Low-altitude airspace conflict detection and resolution method and system

Abstract

The application provides a low-altitude airspace conflict detection and resolution method and system, the system comprises the following steps: step 1, receiving a plurality of exclusive airspace use plans in the low-altitude airspace; step 2, data preprocessing is performed on the received airspace use plans; step 3, a multi-dimensional conflict detection is performed on the preprocessed airspace use plans by adopting a hierarchical detection strategy; step 4, based on the adjustable airspace with conflicts in the detection results, an available time axis model is constructed, time scheduling optimization is performed, and a conflict resolution scheme is obtained; step 5, the conflict detection results and the resolution scheme are output, and the low-altitude airspace conflict detection and resolution are completed; the system is used for realizing the method. The application decouples the complex four-dimensional avoidance and reduces the dimension, completely blocks the chain secondary conflict in the multi-concurrent scene in a single scheduling traversal, greatly reduces the calculation cost, and realizes the efficient utilization and anti-deadlock intelligent scheduling of the low-altitude airspace resources.

Description

Technical Field

[0001] This invention relates to a method and system for detecting and resolving airspace conflicts, and particularly to a method and system for detecting and resolving low-altitude airspace conflicts. Background Technology

[0002] With the rapid development of the low-altitude economy, low-altitude application scenarios such as drone logistics, urban air traffic, and emergency rescue are increasing, leading to an explosive growth in airspace usage demand. Traditional airspace management methods, primarily designed for high-altitude civil aviation flights, are ill-suited to the characteristics of low-altitude aircraft, which are numerous, dense, and highly flexible.

[0003] In existing technologies, low-altitude airspace conflict detection mainly suffers from the following problems:

[0004] 1. Detection often uses pairwise comparison methods, which have high computational complexity and cannot cope with large-scale airspace plans;

[0005] 2. Detection often only considers the spatial or temporal dimensions, lacking comprehensive conflict detection across multiple dimensions such as space, height, and time;

[0006] 3. There is a lack of effective automatic resolution methods after a conflict is detected, and it mainly relies on manual adjustment, which is inefficient and prone to errors. Summary of the Invention

[0007] Purpose of the invention: The technical problem to be solved by the present invention is to provide a method and system for detecting and resolving low-altitude airspace conflicts, in order to address the shortcomings of the existing technology.

[0008] To address the aforementioned technical problems, this invention discloses a method and system for detecting and resolving low-altitude airspace conflicts, wherein the method includes the following steps:

[0009] Step 1: Receive multiple exclusive airspace usage plans in the low-altitude airspace;

[0010] Step 2: Preprocess the received airspace usage plan data;

[0011] Step 3: Use a layered detection strategy to perform multi-dimensional conflict detection on the preprocessed airspace usage plan;

[0012] Step 4: Based on the adjustable spatial domains with conflicts in the detection results, construct an available time axis model, optimize the time scheduling, and obtain a conflict resolution scheme.

[0013] Step 5: Output the conflict detection results and resolution plan to complete the low-altitude airspace conflict detection and resolution.

[0014] Furthermore, the exclusive airspace usage plan described in step 1 includes:

[0015] Spatial range information, altitude range information, time range information, and airspace type information; wherein, the airspace type includes fixed airspace and adjustable airspace; the spatial range information includes different geometric shapes.

[0016] Furthermore, the data preprocessing described in step 2 includes:

[0017] Step 2-1: Convert the spatial extent information in the airspace usage plan into a standardized polygon representation, as follows:

[0018] For linear spatial regions, buffer polygons are generated based on a preset width; for circular spatial regions, a polygon approximation algorithm is used to convert them into standardized polygons; for existing polygons, their vertex sequences are closed to form valid geometric objects.

[0019] Step 2-2: Parse the time range information and height range information into a unified numerical format; for the height range information, introduce a boundary relaxation operator; when the original height range information is an open interval, introduce a floating-point offset; and convert all heterogeneous height boundaries into standard closed intervals.

[0020] Step 2-3: Following the preprocessing in Step 2-1 and Step 2-2, each airspace usage plan includes a labeled polygonal airspace, height range, and time range.

[0021] Furthermore, the layered detection strategy described in step 3 includes:

[0022] Spatial dimension conflict detection, height dimension conflict detection, and temporal dimension conflict detection are performed sequentially; among them...

[0023] Spatial detection involves constructing a spatial index for all polygonal spatial domains, retrieving candidate spatial domain pairs where bounding boxes intersect, performing polygonal geometry calculations on the candidate spatial domain pairs, confirming planar spatial conflicts, and performing height detection if a conflict is found.

[0024] Height detection involves determining the height range for confirmed planar spatial conflict pairs. Let the height ranges of these pairs be represented as intervals (h1_min, h1_max) and (h2_min, h2_max), where h1_min and h2_min are the minimum values ​​of the two height ranges, and h1_max and h2_max are the maximum values ​​of the two height ranges. If the following conditions are met:

[0025] max(h1_min, h2_min) < min(h1_max, h2_max)

[0026] If it is determined to be a high-level conflict, a time check will be performed, where max() means to take the maximum value and min() means to take the minimum value;

[0027] Time detection involves performing time window overlap analysis on spatial pairs that are determined to conflict in both space and altitude. Let the time ranges of the two spatial pairs be represented as intervals (t1_start, t1_end) and (t2_start, t2_end), where t1_start and t2_start are the starting values ​​of the two time ranges, and t1_end and t2_end are the ending values ​​of the two time ranges. If the following conditions are met:

[0028] t1_start < t2_end and t2_start < t1_end

[0029] If the two pairs are identified as spatiotemporal conflicts, the spatial domain pairs will be stored as spatiotemporal conflict pairs in the spatiotemporal conflict pair set.

[0030] Furthermore, the construction of a usable timeline model described in step 4 specifically includes:

[0031] Spatial constraints are extracted from the adjustable spatial domains in all spatiotemporal conflict pairs, that is, all fixed spatial domains that have spatial or height overlap with the adjustable spatial domain are extracted as potential spatial obstacles.

[0032] Obtain the planned time periods for all potential spatial obstacles and project them onto a preset scheduling time axis. Calculate the complement on the time axis to generate an available time axis model for the adjustable spatial domain.

[0033] Furthermore, the time scheduling optimization described in step 4 involves adopting an optimization strategy based on incremental state updates, with the goal of minimizing the overall time adjustment amount, and performing discretization optimization.

[0034] Furthermore, the aforementioned discretization optimization specifically includes:

[0035] Within the available time axis model of the adjustable spatial domain to be scheduled, a discretized grid search is performed with a preset time step.

[0036] Construct a loss function with the absolute deviation between the candidate start time and the original planned start time as the independent variable;

[0037] The candidate point that minimizes the loss function value is selected as the starting time for the new assignment.

[0038] Furthermore, the optimization strategy based on incremental state updates specifically includes:

[0039] Maintain a global dynamic scheduling state database. Once a new time period is successfully allocated to the current adjustable airspace, immediately synchronize its parameters to the global dynamic scheduling state database.

[0040] When constructing a usable time axis model for subsequent adjustable spatial domains, in addition to extracting fixed spatial domains as constraints, dynamically extracting the already scheduled adjustable spatial domains that have spatial overlap with it from the global dynamic scheduling state library as constraints.

[0041] Furthermore, the time scheduling optimization described in step 4 also includes heuristic scheduling and sorting, specifically including:

[0042] Before performing discretization optimization, spatial topology congestion weights are calculated based on the number of fixed spatial domains with adjustable spatial domain overlap and the number of other adjustable spatial domains.

[0043] Based on the spatial topology congestion weight, the priority scheduling order of adjustable spatial domains is determined by sorting them in descending order.

[0044] This invention also proposes a low-altitude airspace conflict detection and resolution system for implementing the aforementioned method, comprising:

[0045] The system comprises a data receiving and preprocessing module, a multi-dimensional conflict detection module, an intelligent scheduling module, and a result output module; among which,

[0046] The data receiving and preprocessing module is used to receive and standardize airspace usage plans.

[0047] The multi-dimensional conflict detection module employs a hierarchical detection strategy to detect conflicts in the spatial, height, and temporal dimensions.

[0048] The intelligent scheduling module, based on the conflict detection results, constructs a personalized available time axis model for the adjustable airspace with conflicts, and adopts an optimization algorithm based on incremental state updates to optimize time scheduling to resolve conflicts with the goal of minimizing the overall time adjustment.

[0049] The results output module is used to output the detection and resolution results.

[0050] Beneficial effects:

[0051] 1. This invention optimizes the detection complexity from O(n²) to close to O(n log n) through spatial indexing technology and hierarchical filtering strategy, which can handle the real-time detection of thousands of spatial plans, thus greatly improving the detection efficiency.

[0052] 2. This invention can achieve accurate identification of multi-dimensional conflicts by judging conflicts in three dimensions: space, height, and time, thus avoiding misjudgment and omission by single-dimensional detection.

[0053] 3. This invention uses an automatic scheduling and resolution method based on optimization algorithms, which reduces manual intervention and improves resolution efficiency and accuracy.

[0054] 4. This invention achieves efficient utilization of airspace resources and improves overall airspace capacity through the available time axis model and optimization algorithm.

[0055] 5. This invention is applicable to various low-altitude application scenarios such as urban air traffic, drone logistics, and emergency rescue, and has good promotional value. Attached Figure Description

[0056] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the present invention in the above and / or other aspects will become clearer.

[0057] Figure 1 This is a schematic diagram of the overall process of the low-altitude airspace conflict detection and resolution method in this invention.

[0058] Figure 2 This is a schematic diagram of the available timeline model and intelligent scheduling process. Detailed Implementation

[0059] The core idea of ​​this invention lies in a technical approach that combines a hierarchical conflict detection mechanism with an intelligent scheduling optimization algorithm. First, a hierarchical conflict detection mechanism is employed, prioritizing spatial dimensions and utilizing spatial indexing technology to quickly filter potentially overlapping spatial domain pairs, significantly reducing computational load. Then, height range intersection detection is performed on the overlapping spatial domains to eliminate pseudo-conflicts where spatial overlap is not at the height level. Finally, time window overlap analysis confirms the true spatiotemporal conflict. Based on this, intelligent scheduling optimization is used. An available time window model for adjustable spatial domains is constructed based on fixed spatial domain time occupancy. With the objectives of minimizing time adjustment and maximizing resource utilization, an optimization algorithm is used to solve for the optimal scheduling scheme. Simultaneously, a secondary conflict verification is performed on the scheduling results to ensure the effectiveness and reliability of the conflict resolution scheme.

[0060] The first part of this invention provides a method for detecting and resolving low-altitude airspace conflicts. Through a mechanism combining "layered conflict detection" and "intelligent scheduling optimization," it efficiently and accurately identifies and automatically resolves conflicts in airspace usage plans across spatial, altitude, and temporal dimensions. The overall process of the method is as follows: Figure 1 As shown, the specific steps include:

[0061] 1. Data preparation.

[0062] Receive multiple airspace usage plans, each plan containing its spatial extent (such as polygonal, circular, or linear paths), altitude extent (represented as open or closed intervals), planned usage time window, and airspace type (fixed or adjustable).

[0063] First, the input data undergoes standardized preprocessing: for linear spatial domains, buffer polygons are generated based on a preset width (e.g., a safety interval) (a common GIS operation; refer to relevant geographic information system literature); for circular spatial domains, a polygon approximation algorithm (e.g., regular N-gon approximation) is used to convert them into standardized polygons; for existing polygons, their vertex sequences are ensured to be closed at both ends to form valid geometric objects. Simultaneously, time and height information are parsed into a unified numerical format, constructing a complete spatial domain object for each plan, containing spatial polygons, height slots, and time slots. For height information, the system introduces a boundary relaxation operator: when the original declaration is an open interval (e.g., strictly less than a certain height), the system introduces a small floating-point offset (e.g., ±0.0001 meters) at the calculation level, uniformly converting all heterogeneous height boundaries into standard closed intervals, thereby achieving unambiguous and extremely fast comparison in subsequent intersection detection.

[0064] 2. Layered conflict detection.

[0065] This step employs a layered strategy that filters sequentially across three dimensions: space, altitude, and time. This aims to quickly eliminate a large number of obviously non-conflicting airspace pairs and focus on truly potential conflicts, thereby significantly improving detection efficiency.

[0066] 2.1 Rapid Spatial Dimension Conflict Filtering: Construct a spatial index (e.g., R-tree) for the polygons of all spatial objects. This index can efficiently retrieve pairs of spatial regions where bounding boxes intersect, with a time complexity optimized to O(n log n), thus avoiding O(n²) global pairwise comparisons. For candidate spatial region pairs retrieved by the index, perform precise geometric relationship calculations (e.g., determine whether polygons overlap or contain each other) to confirm spatial conflicts. When determining conflicts in this step, perform the height dimension conflict filtering in step 2.2.

[0067] 2.2 Height Dimension Conflict Filtering: For spatial pairs that pass the spatial conflict detection, further determine whether their height ranges intersect. Height ranges can be represented as intervals (h1_min, h1_max) and (h2_min, h2_max). The mathematical condition for determining whether there is substantial overlap is: max(h1_min, h2_min) < min(h1_max, h2_max). Only when this condition is met is it determined to be a height conflict. If a conflict exists in this step, proceed to 2.3 Time Dimension Conflict Confirmation.

[0068] 2.3 Temporal Dimension Conflict Confirmation: For spatial pairs that conflict in the first two dimensions, a temporal overlap analysis is performed last. Let the two time windows be (t1_start, t1_end) and (t2_start, t2_end). The mathematical condition for temporal overlap between them is: t1_start < t2_end and t2_start < t1_end. Spatial pairs that satisfy this condition will be ultimately recorded as "spatiotemporal conflict pairs" and added to the conflict list.

[0069] 3. Conflict resolution and intelligent scheduling.

[0070] 3.1 In a multi-concurrency scheduling scenario, the system first calculates the "spatial topology congestion weight" of each adjustable spatial domain. This congestion weight is used to determine the order in which the adjustable spatial domains are scheduled.

[0071] The formula for calculating this weight is defined as follows:

[0072]

[0073] in This refers to the number of fixed airspaces overlapping the adjustable airspace. The number of other adjustable airspaces that overlap with it. and For the preset penalty coefficient and (This indicates that fixed airspace is an inviolable hard constraint and has a higher priority.)

[0074] Through graph theory topology analysis, the number of static obstacles (fixed airspace) and dynamic obstacles (other adjustable airspace) around each adjustable airspace is counted. Based on the spatial topology congestion weight, the airspace with the most complex spatial environment and the densest constraints is prioritized for scheduling, thereby improving the success rate of global scheduling at the macro level. When determining the individual scheduling time, the algorithm introduces a "discrete time step and loss function optimization" mechanism. The system performs a sliding search within a one-dimensional available time axis dedicated to that airspace, using a preset time step (e.g., 0.1 hours), and explicitly constructs a loss function with "absolute time adjustment" as the independent variable. By traversing the valid discretized time grid, the global minimum value of the loss function is solved, thus ensuring the optimization quality of a single scheduling operation with an accurate mathematical model. Specifically, the following steps are included.

[0075] 3.2 Construction of available time axis based on spatiotemporal dimensionality reduction mapping: In traditional conflict resolution, a trial-and-error mechanism of first adjusting time and then global recalculation is usually adopted, which has a large computational cost and is very easy to produce "chain-like secondary conflict" (i.e., avoiding one spatial domain but colliding with another spatial domain that was not originally in conflict).

[0076] This invention proposes a dimensionality reduction and decoupling strategy: avoiding blind searching directly in the four-dimensional spatiotemporal domain. Specifically, for any tunable spatial domain with conflicts, the system first utilizes intermediate results from previous R-tree spatial indexing and height detection to retrieve all fixed spatial domains that overlap with it in spatial geometry and height intervals (regardless of whether there is a conflict at the current time). These fixed spatial domains constitute the "potential spatial barriers" of the tunable spatial domain. Subsequently, the system extracts the time-occupancy slots of these potential spatial barriers and projects them onto a one-dimensional 24-hour absolute time axis. If the time slots of multiple fixed spatial domains intersect, the system merges them into a large continuous blocking block. Finally, by subtracting these blocking blocks from the overall schedulable time window (i.e., finding the complement), an accurate "safe and usable time window set," i.e., the usable time axis, is obtained.

[0077] Based on the available time axis, the safe available time of each adjustable airspace with conflict is obtained relative to the fixed airspace. Within these safe available times, the time is readjusted one by one according to the congestion weight using an optimization algorithm, and finally the conflicts of all adjustable airspaces are resolved.

[0078] The advantage of this design is that it transforms the complex spatial avoidance logic into a pure time-time constraint in advance. As long as any time is assigned to the adjustable spatial domain within this safe time set, it is mathematically and physically guaranteed that it will never conflict with any fixed spatial domain. This significantly compresses the search space of subsequent optimization algorithms and completely eliminates the risk of secondary conflicts with fixed spatial domains.

[0079] 3.3 Anti-Chain Conflict Scheduling Algorithm Based on Dynamic State Update: In multi-concurrent airspace scheduling scenarios, if only the adjustable airspace is optimized independently, it is very easy to generate "chain secondary conflict" (that is, the adjustment of airspace A leads to a new conflict with airspace B).

[0080] To address this, this invention proposes an optimized scheduling scheme with an incremental constraint locking mechanism. Specifically, the system maintains a global dynamic scheduling state database, which records the occupancy time of currently adjusted adjustable space domains. This ensures that subsequent adjustments to adjustable space domains do not conflict with already adjusted space domains, treating already adjusted adjustable space domains as fixed space domains. Specifically, when allocating time to multiple conflicting adjustable space domains based on priority or a specific order, whenever an adjustable space domain obtains an optimal new time window within the available time axis, the system immediately synchronizes its new spatiotemporal occupancy status to the dynamic scheduling state database.

[0081] When processing subsequent unscheduled spatial domains, the constraints of the optimization algorithm not only include static fixed spatial domains, but also dynamically introduce the latest time parameters of other adjustable spatial domains that have been scheduled and overlap with it in space from the state library. Through this serial processing logic of "optimizing one, locking one, and adding one constraint", the scheduled spatial domains are transformed into dynamic barriers for subsequent scheduling. Thus, concurrent conflicts between adjustable spatial domains can be absolutely avoided in a single traversal of the algorithm, eliminating the need for complex global multi-round iterative backtracking.

[0082] 3.4 Secondary Conflict Verification: For the new airspace plan set generated after scheduling optimization, perform the hierarchical conflict detection in step 2 again to ensure that all conflicts have been successfully resolved and to guarantee the feasibility of the output scheme.

[0083] 4. Output the results.

[0084] Ultimately, the system outputs two key results: first, a detailed conflict detection report listing all identified spatiotemporal conflict pairs; and second, an intelligent resolution plan that clearly indicates the recommended adjusted schedule for each adjustable airspace.

[0085] In the second part, based on the above method, the present invention also proposes a low-altitude airspace conflict management system for implementing the above method, comprising:

[0086] The data receiving and preprocessing module is used to receive and standardize airspace usage plans; the data preprocessing of the received airspace usage plans includes:

[0087] If the spatial domain is linear, a buffer polygon is generated based on the preset line width;

[0088] If the spatial shape is circular, a polygon approximation algorithm is used to generate an approximate circular polygon.

[0089] The multidimensional conflict detection module is configured to use a hierarchical detection strategy to detect conflicts in the spatial, height, and temporal dimensions.

[0090] The intelligent scheduling module is configured to build a personalized available time axis model for the adjustable airspace with conflicts based on the detection results, and to perform time scheduling optimization to resolve conflicts by using an optimization algorithm based on incremental state updates with the goal of minimizing the overall time adjustment.

[0091] The results output module is used to output the detection and resolution results.

[0092] The present invention also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the program to implement the above-described method and system.

[0093] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method and system.

[0094] Example:

[0095] Taking a city-wide drone logistics delivery airspace conflict resolution task as an example, the city's low-altitude flight service center needs to conduct airspace conflict detection and resolution for the delivery tasks of multiple drone logistics companies between 08:00 and 18:00 on October 1, 2024. Figure 2 As shown, the specific implementation process is as follows:

[0096] Step 1: Data Reception and Preprocessing

[0097] The service center received three airspace usage plan applications, the raw data of which are shown in Table 1 below:

[0098] Table 1 Original Data Table

[0099]

[0100] The system performs standardized preprocessing on the above data:

[0101] (1) For airspace A1 (linear): Based on the line width of 0.5 km, take the line segment as the central axis and extend it to both sides by 0.25 km to generate a buffer zone, resulting in a rectangular strip polygon with a width of 0.5 km. The vertex coordinates are [(116.299,39.899), (116.501,40.001), (116.501,40.001), (116.299,39.899)].

[0102] (2) For the spatial domain A2 (polygon): use the coordinates of its four vertices directly and ensure that the first and last vertices are the same to form a closed polygon.

[0103] (3) For the spatial domain A3 (circle): with the center (116.4, 39.95) as the center and a radius of 2 kilometers, a regular 16-sided polygon is used to approximate the polygon with 16 vertex coordinates.

[0104] (4) Time information is uniformly converted to the number of hours starting from 00:00 on the same day: A1 is [8.0, 10.0], A2 is [9.5, 11.0], and A3 is [9.0, 12.0].

[0105] (5) The height range is uniformly represented as an open interval: A1 is (100, 300), A2 is (150, 250), and A3 is (0, 500).

[0106] Step 2: Construct spatial domain objects and spatial indexes

[0107] The system creates a spatial domain object for each preprocessed spatial domain, containing polygon geometry, height range, time slot, and type attributes. Then, an R-tree spatial index is constructed for the polygons in these three spatial domains.

[0108] The bounding box of A1 is (min_x=116.299, min_y=39.899, max_x=116.501, max_y=40.001).

[0109] The bounding box of A2 is (min_x=116.34, min_y=39.91, max_x=116.46, max_y=39.98).

[0110] The bounding box of A3 is (min_x=116.382, min_y=39.932, max_x=116.418, max_y=39.968).

[0111] The R-tree index quickly calculates that the bounding boxes of A1 and A3 intersect, the bounding boxes of A2 and A3 intersect, and the bounding boxes of A1 and A2 do not intersect.

[0112] Step 3: Layered Conflict Detection

[0113] according to Figure 2 The three-layer filtering mechanism shown is used for detection:

[0114] First layer: Spatial conflict detection

[0115] For A1 and A3: Calculate their polygonal geometric relationship and find that the rectangular band of A1 and the approximately circular polygon of A3 have overlapping areas (the overlapping area is about 0.8 square kilometers), which is determined to be a spatial conflict.

[0116] For A2 and A3: Calculate the polygon geometric relationship and find that the polygon of A2 is completely inside the circular polygon of A3, which is determined to be a spatial conflict (inclusion relationship).

[0117] For A1 and A2: Since the R-tree has already determined that the bounding boxes do not intersect, the precise geometric calculation is skipped, and they are directly determined to have no spatial conflict.

[0118] Second layer: High-conflict filtering

[0119] A1 and A3: A1 height (100, 300), A3 height (0, 500). The condition is that max(100, 0) = 100 < min(300, 500) = 300, satisfying the open interval overlap condition, thus indicating a height conflict.

[0120] A2 and A3: A2 height (150, 250), A3 height (0, 500). The condition is met: max(150, 0) = 150 < min(250, 500) = 250. Therefore, the heights are considered to conflict.

[0121] Third layer: Time conflict confirmation

[0122] A1 and A3: A1 time [8.0, 10.0], A3 time [9.0, 12.0]. Since 8.0 < 12.0 and 9.0 < 10.0, the time overlap interval is [9.0, 10.0], indicating a time conflict. Therefore, A1 and A3 have a spatiotemporal conflict.

[0123] A2 and A3: A2 time [9.5, 11.0], A3 time [9.0, 12.0]. Since 9.5 < 12.0 and 9.0 < 11.0, the time overlap is within the range [9.5, 11.0], indicating a time conflict. Therefore, A2 and A3 have a spatiotemporal conflict.

[0124] Two pairs of spatiotemporal conflicts were ultimately detected: (A1, A3) and (A2, A3).

[0125] Step 4: Heuristic sorting and state database initialization based on spatial topological conflict degree

[0126] The system classifies conflict topology based on the inherent attributes of airspace. A3 is marked as an immutable hard constraint, designated "Fixed Airspace," while A1 and A2 are marked as "Adjustable Airspace." The core scheduling engine initializes a dynamic scheduling state database based on the classification results. This database maintains the mapping between each airspace ID and its latest time slot in physical memory using a key-value pair (hash map) data structure, locking the static time slot of Fixed Airspace A3 as a resident spatiotemporal obstacle.

[0127] Before time allocation, the system triggers a heuristic scheduling and sorting mechanism. By retrieving the high-dimensional spatial filtering results from step 3, the system statistically concludes that: adjustable spatial domain A1 is spatially associated with one fixed spatial domain (A3) and one adjustable spatial domain (A2); adjustable spatial domain A2 is also associated with one fixed spatial domain (A3) and one adjustable spatial domain (A1). The system calculates the spatial topology congestion weights for both domains accordingly. In complex, multi-concurrency scenarios with weight differences, the system will force priority scheduling of the spatial domain with the higher congestion weight (i.e., the densest surrounding spatial constraints) to prevent deadlock situations where no time is available in subsequent scheduling. In this embodiment, both domains have the same weight and are processed sequentially according to the system's default queue.

[0128] Step 5: Intelligent Scheduling Based on Spatiotemporal Dimensionality Reduction Mapping and Discretization Loss Function Optimization

[0129] The core scheduling engine employs a combined mechanism of dimensionality reduction mapping, discrete optimization, and incremental locking to resolve conflicts.

[0130] 5.1 Spatiotemporal Dimensionality Reduction and Discretized Grid Optimization for Adjustable Spatial Domain A1

[0131] The system first retrieves the scheduling request for A1 (originally planned for [8.0, 10.0], duration 2.0 hours). The dimensionality reduction mapping module projects the time-occupied interval [9.0, 12.0] of the fixed spatial domain A3, which overlaps with A1's time domain, onto the overall time axis and calculates its complement, thus creating a customized, conflict-free time axis model for A1 after decoupling.

[0132] Available_Axis(A1) = [0.0, 9.0] ∪ [12.0, 24.0]

[0133] The four-dimensional avoidance search was successfully reduced to a one-dimensional safe interval. Subsequently, the optimization algorithm initiated a discretized grid search. The system performs a sliding traversal within the aforementioned available interval set with a preset step size of 0.1 hours, and constructs a target loss function to quantify the cost of time offset:

[0134] Minimize: Loss = |candidate_start - original_start| = |candidate_start - 8.0|

[0135] Within the available interval [0.0, 9.0], the algorithm discretizes the search in 0.1-hour increments, and the latest valid candidate start point (candidate_start) that can accommodate a duration of 2 hours is 7.0. Substituting into the loss function, we get: Loss = |7.0 - 8.0| = 1.0 hour.

[0136] Within the available interval [12.0, 24.0], the earliest valid candidate starting point is 12.0. Substituting this into the loss function, we get:

[0137] Loss = |12.0 - 8.0| = 4.0 hours.

[0138] The optimization algorithm compares the solutions of all discrete grid points and selects the optimal solution with the smallest loss function value (Loss=1.0), and establishes the new time window of A1 as [7.0, 9.0].

[0139] 5.2 Dynamic Incremental Locking and Joint Optimization for Adjustable Spatial Domain A2

[0140] When A1 is assigned a new time [7.0, 9.0], the system triggers an incremental locking mechanism, immediately writing A1's latest scheduling state into the dynamic scheduling state database. In subsequent viewpoints, A1 is effectively promoted to a "dynamic spatiotemporal barrier".

[0141] The algorithm then processes A2 (originally planned [9.5, 11.0], duration 1.5 hours). The dimensionality reduction mapping module not only pulls the static obstacle A3 [9.0, 12.0], but also dynamically pulls the latest blocking segment [7.0, 9.0] of A1 in the state library that has potential spatial overlap with A2. The algorithm merges the two to generate a joint blocking block [7.0, 12.0], and generates a chain-avoidance conflict-safe time axis model for A2 using the complement set operator: Available_Axis(A2) = [0.0, 7.0] ∪ [12.0, 24.0]

[0142] The system continues to use a 0.1-hour step size for discretization optimization, constructing the loss function for A2:

[0143] Minimize: Loss = |candidate_start - 9.5|.

[0144] Within [0.0, 7.0], the latest valid candidate starting point is 5.5 (5.5+1.5=7.0), and the loss is calculated as |5.5 - 9.5| = 4.0 hours.

[0145] Within [12.0, 24.0], the earliest valid candidate starting point is 12.0, and the calculation loss is |12.0 - 9.5| = 2.5 hours.

[0146] The algorithm selects a local minimum (Loss=2.5) to precisely lock the new time window of A2 to [12.0, 13.5]. At this point, the single scheduling traversal based on queuing weight and discrete loss optimization is complete, completely avoiding the occurrence of chain-like secondary collisions.

[0147] Step 6: Output Results

[0148] The system outputs the final scheduling scheme:

[0149] Conflict detection results:

[0150] 1. There is a time and space conflict between A1 (medical emergency access) and A3 (key area protection).

[0151] 2. There is a time and space conflict between A2 (logistics trunk line) and A3 (key area protection).

[0152] Intelligent digestion solution:

[0153] 1. The A1 time has been adjusted from 08:00-10:00 to 07:00-09:00 (1 hour earlier).

[0154] 2. The A2 time has been adjusted from 09:30-11:00 to 12:00-13:30 (delayed by 2.5 hours).

[0155] Verification result: All conflicts have been successfully resolved.

[0156] The following is an extended description of this embodiment:

[0157] For more complex scenarios involving a large amount of spatial domains, the advantages of R-tree indexes become even more apparent. Assuming there are 1000 spatial domains, traditional pairwise comparisons would require approximately 500,000 computations. Using the R-tree index of this invention, bounding box filtering is performed first, potentially generating only about 10,000 candidate pairs. Then, precise geometric calculations, height and time determinations are performed, reducing the total computational load by approximately 98% and enabling detection to be completed within seconds.

[0158] In its specific implementation, this application provides a computer storage medium and a corresponding data processing unit. The computer storage medium is capable of storing a computer program, which, when executed by the data processing unit, can run the invention's content regarding a low-altitude airspace conflict detection and resolution method and system, as well as some or all of the steps in various embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0159] Those skilled in the art will clearly understand that the technical solutions in the embodiments of the present invention can be implemented using computer programs and their corresponding general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of computer programs, i.e., software products. These computer program software products can be stored in a storage medium and include several instructions to cause a device containing a data processing unit (which may be a personal computer, server, microcontroller, MCU, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0160] This invention provides a method and system for detecting and resolving conflicts in low-altitude airspace. Many methods and approaches exist for implementing this technical solution; the above description is merely a preferred embodiment. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A method for detecting and resolving conflicts in low-altitude airspace, characterized in that, Includes the following steps: Step 1: Receive multiple exclusive airspace usage plans in the low-altitude airspace; Step 2: Preprocess the received airspace usage plan data; Step 3: Use a layered detection strategy to perform multi-dimensional conflict detection on the preprocessed airspace usage plan; Step 4: Based on the adjustable spatial domains with conflicts in the detection results, construct an available time axis model, optimize the time scheduling, and obtain a conflict resolution scheme. Step 5: Output the conflict detection results and resolution plan to complete the low-altitude airspace conflict detection and resolution.

2. The method for low-altitude airspace conflict detection and resolution according to claim 1, characterized in that, The exclusive airspace usage plan described in step 1 includes: Spatial range information, altitude range information, time range information, and airspace type information; wherein, the airspace type includes fixed airspace and adjustable airspace; the spatial range information includes different geometric shapes.

3. The method for low-altitude airspace conflict detection and resolution according to claim 2, characterized in that, The data preprocessing described in step 2 includes: Step 2-1: Convert the spatial extent information in the airspace usage plan into a standardized polygon representation, as follows: For linear spatial regions, buffer polygons are generated based on a preset width; for circular spatial regions, a polygon approximation algorithm is used to convert them into standardized polygons; for existing polygons, their vertex sequences are closed to form valid geometric objects. Step 2-2: Parse the time range information and height range information into a unified numerical format; for the height range information, introduce a boundary relaxation operator; when the original height range information is an open interval, introduce a floating-point offset; and convert all heterogeneous height boundaries into standard closed intervals. Step 2-3: Following the preprocessing in Step 2-1 and Step 2-2, each airspace usage plan includes a labeled polygonal airspace, height range, and time range.

4. The method for low-altitude airspace conflict detection and resolution according to claim 3, characterized in that, The hierarchical detection strategy described in step 3 includes: Spatial dimension conflict detection, height dimension conflict detection, and temporal dimension conflict detection are performed sequentially; among them... Spatial detection involves constructing a spatial index for all polygonal spatial domains, retrieving candidate spatial domain pairs where bounding boxes intersect, performing polygonal geometry calculations on the candidate spatial domain pairs, confirming planar spatial conflicts, and performing height detection if a conflict is found. Height detection involves determining the height range for confirmed planar spatial conflict pairs. Let the height ranges of these pairs be represented as intervals (h1_min, h1_max) and (h2_min, h2_max), where h1_min and h2_min are the minimum values ​​of the two height ranges, and h1_max and h2_max are the maximum values ​​of the two height ranges. If the following conditions are met: max(h1_min, h2_min) < min(h1_max, h2_max) If it is determined to be a high-level conflict, a time check will be performed, where max() means to take the maximum value and min() means to take the minimum value; Time detection involves performing time window overlap analysis on spatial domain pairs determined to conflict in both space and altitude. Let the time ranges of the two spatial domain pairs be represented as intervals (t1_start, t1_end) and (t2_start, t2_end), where t1_start and t2_start are the starting values ​​of the two time ranges, and t1_end and t2_end are the ending values ​​of the two time ranges. If the following conditions are met: t1_start < t2_end and t2_start < t1_end If the two pairs are identified as spatiotemporal conflicts, the spatial domain pairs will be stored as spatiotemporal conflict pairs in the spatiotemporal conflict pair set.

5. The method for low-altitude airspace conflict detection and resolution according to claim 4, characterized in that, Step 4, which describes building a usable timeline model, specifically includes: Spatial constraints are extracted from the adjustable spatial domains in all spatiotemporal conflict pairs, that is, all fixed spatial domains that have spatial or height overlap with the adjustable spatial domain are extracted as potential spatial obstacles. Obtain the planned time periods for all potential spatial obstacles and project them onto a preset scheduling time axis. Calculate the complement on the time axis to generate an available time axis model for the adjustable spatial domain.

6. The method for low-altitude airspace conflict detection and resolution according to claim 5, characterized in that, The time scheduling optimization described in step 4 involves using an optimization strategy based on incremental state updates, with the goal of minimizing the overall time adjustment, and performing discretization optimization.

7. The method for low-altitude airspace conflict detection and resolution according to claim 6, characterized in that, The aforementioned discretization optimization specifically includes: Within the available time axis model of the adjustable spatial domain to be scheduled, a discretized grid search is performed with a preset time step. Construct a loss function with the absolute deviation between the candidate start time and the original planned start time as the independent variable; The candidate point that minimizes the loss function value is selected as the starting time for the new assignment.

8. A method for detecting and resolving low-altitude airspace conflicts according to claim 6, characterized in that, The optimization strategy based on incremental state updates specifically includes: Maintain a global dynamic scheduling state database. Once a new time period is successfully allocated to the current adjustable airspace, immediately synchronize its parameters to the global dynamic scheduling state database. When constructing a usable time axis model for subsequent adjustable spatial domains, in addition to extracting fixed spatial domains as constraints, dynamically extracting the already scheduled adjustable spatial domains that have spatial overlap with it from the global dynamic scheduling state library as constraints.

9. A method for detecting and resolving low-altitude airspace conflicts according to claim 6, characterized in that, Step 4, which involves time scheduling optimization, also includes heuristic scheduling and sorting, specifically: Before performing discretization optimization, spatial topology congestion weights are calculated based on the number of fixed spatial domains with adjustable spatial domain overlap and the number of other adjustable spatial domains. Based on the spatial topology congestion weight, the priority scheduling order of adjustable spatial domains is determined by sorting them in descending order.

10. A low-altitude airspace conflict detection and resolution system, characterized in that, To implement any of the methods described in claims 1-9, comprising: The system comprises a data receiving and preprocessing module, a multi-dimensional conflict detection module, an intelligent scheduling module, and a result output module; among which, The data receiving and preprocessing module is used to receive and standardize airspace usage plans. The multi-dimensional conflict detection module employs a hierarchical detection strategy to detect conflicts in the spatial, height, and temporal dimensions. The intelligent scheduling module, based on the conflict detection results, constructs a personalized available time axis model for the adjustable airspace with conflicts, and adopts an optimization algorithm based on incremental state updates to optimize time scheduling to resolve conflicts with the goal of minimizing the overall time adjustment. The results output module is used to output the detection and resolution results.

Images