Aircraft route conflict detection and rapid resolution method under established road network structure

Abstract

The invention discloses an aircraft route conflict detection and rapid resolution method under a set road network structure, which can establish a conflict detection model based on the set road network structure through a consistency space-time analysis method and by considering constraint conditions in a road network from the aspects of same-direction chasing, reverse crossing, channel saturation and the like. Qualitative and detection of conflicts among airways of a plurality of aircrafts are realized. And constructing a conflict tree based on a conflict detection result, and searching a feasible solution of an air route conflict resolution adjustment scheme through a Monte Carlo tree search and focusing subtraction method, so as to realize the conflict-free air route adjustment generation of the plurality of aircrafts. According to the method, the possible collision risk is found in advance by detecting the planned airways between the aircrafts, and a quick and feasible airway adjustment scheme is provided to eliminate the conflict risk, so that conflict-free airway planning of a large number of aircrafts under a set road network structure is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of air traffic control, and particularly relates to a method for detecting and quickly resolving flight path conflicts of aircraft under a given road network structure. Background Art

[0002] In some air traffic control scenarios, aircraft flights need to be carried out on specified fixed flight paths, and the task of maintaining the intervals between aircraft is completed by air traffic controllers. If flight tracks need to be changed in special situations such as conflicts, no-fly zones, danger zones, and extremely bad weather, it is necessary to strictly follow the instructions of air traffic controllers. However, with the continuous increase in air traffic flow, the flight paths will become more crowded, increasing the conflict risk of aircraft, thereby reducing the operating efficiency of the air traffic system and threatening aviation flight safety. In current control work, the detection of conflicts uses the minimum safety interval method, which often relies on the controller's prediction of the real-time flight path. For the resolution of conflicts, it also often relies on manual instruction mediation. These methods greatly reduce the control efficiency of the current airspace, especially when the number of aircraft in the road network increases, the difficulty of discovery and adjustment also increases exponentially. Therefore, for the pre-planned flight paths of aircraft, it is particularly important to determine and resolve conflicts between them.

[0003] Currently, for the conflict detection and conflict resolution methods of flight path tasks, due to the complex, continuous and mutually coupled characteristics of the time, space and frequency domains, the existing conflicts are difficult to decouple, and it is easy to affect the whole situation by pulling one hair. When a large number of flight paths are planned, this chain effect will lead to an "explosive" increase in the time-consuming of conflict resolution, and it is impossible to form a conflict-free large-scale aircraft planning scheme within a limited time. Summary of the Invention

[0004] Object of the Invention: The object of the present invention is to provide a method for detecting and quickly resolving flight path conflicts of aircraft under a given road network structure, to realize a flight path control method for a large number of aircraft pouring into the airspace with a determined road network structure, and to realize the rapid detection of conflicts between them and provide a resolution scheme.

[0005] Technical Solution: A method for detecting and quickly resolving flight path conflicts of aircraft under a given road network structure according to the present invention includes the following steps: Step 1: Define the elements in the given road network structure in the current spatial area, and the elements include channels, intersections and temporary channels; Step 2: Establish a conflict detection model based on consistent spatio-temporal analysis, and quantitatively analyze the same-direction pursuit conflicts, reverse intersection conflicts and channel saturation conflicts between the flight paths of multiple aircraft through spatio-temporal constraint modeling; Step 3: Construct a conflict resolution experience knowledge base to store the conflict resolution schemes for pairwise flight paths based on artificial experience; Step 4: Construct a conflict tree based on the conflict detection results in Step 2. Combining with the conflict resolution experience knowledge base in Step 3, use the Monte Carlo tree search and focused pruning method to search for feasible solutions for route adjustment; Step 5: Within a given time, select the optimal conflict-free route plan based on the total flight distance and total flight time evaluation indicators of the leaf nodes of the conflict tree, and adjust the flight route of the aircraft.

[0006] Furthermore, in Step 1, the channel is defined as a straight-line horizontal area connecting two points at the same altitude, including the upper bound altitude, lower bound altitude, channel width, starting longitude and latitude, ending longitude and latitude, and channel capacity attributes; the intersection is defined as a point in the channel, including the upper bound altitude, lower bound altitude, width, longitude, latitude, and safe passage time attributes, and each channel includes at least a starting intersection and an ending intersection; the temporary channel is defined as a straight-line horizontal area connecting intersections at the same altitude between different channels, including the starting longitude and latitude, ending longitude and latitude, altitude, and channel width attributes.

[0007] Furthermore, in Step 2, the spatio-temporal constraint modeling is as follows: Channel flow constraint model: ; where is the air traffic flow from the starting intersection of the route to a certain fixed intersection ; is the flow from a certain fixed intersection to the ending intersection of the route ; is the flow from the starting intersection of the route directly to the ending intersection of the route ; is the maximum capacity value of the intersection ; is the maximum capacity value of the channel ; is the total number of intersections; is the total number of starting intersections of the route; is the total number of ending intersections of the route; The safety constraint model includes: node flight interval constraint, route channel constraint; Task condition constraint model: ; In the formula, is the arrival time at the end of the route, is the task required time.

[0008] Furthermore, in step 2, the qualitative conflict analysis includes the following: same-direction overtaking conflict: two aircraft cross the same channel at the same altitude in the same direction, and the one that enters first exits last; opposite-direction crossing conflict: two aircraft cross the same channel at the same altitude in opposite directions; channel saturation conflict: the time interval between two aircraft passing the same intersection is less than the safe passage time.

[0009] Furthermore, in step 3, the conflict resolution experience knowledge base includes: same-direction pursuit conflict resolution solutions: speed adjustment, height adjustment or point adjustment for detour; reverse cross conflict resolution solutions: height adjustment, speed adjustment or point adjustment for detour; channel saturation conflict resolution solutions: speed adjustment or point adjustment for detour.

[0010] Furthermore, in step 4, the conflict tree node includes: conflict number, conflict cause, conflict aircraft number, and conflict channel number; the expansion process of the conflict tree is expressed as: ; in, Representative is located at Layer Node No., which contains the aircraft Conflict between ; The representative randomly extracts a solution from the conflict resolution solution library in a Monte Carlo manner and obtains the Layer Node .

[0011] Furthermore, the focused branch reduction method is as follows: Calculate the leaf node evaluation value: ; in, Represents the relevance of the conflict within the leaf node to all routes, Represents the impact of conflicts within leaf nodes on key tasks, Represents the airspace resources affected by the conflict within the leaf node; calculates the Q value for each leaf node and retains the node with the largest Q value in the current layer, and prunes the remaining nodes in the same layer.

[0012] Furthermore, step 5 is as follows: search for a conflict-free leaf node set T within time t; calculate the total range and total flight time evaluation values of each solution , select the minimum value Corresponding solutions As the final adjustment plan.

[0013] An electronic device according to the present invention includes a processor and a memory, wherein the memory stores a computer program, and the processor implements any one of the methods when executing the computer program.

[0014] A computer-readable storage medium according to the present invention stores a computer program, and when the computer program is executed by a processor, any one of the above methods is implemented.

[0015] Beneficial effects: Compared with the prior art, the present invention has the following remarkable advantages: (1) For a given road network structure, the element definitions and attributes included in the elements under this road network structure are clarified, laying a foundation for the subsequent solution of the conflict detection and conflict resolution problems of air routes; (2) A qualitative and quantitative detection method for the conflict relationship between air routes based on consistent spatio-temporal modeling is proposed, realizing a reasonable modeling of the spatio-temporal resources occupied by the flight routes of aircraft, and quickly and efficiently analyzing all conflicts existing between air routes; (3) A conflict tree structure is constructed to describe the current conflict of air route methods. Based on the Monte Carlo tree search method, the subtrees for conflict resolution at the lower layer are explored, and finally the leaf nodes without conflicts between air routes are obtained, transforming the air route conflict resolution problem into the problem of finding the optimal leaf nodes of the conflict tree; (4) Through the heuristic key conflict focusing method, the core conflicts are quickly located to prune the conflict tree nodes, reducing the subsequent search scale while ensuring the optimality of the subsequent search results as much as possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is the algorithm flowchart of the consistent spatio-temporal flight conflict model of the present invention; Figure 2 is the implementation method diagram of multi-directional conflict detection of the present invention; Figure 3 is an example diagram of the same-direction pursuit relationship of the present invention; Figure 4 is an example diagram of the reverse intersection relationship of the present invention; where (A) represents that aircraft 1 has a reverse intersection conflict with level-flying aircraft 2 when descending; (B) represents that aircraft 1 has a reverse intersection conflict with level-flying aircraft 2 when ascending; (C) represents that aircraft 1 and aircraft 2 have a conflict when flying in reverse at the same altitude; (D) represents that aircraft 1 has a reverse intersection conflict with descending aircraft 2 when ascending; (E) represents that aircraft 1 has a reverse intersection conflict with ascending aircraft 2 when ascending; (F) represents that aircraft 1 has a reverse intersection conflict with descending aircraft 2 when descending; Figure 5 is a schematic diagram of the conflict tree construction based on the conflict resolution solution library of the present invention; Figure 6 is the conflict tree search flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0017] The technical solutions of the present invention will be further described below with reference to the accompanying drawings.

[0018] As Figure 1As shown, an embodiment of the present invention provides a method for detecting and quickly resolving aircraft route conflicts under a given road network structure, and the specific steps are as follows: S1, determine the definitions of various elements in the established road network structure within the current space that can be used for the safe passage of aircraft. The elements in the road network include channels, intersections, temporary channels, etc.

[0019] (11) Elements in the road network include channels, intersections, temporary channels, etc. The specific definition of a channel is a straight horizontal area connecting two points at the same height in space, which includes attributes such as upper boundary height, lower boundary height, channel width, starting point longitude and latitude, end point longitude and latitude, and channel capacity; (12) The specific definition of an intersection is a point in a channel, including attributes such as upper bound height, lower bound height, width, longitude, latitude, and safe passing time. A channel includes at least two intersections, namely the starting intersection and the ending intersection; (13) The specific definition of a temporary channel is a straight horizontal area between different channels that connects intersections at the same height. Its height is consistent with the intersection height and includes the latitude and longitude of the starting point, the longitude and longitude of the end point, the height, the channel width and other attributes.

[0020] S2, determine the type of airspace conflict under the current established road network structure, design a conflict detection model based on consistent spatiotemporal analysis, and through modeling the spatiotemporal constraints of the established road network structure, achieve rapid qualitative and quantitative analysis of conflicts such as same-direction overtaking and reverse intersection between the routes planned by multiple aircraft.

[0021] The algorithm flow of the consistent space-time flight conflict model is as follows Figure 1 First, the established road network structure is modeled as a directed graph Secondly, a spatiotemporal constraint model is established under the given road network structure, which includes channel flow constraints, two types of safety constraints, and mission condition constraints. The specific definitions are as follows: Channel flow constraint model: ; In the formula, The starting intersection of the route Arrive at a fixed intersection air traffic flow; For a fixed intersection To the end of the route Traffic volume; The starting point of the route Go directly to the end of the route Traffic volume; For intersection The maximum capacity value; For channel The maximum capacity value; is the total number of intersections; is the total number of route starting intersections; is the total number of route terminal intersections.

[0022] Two types of security constraint models: Category 1: Node flight interval constraints: ; In the formula, For the intersection aircraft and the The arrival time difference of the aircraft, Provide safe crossing time intervals at intersections; Category 2: Route channel constraints: ; In the formula, For the The flow rate of the airway channel, For the The maximum capacity of each airway channel; Task condition constraint model: ; In the formula, is the time of arrival at the route destination, Ask for time for tasks.

[0023] Finally, multi-directional conflict detection and classification of intersection conflicts and waterway conflicts are realized. The specific implementation method is as follows: Figure 2 shown.

[0024] Before conflict detection, all aircraft conflict scenarios are divided into the following 13 categories based on the heading relationship between aircraft routes and the motion status of aircraft on the routes, as shown in Table 1.

[0025] Table 1 Classification of two-aircraft conflict scenarios ; Flight aircraft conflict detection and resolution first determines the state relationship and position relationship of the aircraft, establishes a decision tree for conflict type judgment, and after determining the corresponding conflict type, selects the corresponding conflict resolution method from the established resolution rule library to resolve it.

[0026] (a) Overtaking in the same direction Same-direction pursuit conflicts are a common type of aircraft conflict, often occurring during the en-route flight phase, or cruising phase. Same-track conflicts can be resolved by converting them into pursuit issues. These conflicts typically occur when the longitudinal separation between aircraft decreases due to the following aircraft's speed exceeding the leading aircraft's.

[0027] Taking the case of two aircraft in the same direction in a climbing state in a chasing conflict as an example, the conflict detection process can be described as follows: First, based on the heading relationship between the two aircraft, it is determined that the two aircraft are in the same direction. After 5 minutes, the horizontal distance between the two aircraft and the altitude difference are taken. and are compared with the conflict standard value. If and are both less than the conflict standard value, it indicates that a conflict is detected. The relationship between the two aircraft in the same-track conflict is as shown in Figure 3 .

[0028] (b) Reverse crossing Similar to the chasing conflict in the same direction, the reverse crossing conflict mostly occurs during the en-route cruise phase and is also likely to occur during the aircraft takeoff and landing phases. The conflict detection of reverse crossing can be summarized as an encounter problem.

[0029] Taking the case of two aircraft in a climbing state in a reverse crossing conflict as an example, the conflict detection process can be described as follows: First, based on the heading relationship between the two aircraft, it is determined that the two aircraft are in a reverse relationship. After 5 minutes, the horizontal distance between the two aircraft and the altitude difference are taken. and are compared with the conflict standard value. If and are both less than the conflict standard value, it indicates that a conflict is detected. The relationship between the two aircraft in the reverse crossing conflict is as shown in group Figure 4 ; among them, (A) represents that aircraft 1 has a reverse crossing conflict with level-flying aircraft 2 when aircraft 1 descends in altitude; (B) represents that aircraft 1 has a reverse crossing conflict with level-flying aircraft 2 when aircraft 1 ascends in altitude; (C) represents that aircraft 1 and aircraft 2 have a conflict when flying in reverse at the same altitude; (D) represents that aircraft 1 has a reverse crossing conflict with aircraft 2 that is descending in altitude when aircraft 1 ascends in altitude; (E) represents that aircraft 1 has a reverse crossing conflict with aircraft 2 that is ascending in altitude when aircraft 1 ascends in altitude; (F) represents that aircraft 1 has a reverse crossing conflict with aircraft 2 that is descending in altitude when aircraft 1 descends in altitude.

[0030] (c) Intersection congestion conflict Regarding the intersection , if the time interval between the passing aircraft and the aircraft is less than 5 minutes, it is considered that there is an intersection congestion conflict, that is: .

[0031] S3. Based on manual experience, conflict resolution schemes for pairwise air routes are set up to obtain a conflict resolution experience knowledge base.

[0032] For different conflict types, different resolution solution libraries are constructed for subsequent selection when resolving route conflicts.

[0033] (d) Same-direction pursuit and conflict resolution solutions Option 1: Speed control, speed up the path of the aircraft in front and slow down the path of the aircraft behind; Option 2: Adjust the height, lower the path of the aircraft in front; raise the path of the aircraft in front; raise the path of the aircraft behind; lower the path of the aircraft behind; Option three: Adjust the route and take another detour.

[0034] (e) Reverse cross-conflict resolution scheme Option 1: Adjust the height, lower the route of the aircraft in front and raise the route of the aircraft behind; raise the route of the aircraft in front and lower the route of the aircraft behind; Option 2: Speed adjustment, accelerating the aircraft in one direction and decelerating the aircraft in another direction; Option three: Adjust the route and take another detour.

[0035] (f) Channel saturation conflict Option 1: Speed control, speed up the path of the aircraft in front and slow down the path of the aircraft behind; Option 2: Adjust the route and take another detour.

[0036] S4, based on the conflict results given in the second step, establish a conflict tree structure, and search for feasible solutions for the route adjustment plan based on the conflict resolution experience knowledge base set in the third step and the Monte Carlo tree search and focused branch reduction method.

[0037] Construct a conflict tree, in which each node describes the conflicts between multiple routes, such as cross-collision, same-direction overtaking, and channel saturation, in the form of numbers. The specific attributes include conflict sequence number, conflict cause, aircraft Serial number, aircraft Serial number, aircraft Channel number and aircraft number where the route conflicts The information of the channel number where the route conflicts occurs is used to build a conflict tree based on the conflict resolution solution library, such as Figure 5As shown below. After that, Monte Carlo tree search is used to search for and solve the conflict tree. The Monte Carlo tree search algorithm is a method that can take random samples in a specific decision space and build a search tree based on the obtained results to find the optimal decision. This algorithm has wide applicability in the problem of sequential decision trees. The basis of the Monte Carlo tree search algorithm is Monte Carlo simulation, which finds the "best" action in the current state by constructing an asymmetric optimal search tree. The advantage of the Monte Carlo tree search algorithm is that it does not need to know the complete Markov decision process model in advance, but only requires a low-level generation model - that is, a simulator or emulator. The main idea of the Monte Carlo tree search algorithm is to evaluate a state based on the simulation data obtained from a state node. Specifically, for each state node, the Monte Carlo tree search algorithm selects an action according to the current action selection strategy; and executes this action through Monte Carlo simulation; observes the new state. If this state already exists in the tree, recursively evaluate this state, otherwise insert this state node into the tree and simulate the execution of the default strategy from this state until the termination condition is encountered; finally, update the statistics of each state on the path through backtracking.

[0038] The schematic diagram of the Monte Carlo algorithm search process for the entire airway conflict tree is shown in Figure 6 , and the search of this Monte Carlo tree includes 6 steps: selection, expansion, simulation, focused search pruning, end, and termination, which are specifically described as follows: (41) Selection In the selection stage, starting from the root node, that is, an initial situation is assigned to the airspace grid that needs to be decided , and then a most urgent node to be expanded is selected downward , and the situation is the first node to be checked in each iteration; the inspection results include three possibilities: (411) all the actionable actions in this node have been expanded; (412) this node has actionable actions but has not been expanded; (413) this node has been expanded and has ended. For the above three possibilities, the following treatments are carried out respectively: (i) If all actionable actions have been expanded, calculate the performance index values of all the child nodes contained in this node, and find the child node with the largest performance index value, and then continue to check the next child node. And repeatedly iterate and calculate the next child node.

[0039] (ii) If there are still unexpanded child nodes in the inspected situation (for example, a node includes 30 actionable actions, but only 12 child nodes have been created in the search tree), then this child node is considered the target node of this iteration , and find out An action that has not been expanded , and then perform the expansion step.

[0040] (iii) If the node being examined is an ended node. Then directly perform the backpropagation step from this node.

[0041] (42) Expansion At the end of the selection phase, find a node that needs to be expanded , and an action that has not been expanded . At the same time, create a new node in the search tree as a new child node of. The situation of is the node after performing the action . The action of the expansion step represents using the method in different conflict resolution schemes to resolve any pair of conflicts. Then the expansion process of this conflict tree is: ; Among them, represents the j-th node at the i-th layer, which contains the aircraft the conflict between , represents randomly selecting a scheme from the conflict resolution scheme library in a Monte Carlo manner to obtain the k-th node at the i + 1-th layer ; (43) Simulation To let get an initial score. At the start of from , it is randomly performed until an end state is obtained, and this end state will be used as the initial score. Here, generally the success / failure of resolution is used as the scoring criterion for the result, that is, 1 or 0.

[0042] (44) Focused search pruning Based on the focused search heuristic, couple the route relevance in the actual large - scale aircraft routes , the impact on critical missions , the airspace crossing and other influencing factors to obtain the focused search heuristic , implement the evaluation of each leaf node, traverse the values of all nodes within the current layer of the conflict tree, retain the node with the largest value, prune all other nodes in the same layer, and continue the search based on the current node.

[0043] (45) End At After the simulation of ends, its parent node and all nodes on the path from the root node to

[0044] will add their cumulative scores according to the result of this simulation. If an ending is directly found in the selection step, the score can be updated according to this ending.

[0045] (46)Termination The constructed tree is essentially a multi-way search tree, and its expansion method is based on the resolution stack, so it is not complete. It may not be able to reach the leaf node due to resource deadlock. Therefore, when the search depth is greater than the set value and the number of existing conflicts is less than the set value , the search terminates and the result of the current node is returned.

[0046] S5. Within the established time, based on the leaf nodes of multiple conflict trees searched, determine the better leaf node with the evaluation indicators of the total route and total flight time as the solution plan for this conflict resolution process.

[0047] Determine the time , search for the conflict tree by the method of Monte Carlo tree search and focused pruning within the effective time to obtain a set of complex leaf nodes , calculate the total flight time and total voyage of each conflict-free route plan in this set , take the minimum value , obtain the conflict-free route plan with the minimum total flight time and total voyage , determine this plan as the final plan for this conflict resolution, and adjust the routes of each aircraft.

Claims

1. A method for detecting and quickly resolving flight path conflicts of aircraft under a given road network structure, characterized in that It includes the following steps: Step 1: Define the elements in the established road network structure within the current spatial region. The elements include channels, intersections, and temporary channels; Step 2: Establish a conflict detection model based on consistent spatio-temporal analysis, and quantitatively analyze the same-direction pursuit conflicts, reverse intersection conflicts, and channel saturation conflicts among the routes of multiple aircraft through spatio-temporal constraint modeling; Step 3: Construct a conflict resolution experience knowledge base to store the conflict resolution schemes for pairwise route conflicts based on manual experience; Step 4: Construct a conflict tree according to the conflict detection results in Step 2. Combining with the conflict resolution experience knowledge base in Step 3, use the Monte Carlo tree search and focused pruning method to search for feasible solutions for route adjustment; Step 5: Within a given time, select the optimal conflict-free route plan based on the total flight distance and total flight time evaluation indicators of the leaf nodes of the conflict tree, and adjust the flight routes of the aircraft.

2. The method for detecting and quickly resolving flight path conflicts of an aircraft under a given road network structure according to claim 1, characterized in that, In Step 1, the channel is defined as a straight-line horizontal area connecting two points at the same height, including upper bound height, lower bound height, channel width, starting longitude and latitude, ending longitude and latitude, and channel capacity attributes; the intersection is defined as a point in the channel, including upper bound height, lower bound height, width, longitude, latitude, and safe passing time attributes. Each channel includes at least a starting intersection and an ending intersection; the temporary channel is defined as a straight-line horizontal area connecting intersections at the same height between different channels, including starting longitude and latitude, ending longitude and latitude, height, and channel width attributes.

3. A method for detecting and quickly resolving flight path conflicts under a given road network structure according to claim 1, characterized in that, In Step 2, the spatio-temporal constraint modeling is as follows: Channel flow constraint model: ； Among them, is the starting intersection of the route to reach a certain fixed intersection of the air traffic flow; is a certain fixed intersection to the ending intersection of the route of the flow; is the starting intersection of the route directly to the ending intersection of the route of the flow; is the intersection of the maximum capacity value; is the channel of the maximum capacity value; is the total number of intersections; is the total number of starting intersections of the route; is the total number of ending intersections of the route; The safety constraint model includes: Node flight interval constraint, route channel constraint; Task condition constraint model: ; In the formula, is the arrival time at the route end point, is the mission required time.

4. A method for detecting and quickly resolving flight path conflicts under a given road network structure according to claim 1, characterized in that, In Step 2, the conflict qualitative analysis includes: Same-direction pursuit conflict: Two aircraft cross the same channel at the same height in the same direction, and the one that enters first exits later; Reverse intersection conflict: Two aircraft cross the same channel at the same height in opposite directions; Channel saturation conflict: The time interval between two aircraft passing through the same intersection is less than the safe passing time.

5. A method for detecting and quickly resolving flight path conflicts of an aircraft under a given road network structure according to claim 1, characterized in that, In Step 3, the conflict resolution experience knowledge base includes: Same-direction pursuit conflict resolution scheme: Speed adjustment, altitude adjustment, or point adjustment for detour; Reverse intersection conflict resolution scheme: Altitude adjustment, speed adjustment, or point adjustment for detour; Channel saturation conflict resolution scheme: Speed adjustment or point adjustment for detour.

6. The method for detecting and quickly resolving flight path conflicts of an aircraft under a given road network structure according to claim 1, wherein In Step 4, the conflict tree nodes include: Conflict serial number, conflict reason, conflict aircraft serial number, conflict channel serial number; The expansion process of the conflict tree is expressed as: ； Among them, represents the th layer of node No., which contains the conflict between aircraft ; represents randomly extracting a solution from the conflict resolution solution library in a Monte Carlo manner to obtain the th layer of node No. .

7. A method for detecting and quickly resolving flight path conflicts of an aircraft under a given road network structure according to claim 1, characterized in that, In Step 4, the focused pruning method is as follows: Calculate the evaluation value of the leaf node: ； Among them, represents the relevance of conflicts within leaf nodes to all air routes, represents the impact of conflicts within leaf nodes on critical tasks, represents the airspace resource situation affected by conflicts within leaf nodes; calculate the Q value for each leaf node and retain the node with the largest Q value in the current layer, and prune the remaining nodes in the same layer.

8. A method for detecting and quickly resolving flight path conflicts of an aircraft under a given road network structure according to claim 1, characterized in that, Step 5 is specifically as follows: Search for the conflict-free leaf node set T within time t; Calculate the total voyage and total flight time evaluation values of each plan , and select the minimum value corresponding plan as the final adjustment plan.

9. An electronic device, characterized in that, It includes a processor and a memory. The memory stores a computer program, and when the processor executes the computer program, it implements the method described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, A computer program is stored, and when the computer program is executed by a processor, it implements the method described in any one of claims 1-8.

Images