CN-121999649-A - Unmanned aerial vehicle cluster airspace intelligent management method and system

CN121999649ACN 121999649 ACN121999649 ACN 121999649ACN-121999649-A

Abstract

The invention relates to the technical field of unmanned aerial vehicle cluster airspace management, and provides an unmanned aerial vehicle cluster airspace intelligent management method and system. The method comprises the steps of monitoring a communication link state between a target unmanned aerial vehicle and a central system, determining a communication quality parameter of the target unmanned aerial vehicle, determining an instruction execution uncertainty degree of the target unmanned aerial vehicle according to the communication quality parameter, predicting a track range of the target unmanned aerial vehicle in a future period according to the instruction execution uncertainty degree, responding to a flight task of the unmanned aerial vehicle with high priority, evaluating the overlapping possibility of a flight path of the unmanned aerial vehicle with high priority and the track range of the target unmanned aerial vehicle in the future period, and sending an avoidance instruction to the target unmanned aerial vehicle according to the instruction execution uncertainty degree if the overlapping possibility exceeds a preset threshold. Therefore, the track range of the unmanned aerial vehicle under the communication uncertainty can be predicted more accurately, and the safety, reliability and efficiency of the urban low-altitude airspace unmanned aerial vehicle cluster management are obviously improved.

Inventors

SUN YING

Assignees

中世智航(北京)科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (10)

1. An intelligent management method for an unmanned aerial vehicle cluster airspace is used for managing a central system of the unmanned aerial vehicle cluster in a low-altitude airspace of a city, and is characterized by comprising the following steps: the method comprises the steps of monitoring a communication link state between a target unmanned aerial vehicle and a central system, and determining a communication quality parameter of the target unmanned aerial vehicle, determining the degree of instruction execution uncertainty of the target unmanned aerial vehicle according to the communication quality parameter, wherein the communication quality parameter comprises received signal strength, a data packet loss rate and transmission delay; Predicting a track range of the target unmanned aerial vehicle in a future period according to the instruction execution uncertainty degree; In response to the presence of a flight mission of a high priority unmanned aerial vehicle, assessing a likelihood of overlap of a flight path of the high priority unmanned aerial vehicle with a trajectory range of the target unmanned aerial vehicle within a future period of time; And if the overlapping possibility exceeds a preset threshold, sending an avoidance instruction to the target unmanned aerial vehicle according to the instruction execution uncertainty degree.
2. The unmanned aerial vehicle cluster airspace intelligent management method of claim 1, wherein predicting the track range of the target unmanned aerial vehicle in the future period according to the instruction execution uncertainty degree comprises: Carrying out track prediction on the target unmanned aerial vehicle by adopting a state estimation filter, and outputting a prediction state and a covariance matrix; increasing a process noise covariance of the state estimation filter according to the instruction execution uncertainty level such that the covariance matrix reflects a greater position uncertainty; and determining a spatial confidence region of the target unmanned aerial vehicle under a preset confidence level based on the prediction state and the covariance matrix, and taking the spatial confidence region as the track range.
3. The unmanned aerial vehicle cluster airspace intelligent management method of claim 2, wherein evaluating the likelihood of overlap of the flight path of the high-priority unmanned aerial vehicle with the trajectory range of the target unmanned aerial vehicle over the future period of time comprises: selecting reference positions of a plurality of time points along the flight path of the high-priority task unmanned aerial vehicle; for each reference position, calculating the mahalanobis distance or the space inclusion relation between the reference position and the track range at the same moment; based on the mahalanobis distance or spatial inclusion relationship, a cumulative probability of the flight path coming into contact with the trajectory range over a predicted time window is counted as the overlap probability.
4. The unmanned aerial vehicle cluster airspace intelligent management method of claim 1, wherein the method further comprises: Responding to the target unmanned aerial vehicle receiving a flight control instruction, generating an instruction execution plan, and transmitting abstract information of the instruction execution plan back to the central system; comparing the summary information with a standard execution plan generated based on the flight control instruction to determine a systematic deviation of the target unmanned aerial vehicle; and predicting the future flight trajectory of the target unmanned aerial vehicle based on the systematic deviation to generate a flight range as the trajectory range of the target unmanned aerial vehicle in a future period.
5. The unmanned aerial vehicle cluster airspace intelligent management method of claim 4, wherein comparing the summary information with a standard execution plan generated based on the flight control instructions to determine the systematic deviation of the target unmanned aerial vehicle comprises: Establishing a one-to-one mapping relation between each discrete expected state point contained in the abstract information and the expected state point with the same or corresponding timestamp in the standard execution plan; extracting and comparing parameter values of both in the same state dimension for each pair of mapped expected state points; And calculating parameter deviation between the expected state parameter value in the abstract information and the expected state parameter value in the standard execution plan according to the comparison result, wherein the state dimension comprises at least one of a three-dimensional position, a three-dimensional speed and a three-dimensional posture.
6. The unmanned aerial vehicle cluster airspace intelligent management method of claim 5, wherein comparing the summary information with a standard execution plan generated based on the flight control instructions to determine a systematic deviation of the target unmanned aerial vehicle, further comprises: Acquiring parameter deviations at a plurality of continuous time points, and respectively forming deviation sequences according to state dimensions; Carrying out statistical analysis on the deviation sequence of each state dimension, and judging that systematic deviation exists in the state dimension if the numerical distribution center of the deviation sequence continuously deviates from zero and the change of the deviation sequence with time shows directionality; Based on the parameter bias at the successive time points, a bias vector for quantifying the systematic bias is calculated.
7. The method of claim 6, wherein predicting future flight trajectories of the target unmanned aerial vehicle based on the systematic deviation to generate a flight range comprises: Introducing the deviation vector as a correction term into a state transition model of a state estimation algorithm for track prediction; predicting the future state of the target unmanned aerial vehicle and the probability distribution corresponding to the future state by using the state transition model after the correction term is introduced; And determining a flight area of the target unmanned aerial vehicle at each time position in the future according to the covariance matrix, thereby forming the flight range.
8. The method of intelligent management of unmanned aerial vehicle clusters airspace of claim 7, wherein prior to sending the avoidance command to the target unmanned aerial vehicle, the method further comprises: According to the direction and the amplitude of the deviation vector, determining a reverse compensation quantity superimposed on the basic parameter of the avoidance command to generate a compensated avoidance command, wherein the direction of the compensation quantity is opposite to the direction of the systematic deviation; The sending the avoidance command to the target unmanned aerial vehicle comprises the step of sending the compensated avoidance command to the target unmanned aerial vehicle.
9. The unmanned aerial vehicle cluster airspace intelligent management method of claim 1, wherein sending an avoidance instruction to the target unmanned aerial vehicle according to the degree of instruction execution uncertainty comprises: Determining the safe buffer distance of the target unmanned aerial vehicle according to the instruction execution uncertainty degree, wherein the higher the instruction execution uncertainty degree is, the larger the safe buffer distance is; Generating an avoidance path and/or a speed adjustment instruction based on the safe buffer distance and the flight path of the flight task of the high-priority unmanned aerial vehicle; and sending the avoidance path and/or the speed adjustment instruction to the target unmanned aerial vehicle through a high-priority communication channel.
10. An intelligent management system for an unmanned aerial vehicle cluster airspace, which is used for managing a central system of an unmanned aerial vehicle cluster in a low-altitude airspace of a city, and is characterized in that the intelligent management system comprises: The system comprises a monitoring and determining module, a command execution uncertainty degree determining module and a control module, wherein the monitoring and determining module is used for monitoring the state of a communication link between a target unmanned aerial vehicle and a central system and determining the communication quality parameter of the target unmanned aerial vehicle; the prediction module is used for predicting the track range of the target unmanned aerial vehicle in a future period according to the instruction execution uncertainty degree; An overlap assessment module for assessing a likelihood of overlap of a flight path of a high priority unmanned aerial vehicle with a trajectory range of the target unmanned aerial vehicle within a future period in response to a flight mission of the high priority unmanned aerial vehicle being present; And the avoidance control module is used for sending an avoidance instruction to the target unmanned aerial vehicle according to the instruction execution uncertainty degree if the overlapping possibility exceeds a preset threshold value.

Description

Unmanned aerial vehicle cluster airspace intelligent management method and system Technical Field The application relates to the technical field of unmanned aerial vehicle cluster airspace management, in particular to an unmanned aerial vehicle cluster airspace intelligent management method and system. Background In urban low-altitude areas, the popularity of unmanned aerial vehicle clusters makes airspace management increasingly complex. Existing systems allocate unmanned aerial vehicle paths, avoid collisions, and respond to bursty tasks by dynamically partitioning "flight corridors". However, high-rise dense areas and electromagnetic interference in urban environments often seriously affect the stability of the communication link between the management system and the drone. The link is the basis for the system to master the real-time state of the unmanned aerial vehicle and issue control instructions, and once the communication quality is reduced, the capability of the system for predicting the movement of the unmanned aerial vehicle and preventing conflict is obviously weakened, and particularly, the link is prominent when high-priority emergency tasks are executed. For example, when the system coordinates hundreds of unmanned aerial vehicles to perform daily operations, three-dimensional corridor can be dynamically planned according to airspace flow, weather and task priority, and status is continuously monitored and conflict is predicted. Once the highest priority tasks such as medical material transport occur, the system immediately re-plans the path and instructs nearby drones to avoid. However, if a certain logistics unmanned aerial vehicle is in a signal unstable area, the received avoidance command may be incomplete due to the loss of the data packet, so that the deviation between the actual track and the expected system occurs. Meanwhile, the system still plans a subsequent path for the medical unmanned aerial vehicle based on the original prediction, and determines that the two remain a safe interval. Over time, the track deviation gradually accumulates until the two machines approach, and the proximity sensor triggers an emergency alarm, so that the unmanned aerial vehicle is forced to take a severe evasive action with high overload. The situation not only brings collision risk, but also highlights the defect of the real-time state sensing and prediction capability of the system to the unmanned aerial vehicle in a complex communication environment. In view of the above, there is a need in the art for improvements. Disclosure of Invention The application discloses an intelligent management method and system for an unmanned aerial vehicle cluster airspace, and aims to solve the technical problem that potential conflict risks exist when a high-priority task is executed due to insufficient sensing and predicting capabilities of an unmanned aerial vehicle real state in a complex communication environment of an unmanned aerial vehicle cluster management system in an urban low-altitude airspace. The technical scheme of the application is as follows: In a first aspect, the application discloses an intelligent management method for an unmanned aerial vehicle cluster airspace, which is used for managing a central system of an unmanned aerial vehicle cluster in a low-altitude airspace of a city, and comprises the following steps: The method comprises the steps of monitoring a communication link state between a target unmanned aerial vehicle and a central system, determining a communication quality parameter of the target unmanned aerial vehicle, determining the degree of instruction execution uncertainty of the target unmanned aerial vehicle according to the communication quality parameter, wherein the communication quality parameter comprises received signal strength, data packet loss rate and transmission delay; Predicting the track range of the target unmanned aerial vehicle in a future period according to the instruction execution uncertainty degree; In response to the flight mission of the high priority unmanned aerial vehicle being present, evaluating a likelihood of overlap of a flight path of the high priority unmanned aerial vehicle with a trajectory range of the target unmanned aerial vehicle within a future period of time; and if the overlapping possibility exceeds a preset threshold, sending an avoidance instruction to the target unmanned aerial vehicle according to the instruction execution uncertainty degree. Further, according to the degree of instruction execution uncertainty, predicting the track range of the target unmanned aerial vehicle in a future period, wherein the track range comprises the steps of carrying out track prediction on the target unmanned aerial vehicle by adopting a state estimation filter, outputting a prediction state and a covariance matrix, increasing the process noise covariance of the state estimation filter according to the degree of instruction execution uncer