and the airspace status position annotation data sets at the previous N-1 monitoring time points are combined into an airspace status spatiotemporal analysis data set at the monitoring time point, and then the airspace status spatiotemporal analysis data set and the aircraft attitude data set at the current monitoring time point are sent to the flight strategy adjustment model for processing to output a flight strategy adjustment data set; The flight strategy adjustment model includes an airspace state prediction layer, an uncertainty analysis layer, an uncertainty spatiotemporal propagation layer and a flight strategy adjustment layer. The airspace state prediction layer includes a Transformer unit and a random pruning fully connected unit. The Transformer unit is used to perform time series analysis on the airspace state spatiotemporal analysis data set to output an airspace state spatiotemporal analysis vector. The random pruning fully connected unit is used to perform a random pruning fully connected operation on the airspace state spatiotemporal analysis vector output by the Transformer unit to output an airspace state prediction vector. The uncertainty analysis layer is used to perform a random pruning fully connected operation on the airspace state prediction vector output by several airspace state prediction layers. Uncertainty analysis is performed to construct an uncertainty vector and a reference airspace state prediction vector; the uncertainty spatiotemporal propagation layer is used to perform time series analysis on the uncertainty vector corresponding to the current monitoring time point and the uncertainty vectors corresponding to the previous N-1 monitoring time points to output the uncertainty spatiotemporal propagation vector; the flight strategy adjustment layer is used to splice the aircraft attitude data set, the reference airspace state prediction vector, the uncertainty vector and the uncertainty spatiotemporal propagation vector at the current monitoring time point and send them to the multi-layer perceptron for processing to output the flight strategy adjustment data set; during the use of the flight strategy adjustment model, the flight strategy adjustment model is updated in real time by reinforcement learning.

[0006] Preferably, the random pruning full connection operation specifically includes the following steps: In the process of executing the full connection operation, whenever any neuron node outputs, a random number δ between (0, 1) is generated through a random function, and then the random number δ is compared with the discard probability P. If the random number δ is greater than the discard probability P, the output of the neuron node is set to zero. If the random number δ is not greater than the discard probability P, the zeroing operation is not performed.

[0007] Preferably, uncertainty analysis is performed on the airspace state prediction vectors output by several airspace state prediction layers through an uncertainty analysis layer to construct an uncertainty vector and a reference airspace state prediction vector, which specifically includes the following steps: The operation of sending the spatial state spatiotemporal analysis data set to the spatial state prediction layer for processing is performed several times to obtain several spatial state prediction vectors; Perform a weighted average operation on all airspace state prediction vectors to construct a baseline airspace state prediction vector; then calculate the uncertainty vector based on all airspace state prediction vectors. The uncertainty vector stores the uncertainty value Fi, i=1, 2, 3, ..., I, I is the total number of uncertainty values, which is consistent with the total number of data in the airspace state data set. The uncertainty value Fi is the variance of the i-th data in all airspace state prediction vectors, which is used to characterize the uncertainty of the prediction of the i-th data in the airspace state data set.

[0008] Preferably, during the use of the flight strategy adjustment model, the flight strategy adjustment model is updated in real time by means of reinforcement learning, which specifically includes the following steps: At the monitoring time point, the airspace status spatiotemporal analysis dataset and the aircraft attitude dataset at the current monitoring time point are sent to the target flight strategy adjustment model for processing, and the target flight strategy adjustment dataset is output. In the initial state, the flight strategy adjustment model is consistent with the target flight strategy adjustment model. Then, the airspace status spatiotemporal analysis dataset, the aircraft attitude dataset at the current monitoring time point and the target flight strategy adjustment dataset are spliced ​​to construct reward value analysis data. The reward value analysis data is sent to the reward value analysis network for processing, and the reward value data is output. The gradient value of the reward value data for the target flight strategy adjustment dataset is calculated, and then the parameters of the flight strategy adjustment model are adjusted based on the gradient value using the gradient ascent method to realize real-time update of the flight strategy adjustment model; and the target flight strategy adjustment model is also updated in real time.

[0009] Preferably, the target flight strategy adjustment model is updated in real time, specifically comprising the following steps: Whenever the flight strategy adjustment model is updated, the target flight strategy adjustment model is updated in real time by the following formula: θk(new)=δηk+θk(bef), where k=1, 2, 3,…, K, K is the total number of parameter values ​​in the target flight strategy adjustment model, θk(new) is the kth parameter value in the updated target flight strategy adjustment model, δ is the update rate, which is generally 0.01, ηk is the kth parameter value in the updated flight strategy adjustment model, and θk(bef) is the kth parameter value in the target flight strategy adjustment model before update.

[0010] Preferably, training the flight strategy adjustment model specifically includes the following steps: A number of flight strategy adjustment training samples are obtained, including a data set of airspace status spatiotemporal analysis and a data set of aircraft attitude at corresponding monitoring time points. The flight strategy adjustment training samples are labeled by the flight strategy adjustment data set, and all labeled flight strategy adjustment training samples are combined into a flight strategy adjustment training set. The flight strategy adjustment model is trained by the flight strategy adjustment training set to determine whether the training conditions are met. If the training conditions are met, the trained flight strategy adjustment model is output; otherwise, the flight strategy adjustment model is continued to be trained by the flight strategy adjustment training set.

[0011] Preferably, training the reward value analysis network specifically includes the following steps: Obtain several bonus value analysis positive training samples and bonus value analysis negative training samples, the bonus value analysis positive training samples include airspace state spatiotemporal analysis data set, aircraft attitude data set corresponding to monitoring time points and optimal flight strategy adjustment data set, the bonus value analysis negative training samples include airspace state spatiotemporal analysis data set, aircraft attitude data set corresponding to monitoring time points and the worst flight strategy adjustment data set, annotate the bonus value analysis positive training samples and bonus value analysis negative training samples through bonus value data, form a bonus value analysis training set with all annotated bonus value analysis positive training samples and bonus value analysis negative training samples, train the bonus value analysis network through the bonus value analysis training set, take the bonus value data as the target output, calculate the loss value, and judge whether the loss value is within a preset range. If the loss value is within the preset range, output the trained bonus value analysis network; otherwise, continue to train the bonus value analysis network through the bonus value analysis training set.

[0012] The present invention also provides a flight data processing system, comprising: The data acquisition module is used to obtain the position data and flight data set of the aircraft at the current monitoring time point. The flight data set includes the airspace status data set and the aircraft attitude data set; The flight strategy adjustment module is used to combine the position data of the current monitoring time point and the airspace status data set into an airspace status position annotation data set, and combine the airspace status position annotation data set of the current monitoring time point and the airspace status position annotation data sets of the previous N-1 monitoring time points into an airspace status spatiotemporal analysis data set of the monitoring time point, and then send the airspace status spatiotemporal analysis data set and the aircraft attitude data set of the current monitoring time point into the flight strategy adjustment model for processing to output the flight strategy adjustment data set; The flight strategy adjustment model includes an airspace state prediction layer, an uncertainty analysis layer, an uncertainty spatiotemporal propagation layer and a flight strategy adjustment layer, wherein the airspace state prediction layer includes a Transformer unit and a random pruning fully connected unit, wherein the Transformer unit is used to perform a time series analysis on the airspace state spatiotemporal analysis data set to output an airspace state spatiotemporal analysis vector, and the random pruning fully connected unit is used to perform a random pruning fully connected operation on the airspace state spatiotemporal analysis vector output by the Transformer unit to output an airspace state prediction vector; the uncertainty analysis layer is used to perform uncertainty analysis on the airspace state prediction vectors output by several airspace state prediction layers to construct an uncertainty vector and a reference airspace state prediction vector; the uncertainty spatiotemporal propagation layer is used to perform a time series analysis on the uncertainty vector corresponding to the current monitoring time point and the uncertainty vector corresponding to the previous N-1 monitoring time points to output an uncertainty spatiotemporal propagation vector; the flight strategy adjustment layer is used to splice the aircraft attitude data set at the current monitoring time point, the reference airspace state prediction vector, the uncertainty vector and the uncertainty spatiotemporal propagation vector and send them to a multi-layer perceptron for processing to output a flight strategy adjustment data set; The flight strategy adjustment model update module is used to update the flight strategy adjustment model in real time by means of reinforcement learning during the use of the flight strategy adjustment model.

[0013] The present invention has the following advantages: The present invention determines the prediction uncertainty of airspace status in spatiotemporal analysis through the fully connected operation of random pruning, and guides the subsequent flight strategy adjustment according to the prediction uncertainty of airspace status in spatiotemporal analysis, so that the aircraft is in a safer environment when performing flight missions, thereby ensuring the safety of the aircraft; and analyzes the spatiotemporal propagation of the uncertainty of airspace status prediction, further guiding the subsequent flight strategy adjustment and enhancing the interpretability of airspace status prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A schematic diagram of the structure of a flight data processing system used in an embodiment of the present invention. DETAILED DESCRIPTION

[0015] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0016] Embodiment 1, a flight data processing method, comprising: The position data and flight data set of the aircraft at the current monitoring time point are obtained. The flight data set includes an airspace state data set and an aircraft attitude data set. The airspace state data set here is the environmental data of the position of the aircraft at the current monitoring time point, such as wind speed, visibility and turbulence intensity, etc. These environmental data will affect the aircraft's execution of flight missions. The aircraft attitude data set includes the attitude data of the aircraft itself, such as heading angle and flight speed, etc. The position data and the airspace state data set at the current monitoring time point are combined into an airspace state position annotation data set, and the airspace state position annotation data set at the current monitoring time point and the airspace state position annotation data sets at the previous N-1 monitoring time points are combined into an airspace state spatiotemporal analysis data set at the monitoring time point. The airspace state spatiotemporal analysis data set and the aircraft attitude data set at the current monitoring time point are then sent to a flight strategy adjustment model for processing to output a flight strategy adjustment data set. The flight strategy adjustment data set corresponds to an aircraft attitude data set, such as a heading adjustment strategy and a flight speed adjustment strategy. The aircraft is controlled based on the aircraft attitude data set at the current monitoring time point and the flight strategy adjustment data set output by the flight strategy adjustment model to achieve flight state adjustment of the aircraft that is adaptive to the environment. The flight strategy adjustment model includes an airspace state prediction layer, an uncertainty analysis layer, an uncertainty spatiotemporal propagation layer, and a flight strategy adjustment layer. The airspace state prediction layer includes a Transformer unit and a random pruned fully connected unit. The Transformer unit is established based on the Transformer model and is used to perform time series analysis on the airspace state spatiotemporal analysis dataset to output an airspace state spatiotemporal analysis vector. Since the airspace state position annotation dataset in the airspace state spatiotemporal analysis dataset includes position data, the spatial changes can also be analyzed during the time series analysis. The random pruned fully connected unit is used to The spatial state spatiotemporal analysis vector output by the Transformer unit is randomly pruned and fully connected. The random pruning and fully connected operation here refers to randomly discarding some neuron nodes based on the discard probability during the full connection operation, that is, setting the output of the discarded neuron nodes to zero to output the spatial state prediction vector; the uncertainty analysis layer is used to perform uncertainty analysis on the spatial state prediction vectors output by several spatial state prediction layers to construct an uncertainty vector and a reference spatial state prediction vector. Since the space dimension and time dimension of the aircraft will change during the execution of the flight mission, the conventional Time series analysis cannot make accurate predictions. Therefore, the uncertainty of airspace state prediction is introduced. Based on the uncertainty of airspace state prediction, the aircraft strategy is more reasonably controlled to ensure the safety of the aircraft during flight. If the prediction uncertainty is high, the aircraft should output a more conservative flight strategy, such as a safer flight route and a lower flight speed; the uncertainty spatiotemporal propagation layer is used to perform time series analysis on the uncertainty vector corresponding to the current monitoring time point and the uncertainty vector corresponding to the previous N-1 monitoring time points to output the uncertainty spatiotemporal propagation vector. It should be noted that the uncertainty vectors of N monitoring time points are analyzed in time series. , which can analyze the characteristics of the uncertainty propagation of airspace state prediction in time and space. For example, the uncertain change of wind speed will affect the subsequent wind speed prediction along the time dimension, and affect the surrounding wind speed prediction along the spatial dimension, thereby more accurately guiding the subsequent aircraft strategy adjustment; the flight strategy adjustment layer is used to splice the aircraft attitude data set, the reference airspace state prediction vector, the uncertainty vector and the uncertainty space-time propagation vector at the current monitoring time point and send them to the multi-layer perceptron for processing to output the flight strategy adjustment data set; during the use of the flight strategy adjustment model, the flight strategy adjustment model is updated in real time by reinforcement learning; The present application determines the prediction uncertainty of the airspace state in the spatiotemporal analysis through the fully connected operation of random pruning, and guides the subsequent flight strategy adjustment based on the prediction uncertainty of the airspace state in the spatiotemporal analysis, so that the aircraft is in a safer environment when performing the flight mission, thereby ensuring the safety of the aircraft; and analyzes the spatiotemporal propagation of the uncertainty of the airspace state prediction, further guiding the subsequent flight strategy adjustment and enhancing the interpretability of the airspace state prediction.

[0017] The random pruning fully connected operation includes the following steps: In the process of performing the full connection operation, whenever any neuron node outputs, a random number δ between (0, 1) is generated through a random function, and then the random number δ is compared with the discard probability P. The discard probability P is set by the developer and is generally 0.5. If the random number δ is greater than the discard probability P, the output of the neuron node is set to zero. If the random number δ is not greater than the discard probability P, the zeroing operation is not performed. It should be noted that the full connection operation is generally performed through several fully connected layers, each of which is composed of several neuron nodes. The uncertainty analysis layer performs uncertainty analysis on the airspace state prediction vectors output by several airspace state prediction layers to construct uncertainty vectors and reference airspace state prediction vectors, which specifically includes the following steps: The operation of sending the spatial state spatiotemporal analysis data set to the spatial state prediction layer for processing is performed several times, and in this embodiment, it is performed 100 times to obtain several spatial state prediction vectors. It should be noted that due to the random pruning full connection operation, these spatial state prediction vectors should be consistent when the spatiotemporal changes are stable. If these spatial state prediction vectors are inconsistent, it means that the accuracy of the time series prediction is low, that is, the uncertainty of the prediction; Perform a weighted average operation on all airspace state prediction vectors to construct a benchmark airspace state prediction vector; then calculate an uncertainty vector based on all airspace state prediction vectors, in which the uncertainty value Fi is stored, i=1, 2, 3, ..., I, I is the total number of uncertainty values, which is consistent with the total number of data in the airspace state data set, and the uncertainty value Fi is the variance of the i-th data in all airspace state prediction vectors, which is used to characterize the uncertainty of the prediction of the i-th data in the airspace state data set; During the use of the flight strategy adjustment model, the flight strategy adjustment model is updated in real time through reinforcement learning, which specifically includes the following steps: At the monitoring time point, the airspace state spatiotemporal analysis data set and the aircraft attitude data set at the current monitoring time point are sent to the target flight strategy adjustment model for processing, and the target flight strategy adjustment data set is output. In the initial state, the flight strategy adjustment model is consistent with the target flight strategy adjustment model. Then, the airspace state spatiotemporal analysis data set, the aircraft attitude data set at the current monitoring time point and the target flight strategy adjustment data set are spliced ​​to construct reward value analysis data, and the reward value analysis data is sent to the reward value analysis network for processing. The reward value analysis network is established based on the BP neural network model, and the reward value data is output. The reward value data can reflect the safety of the aircraft after the aircraft executes the flight strategy adjustment data set. The gradient value of the reward value data for the target flight strategy adjustment data set is calculated, and then the parameters of the flight strategy adjustment model are adjusted based on the gradient value using the gradient ascent method to realize the real-time update of the flight strategy adjustment model. It should be noted that the parameter adjustment is performed through the target flight strategy adjustment data set output by the target flight strategy adjustment model, which can ensure that the flight strategy adjustment model is stably updated under the premise that the target flight strategy adjustment model is used as the baseline; and the target flight strategy adjustment model is also updated in real time. The target flight strategy adjustment model is updated in real time, including the following steps: Whenever the flight strategy adjustment model is updated, the target flight strategy adjustment model is updated in real time using the following formula: θ k (new) = δη k +θ k (bef), where k = 1, 2, 3, ..., K, K is the total number of parameter values ​​in the target flight strategy adjustment model, θ k (new) is the kth parameter value in the updated target flight strategy adjustment model, δ is the update rate, which is generally 0.01, and η k Adjust the kth parameter value in the model for the updated flight strategy, θ k (bef) the kth parameter value in the target flight strategy adjustment model before updating; Training the flight strategy adjustment model includes the following steps: A number of flight strategy adjustment training samples are obtained. The flight strategy adjustment training samples include a data set of airspace state spatiotemporal analysis and a data set of aircraft attitude at corresponding monitoring time points. It should be noted that these data sets of airspace state spatiotemporal analysis and the data set of aircraft attitude at corresponding monitoring time points are obtained by developers based on aircraft simulation flight tests. The aircraft simulation flight tests are conducted through flight simulation software (such as Simulink), and the simulation environment is randomly set by probability. The flight strategy adjustment training samples are annotated through the flight strategy adjustment data set. It should be noted that the annotated flight strategy adjustment data set here is the optimal setting obtained by the developer in a random environment based on expert experience, and the aircraft simulation flight test here can be used as a sample source for transfer training. In actual use, further training is generally performed based on the flight strategy adjustment training samples corresponding to the actual use site environment of the aircraft. All annotated flight strategy adjustment training samples are composed of a flight strategy adjustment training set. The flight strategy adjustment model is trained through the flight strategy adjustment training set to determine whether the training conditions are met. The training conditions here are generally to reach a certain number of training times. If the training conditions are met, the trained flight strategy adjustment model is output; otherwise, the flight strategy adjustment model is continued to be trained through the flight strategy adjustment training set.

[0018] Training the reward value analysis network includes the following steps: Obtain several bonus value analysis positive training samples and bonus value analysis negative training samples. The bonus value analysis positive training samples include the airspace status spatiotemporal analysis data set, the aircraft attitude data set at the corresponding monitoring time point, and the optimal flight strategy adjustment data set. The bonus value analysis negative training samples include the airspace status spatiotemporal analysis data set, the aircraft attitude data set at the corresponding monitoring time point, and the worst flight strategy adjustment data set. The worst flight strategy adjustment data set is obtained from an aircraft that has a safety accident in an aircraft simulation flight test. The bonus value analysis positive training samples and the bonus value analysis negative training samples are labeled with the bonus value data. Generally, the reward value data corresponding to the positive training sample of reward value analysis is 1, and the reward value data corresponding to the negative training sample of reward value analysis is 0. All labeled positive training samples and negative training samples of reward value analysis form a reward value analysis training set. The reward value analysis network is trained through the reward value analysis training set. The reward value data is used as the target output, and the loss value is calculated to determine whether the loss value is within the preset range. The preset range is set by the developer based on experience. If the loss value is within the preset range, the trained reward value analysis network is output; otherwise, the reward value analysis network continues to be trained through the reward value analysis training set.

[0019] Embodiment 2, a flight data processing system, such as Figure 1 As shown, including: The data acquisition module is used to obtain the position data and flight data set of the aircraft at the current monitoring time point. The flight data set includes the airspace status data set and the aircraft attitude data set. The airspace status data set here refers to the environmental data of the aircraft's location at the current monitoring time point, such as wind speed, visibility, and turbulence intensity. These environmental data will affect the aircraft's flight mission. The aircraft attitude data set includes the aircraft's own attitude data, such as heading angle and flight speed. The flight strategy adjustment module is used to form an airspace state spatiotemporal analysis data set at the monitoring time point by combining the airspace state position annotation data set at the current monitoring time point and the airspace state position annotation data set at the previous N-1 monitoring time points, and then send the airspace state spatiotemporal analysis data set and the aircraft attitude data set at the current monitoring time point to the flight strategy adjustment model for processing to output a flight strategy adjustment data set. The flight strategy adjustment data set corresponds to the aircraft attitude data set, such as the heading adjustment strategy and the flight speed adjustment strategy. The aircraft is controlled based on the aircraft attitude data set at the current monitoring time point and the flight strategy adjustment data set output by the flight strategy adjustment model to achieve adaptive flight state adjustment of the aircraft to the environment. The flight strategy adjustment model includes an airspace state prediction layer, an uncertainty analysis layer, an uncertainty spatiotemporal propagation layer and a flight strategy adjustment layer, wherein the airspace state prediction layer includes a Transformer unit and a random pruned fully connected unit, wherein the Transformer unit is established based on the Transformer model, and is used to perform time series analysis on the airspace state spatiotemporal analysis data set to output an airspace state spatiotemporal analysis vector. Since the airspace state position annotation data set in the airspace state spatiotemporal analysis data set includes position data, the spatial changes can also be analyzed during the time series analysis. The random pruned fully connected unit is used to perform a random pruning fully connected operation on the airspace state spatiotemporal analysis vector output by the Transformer unit. The random pruning fully connected operation here refers to randomly discarding some neuron nodes based on the discard probability during the full connection operation, that is, setting the output of the discarded neuron nodes to zero to output the airspace state prediction vector; the uncertainty analysis layer is used to perform uncertainty analysis on the airspace state prediction vectors output by several airspace state prediction layers to construct an uncertainty vector and a reference airspace state prediction vector. Since the aircraft is performing a flight mission, the aircraft Both the spatial dimension and the temporal dimension will change, and conventional time series analysis cannot accurately predict. Therefore, the uncertainty of airspace state prediction is introduced, and the aircraft strategy is more reasonably controlled based on the uncertainty of airspace state prediction to ensure the safety of the aircraft during flight. If the prediction uncertainty is high, the aircraft should output a more conservative flight strategy, such as a safer flight route and a lower flight speed; the uncertainty spatiotemporal propagation layer is used to perform time series analysis on the uncertainty vector corresponding to the current monitoring time point and the uncertainty vector corresponding to the previous N-1 monitoring time points to output the uncertainty spatiotemporal propagation vector. It should be noted that the time series analysis of the uncertainty vectors of N monitoring time points can analyze the characteristics of the uncertainty propagation of airspace state prediction in time and space. For example, the uncertain change of wind speed will affect the subsequent wind speed prediction with the time dimension, and affect the surrounding wind speed prediction with the change of spatial dimension, thereby more accurately guiding the subsequent aircraft strategy adjustment; the flight strategy adjustment layer is used to splice the aircraft attitude data set, the reference airspace state prediction vector, the uncertainty vector and the uncertainty spatiotemporal propagation vector at the current monitoring time point and send them to the multi-layer perceptron for processing to output the flight strategy adjustment data set; The flight strategy adjustment model update module is used to update the flight strategy adjustment model in real time by means of reinforcement learning during the use of the flight strategy adjustment model.

[0020] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A flight data processing method, characterized in that: include: Acquire the position data and flight data set of the aircraft at the current monitoring time point, the flight data set includes an airspace status data set and an aircraft attitude data set, the position data and the airspace status data set at the current monitoring time point are combined into an airspace status position annotation data set, and the airspace status position annotation data set at the current monitoring time point and the airspace status position annotation data sets at the previous N-1 monitoring time points are combined into an airspace status spatiotemporal analysis data set at the monitoring time point, and then the airspace status spatiotemporal analysis data set and the aircraft attitude data set at the current monitoring time point are sent to the flight strategy adjustment model for processing to output a flight strategy adjustment data set; The flight strategy adjustment model includes an airspace state prediction layer, an uncertainty analysis layer, an uncertainty spatiotemporal propagation layer and a flight strategy adjustment layer. The airspace state prediction layer includes a Transformer unit and a random pruning fully connected unit. The Transformer unit is used to perform time series analysis on the airspace state spatiotemporal analysis data set to output an airspace state spatiotemporal analysis vector. The random pruning fully connected unit is used to perform a random pruning fully connected operation on the airspace state spatiotemporal analysis vector output by the Transformer unit to output an airspace state prediction vector. The uncertainty analysis layer is used to perform a random pruning fully connected operation on the airspace state prediction vector output by several airspace state prediction layers. Uncertainty analysis is performed to construct an uncertainty vector and a reference airspace state prediction vector; the uncertainty spatiotemporal propagation layer is used to perform time series analysis on the uncertainty vector corresponding to the current monitoring time point and the uncertainty vectors corresponding to the previous N-1 monitoring time points to output the uncertainty spatiotemporal propagation vector; the flight strategy adjustment layer is used to splice the aircraft attitude data set, the reference airspace state prediction vector, the uncertainty vector and the uncertainty spatiotemporal propagation vector at the current monitoring time point and send them to the multi-layer perceptron for processing to output the flight strategy adjustment data set; during the use of the flight strategy adjustment model, the flight strategy adjustment model is updated in real time by reinforcement learning.

2. A flight data processing method according to claim 1, characterized in that: The random pruning fully connected operation includes the following steps: In the process of executing the full connection operation, whenever any neuron node outputs, a random number δ between (0, 1) is generated through a random function, and then the random number δ is compared with the discard probability P. If the random number δ is greater than the discard probability P, the output of the neuron node is set to zero. If the random number δ is not greater than the discard probability P, the zeroing operation is not performed.

3. A flight data processing method according to claim 2, characterized in that: The uncertainty analysis layer performs uncertainty analysis on the airspace state prediction vectors output by several airspace state prediction layers to construct uncertainty vectors and reference airspace state prediction vectors, which specifically includes the following steps: The operation of sending the spatial state spatiotemporal analysis data set to the spatial state prediction layer for processing is performed several times to obtain several spatial state prediction vectors; Perform a weighted average operation on all airspace state prediction vectors to construct a baseline airspace state prediction vector; then calculate the uncertainty vector based on all airspace state prediction vectors. The uncertainty vector stores the uncertainty value Fi, i=1, 2, 3, ..., I, I is the total number of uncertainty values, which is consistent with the total number of data in the airspace state data set. The uncertainty value Fi is the variance of the i-th data in all airspace state prediction vectors, which is used to characterize the uncertainty of the prediction of the i-th data in the airspace state data set.

4. A flight data processing method according to claim 3, characterized in that: During the use of the flight strategy adjustment model, the flight strategy adjustment model is updated in real time through reinforcement learning, which specifically includes the following steps: At the monitoring time point, the airspace status spatiotemporal analysis dataset and the aircraft attitude dataset at the current monitoring time point are sent to the target flight strategy adjustment model for processing, and the target flight strategy adjustment dataset is output. In the initial state, the flight strategy adjustment model is consistent with the target flight strategy adjustment model. Then, the airspace status spatiotemporal analysis dataset, the aircraft attitude dataset at the current monitoring time point and the target flight strategy adjustment dataset are spliced ​​to construct reward value analysis data. The reward value analysis data is sent to the reward value analysis network for processing, and the reward value data is output. The gradient value of the reward value data for the target flight strategy adjustment dataset is calculated, and then the parameters of the flight strategy adjustment model are adjusted based on the gradient value using the gradient ascent method to realize real-time update of the flight strategy adjustment model; and the target flight strategy adjustment model is also updated in real time.

5. A flight data processing method according to claim 4, characterized in that: The target flight strategy adjustment model is updated in real time, including the following steps: Whenever the flight strategy adjustment model is updated, the target flight strategy adjustment model is updated in real time using the following formula: θ k (new) = δη k +θ k (bef), where k = 1, 2, 3, ..., K, K is the total number of parameter values ​​in the target flight strategy adjustment model, θ k (new) is the kth parameter value in the updated target flight strategy adjustment model, δ is the update rate, η k Adjust the kth parameter value in the model for the updated flight strategy, θ k (bef) The kth parameter value in the target flight strategy adjustment model before update.

6. A flight data processing method according to claim 5, characterized in that: Training the flight strategy adjustment model includes the following steps: A number of flight strategy adjustment training samples are obtained, including a data set of airspace status spatiotemporal analysis and a data set of aircraft attitude at corresponding monitoring time points. The flight strategy adjustment training samples are labeled by the flight strategy adjustment data set, and all labeled flight strategy adjustment training samples are combined into a flight strategy adjustment training set. The flight strategy adjustment model is trained by the flight strategy adjustment training set to determine whether the training conditions are met. If the training conditions are met, the trained flight strategy adjustment model is output; otherwise, the flight strategy adjustment model is continued to be trained by the flight strategy adjustment training set.

7. A flight data processing method according to claim 6, characterized in that: Training the reward value analysis network includes the following steps: Obtain several bonus value analysis positive training samples and bonus value analysis negative training samples, the bonus value analysis positive training samples include airspace state spatiotemporal analysis data set, aircraft attitude data set corresponding to monitoring time points and optimal flight strategy adjustment data set, the bonus value analysis negative training samples include airspace state spatiotemporal analysis data set, aircraft attitude data set corresponding to monitoring time points and the worst flight strategy adjustment data set, annotate the bonus value analysis positive training samples and bonus value analysis negative training samples through bonus value data, form a bonus value analysis training set with all annotated bonus value analysis positive training samples and bonus value analysis negative training samples, train the bonus value analysis network through the bonus value analysis training set, take the bonus value data as the target output, calculate the loss value, and judge whether the loss value is within a preset range. If the loss value is within the preset range, output the trained bonus value analysis network; otherwise, continue to train the bonus value analysis network through the bonus value analysis training set.

8. A flight data processing system, characterized in that: The system applies a flight data processing method according to any one of claims 1 to 7, including: The data acquisition module is used to obtain the position data and flight data set of the aircraft at the current monitoring time point. The flight data set includes the airspace status data set and the aircraft attitude data set; The flight strategy adjustment module is used to combine the position data of the current monitoring time point and the airspace status data set into an airspace status position annotation data set, and combine the airspace status position annotation data set of the current monitoring time point and the airspace status position annotation data sets of the previous N-1 monitoring time points into an airspace status spatiotemporal analysis data set of the monitoring time point, and then send the airspace status spatiotemporal analysis data set and the aircraft attitude data set of the current monitoring time point into the flight strategy adjustment model for processing to output the flight strategy adjustment data set; The flight strategy adjustment model includes an airspace state prediction layer, an uncertainty analysis layer, an uncertainty spatiotemporal propagation layer and a flight strategy adjustment layer, wherein the airspace state prediction layer includes a Transformer unit and a random pruning fully connected unit, wherein the Transformer unit is used to perform a time series analysis on the airspace state spatiotemporal analysis data set to output an airspace state spatiotemporal analysis vector, and the random pruning fully connected unit is used to perform a random pruning fully connected operation on the airspace state spatiotemporal analysis vector output by the Transformer unit to output an airspace state prediction vector; the uncertainty analysis layer is used to perform uncertainty analysis on the airspace state prediction vectors output by several airspace state prediction layers to construct an uncertainty vector and a reference airspace state prediction vector; the uncertainty spatiotemporal propagation layer is used to perform a time series analysis on the uncertainty vector corresponding to the current monitoring time point and the uncertainty vector corresponding to the previous N-1 monitoring time points to output an uncertainty spatiotemporal propagation vector; the flight strategy adjustment layer is used to splice the aircraft attitude data set at the current monitoring time point, the reference airspace state prediction vector, the uncertainty vector and the uncertainty spatiotemporal propagation vector and send them to a multi-layer perceptron for processing to output a flight strategy adjustment data set; The flight strategy adjustment model update module is used to update the flight strategy adjustment model in real time by means of reinforcement learning during the use of the flight strategy adjustment model.

Images