Airport scene sliding path generation method, electronic equipment and storage medium

Abstract

The invention provides an airport scene taxiing path generation method, electronic equipment and a storage medium, and the method comprises the steps: grouping aircrafts in an unplanned flight sequence, randomly initializing a path planning sequence of the aircrafts in each to-be-planned group, and generating a taxiing path of each aircraft in each to-be-planned group; determining whether path conflicts exist among all aircrafts in the current unplanned flight sequence and the current planned flight sequence, if so, updating the current unplanned flight sequence and the current planned flight sequence, otherwise, acquiring the average sliding time of all the current aircrafts, and adding the average sliding time into a current average sliding time list; and after iteration is finished, taking the sliding path generation information corresponding to the minimum average sliding time in the current average sliding time list as target sliding path generation information. According to the invention, the taxiing efficiency and safety performance of the aircraft on the airport scene can be improved.

Description

Technical Field

[0001] The invention relates to the field of aircraft taxiing path optimization, and in particular to an airport taxiing path generation method, electronic equipment and storage medium. Background Art

[0002] With the rapid development of civil aviation transportation, airport business volume and the number of flight takeoffs and landings have continued to rise. However, compared with the rapid growth of air transportation volume, due to resource constraints and cost pressure, the speed of airport infrastructure construction is relatively slow, and airport surface resources are gradually in short supply, resulting in problems such as low aircraft surface taxiing speed, poor aviation economy, queuing and congestion delays. The taxiway is the only path connecting the runway and the apron. In large airports with multiple runways, the structure of the taxiway is very complex, and the average taxiing time of aircraft is relatively long. How to improve the operating efficiency of the taxiway system and shorten the average taxiway occupation time of aircraft as much as possible has become a key issue for the efficient operation of large airport flight areas. In current practice, airports use complex decision support tools to effectively manage ground traffic. If there is no flight change, the scheduled taxiing plan can be smoothly executed. However, flights are not only easily affected by uncontrollable factors such as weather, but may also be subject to sudden delays, cancellations or changes due to various reasons such as air traffic control, mechanical failures, personnel scheduling, airport congestion, etc., making the actual operation of air transportation more complex and unpredictable. Therefore, it is very necessary to readjust the taxiing strategy of the flight in time after an emergency occurs. Specifically, when a flight is delayed or cancelled, it is necessary to quickly release the occupied nodes and adjust the overall taxiing strategy. In special cases such as flight diversion, it is particularly important to arrange a suitable taxiing path in time. However, how to quickly and effectively allocate taxiing paths remains a huge challenge. At present, research on the problem of airport surface path optimization has achieved certain results. These studies usually generate the optimal taxiing path for each aircraft in turn on the basis of ensuring that the taxiing path is conflict-free. However, this method has the restriction that the path generated by the preceding aircraft will affect the path generated by the subsequent aircraft, resulting in the final path often being a local optimal solution rather than a global optimal solution. If multiple aircraft can be jointly planned, the taxiing paths of each aircraft will not be affected by the generation order, but the taxiing sequence combination of multiple aircraft will be generated simultaneously within a period of time. This can effectively avoid the aircraft taxiing sequence combination from falling into the local optimal solution. Therefore, how to improve the performance of the airport surface path optimization algorithm from the perspective of multi-aircraft joint path planning is crucial to improving aircraft taxiing efficiency and ensuring safety. Summary of the invention

[0003] In view of the above technical problems, the technical solution adopted by the present invention is:

[0004] According to a first aspect of the present invention, a method for generating an airport surface taxi path is provided, the method comprising the following steps:

[0005] S100, constructing a taxi network quadruple (G, q, o, p) based on the airport map and the information of arriving and departing aircraft within a set time period, wherein G is the airport taxiway graph structure, G = (V, E, W), V is a node set formed by nodes on the airport surface, E is an edge set, the edges in E represent the taxiways connecting the nodes, W is a taxi time set formed by the taxi time between adjacent nodes, q is a source node set formed by the source nodes of the aircraft within the set time period, o is a target node set formed by the target nodes of the aircraft within the set time period, and p is a departure time set formed by the timestamps of the aircraft departing from the source nodes within the set time period, and the nodes represent the intersections between parking stands and taxiways on the airport surface, or the intersections between runways and taxiways.

[0006] S200, setting the iteration counter C=1.

[0007] S300, if C≤C0, add all aircraft within the set time period to the current unplanned flight sequence and set the current planned flight sequence to empty, and execute S400; otherwise, execute S800; C0 is the maximum number of iterations.

[0008] S400, grouping the aircraft in the current unplanned flight sequence to obtain a plurality of groups to be planned, and randomly initializing the path planning order of the aircraft in each group to be planned, and generating a taxi path for each aircraft in each group to be planned based on a preset objective function and preset constraints, to obtain current taxi path generation information.

[0009] S500, based on the current taxi path generation information, determine whether there is a path conflict between all aircrafts, if there is a path conflict, execute S600; if there is no path conflict, execute S700.

[0010] S600, update the current unplanned flight sequence and the current planned flight sequence, obtain the updated unplanned flight sequence and the updated planned flight sequence, and use the updated unplanned flight sequence and the updated planned flight sequence as the current unplanned flight sequence and the current planned flight sequence respectively, and execute S400.

[0011] S700, based on the current taxi path generation information, obtain the current average taxi time of all aircraft, and add the obtained current average taxi time of all aircraft to the current average taxi time list; set C=C+1, and execute S300; wherein, the initial value of the current average taxi time list is empty.

[0012] S800: Use the glide path generation information corresponding to the minimum average glide time in the current average glide time list as the target glide path generation information. According to a second aspect of the present invention, there is provided an electronic device, comprising a processor and a memory; the processor is used to execute the steps of the method according to the first aspect of the present invention by calling a program or instruction stored in the memory.

[0013] According to a second aspect of the present invention, there is provided a computer-readable storage medium storing a program or instructions, wherein the program or instructions enable a computer to execute the steps of the method according to the first aspect of the present invention.

[0014] The present invention has at least the following beneficial effects:

[0015] The method for generating a taxiway path at an airport provided by an embodiment of the present invention comprises: grouping aircraft in a current unplanned flight sequence to obtain a plurality of groups to be planned, and randomly initializing the path planning order of the aircraft in each group to be planned, and generating the taxiway path of each aircraft in each group to be planned based on a preset objective function and preset constraints to obtain current taxiway path generation information; based on the current taxiway path generation information, determining whether there is a path conflict between all aircraft in the current unplanned flight sequence and the current planned flight sequence, if there is a path conflict, updating the current unplanned flight sequence and the current planned flight sequence, if there is no path conflict, obtaining the current average taxiing time of all aircraft based on the current taxiway path generation information, and adding the obtained current average taxiing time of all aircraft to the current average taxiing time list; after the iteration is completed, the taxiway path generation information corresponding to the minimum average taxiing time in the current average taxiing time list is used as the target taxiway path generation information. The present invention can generate taxiway sequence combinations of multiple aircraft at the same time within a period of time, so that the aircraft taxiway sequence combination can be effectively prevented from falling into a local optimal solution, thereby improving the taxiing efficiency and safety performance of aircraft at the airport.

[0016] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A flow chart of a method for generating an airport surface taxi path provided by an embodiment of the present invention;

[0019] Figure 2 Generate schematic diagrams for glide paths;

[0020] Figure 3a to Figure 3d This is a comparison of the experimental results;

[0021] Figures 4a to 4c A schematic diagram of the effect of selecting a flight sequence to be planned in the airport surface taxiway path generation method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0024] It should be noted that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the steps as sequential processes, many of the steps can be implemented in parallel, concurrently or simultaneously. In addition, the order of the steps can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0025] In the path planning of aircraft arriving and departing from the airport, each aircraft can be regarded as a separate intelligent agent, taxiing at a constant speed on the taxiway. The airport taxiway is a shared environment, and all aircraft arriving and departing within a period of time constitute a multi-agent system. When an aircraft anticipates a conflict, it waits at a node. Therefore, the joint taxiing path planning of all aircraft within a period of time can be regarded as a multi-agent path finding (MAPF) problem, and the optimization of the path planning of aircraft arriving and departing can be converted into a multi-agent path finding problem for solution.

[0026] However, compared with the traditional MAPF problem, the aircraft taxiing path planning problem has the following characteristics.

[0027] (1) The departure time of agents is different: In the traditional MAPF problem, all agents start at the same time; however, in aircraft taxiway path planning, the arrival and departure times of aircraft are different, that is, the departure time of aircraft entering the taxiway is different. Specifically, for an incoming aircraft, the landing time on the runway is used as its departure time, and for a departing aircraft, the off-block time is used as its departure time.

[0028] (2) The distances between adjacent nodes are different: In the traditional MAPF problem, the distance between any two adjacent nodes is equal, which only requires one time step of the agent; however, in aircraft taxi path planning, the distances between nodes are different, and the taxiing time of the aircraft between different taxiway nodes is different;

[0029] (3) Resources may be released after reaching the target node: In the traditional MAPF problem, after the agent reaches the corresponding target node, it will always occupy the target node; while in the aircraft taxi path planning, the aircraft will leave the taxiway after reaching the runway entrance, releasing the node resources. Therefore, the present invention improves the traditional MAPF modeling method and constructs a MAPF model of aircraft taxiing in view of the characteristics of the aircraft taxi path planning problem. Specifically, the embodiment of the present invention provides a method for generating an airport taxi path, such as Figure 1 As shown, the method may include the following steps:

[0030] S100, constructing a taxi network quadruple (G, q, o, p) based on the airport map and the information of arriving and departing aircraft within a set time period, wherein G is the airport taxiway graph structure, G = (V, E, W), V is a node set formed by nodes on the airport surface, E is an edge set, the edges in E represent the taxiways connecting the nodes, W is a taxi time set formed by the taxi time between adjacent nodes, q is a source node set formed by the source nodes of the aircraft within the set time period, o is a target node set formed by the target nodes of the aircraft within the set time period, and p is a departure time set formed by the timestamps of the aircraft departing from the source nodes within the set time period, and the nodes represent the intersections between parking stands and taxiways on the airport surface, or the intersections between runways and taxiways.

[0031] In the embodiment of the present invention, the aircraft taxiing network is abstracted into a four-tuple (G, q, o, p) by using graph theory. Assuming that there are n aircraft arriving and departing in a certain time period, the output of the taxiing path planning problem is a set of two-tuple sequences ((π1, u1), ..., (π b ,u b ), ..., (π n ,u n)), where π b =(a1, ..., a b , ..., a n ), is the taxiing sequence of aircraft b from the source point s(b) to the target node o(b), u b =(t1, ..., t b ,……,t n ) records the taxiing time of aircraft b on each taxiway from departure to taxiing end. a is the taxiing function of the aircraft, a(v) = v next Indicates that the aircraft at node v performs taxiing action a and moves to node v next , (v, v next )∈E.π b [x] = a x (a x-1 (···a1( s(b) ))), indicating that the aircraft starts from the source point s(b) and executes π b The position of the first x segment after sliding. b [x] = p(b) + t1 + ... + t x Indicates that the aircraft has completed π from the source point b The time after the first x taxi segments in the equation is p(b), and p(b) is the departure time of aircraft b. If π b [|π b |]=o(b), then the binary sequence (π b ,u b ) is the taxi path planning for aircraft b.

[0032] In the embodiment of the present invention, the set time period can be set based on actual needs, such as one day.

[0033] S200, setting the iteration counter C=1.

[0034] S300, if C≤C0, add all aircraft within the set time period to the current unplanned flight sequence and set the current planned flight sequence to empty, and execute S400; otherwise, execute S800; C0 is the maximum number of iterations.

[0035] In the embodiment of the present invention, C0 can be set based on actual needs.

[0036] S400, grouping the aircraft in the current unplanned flight sequence to obtain a plurality of groups to be planned, and randomly initializing the path planning order of the aircraft in each group to be planned, and generating a taxi path for each aircraft in each group to be planned based on a preset objective function and preset constraints, to obtain current taxi path generation information.

[0037] In the embodiment of the present invention, the number of each group to be planned may be the same. The taxi path generation information may be a set of two-tuple sequences formed by the node IDs on the taxi paths of all aircraft and the node time of reaching each node.

[0038] In the embodiment of the present invention, the objective function is to minimize the average taxiing time of all aircraft arriving and departing from the airport within a certain period of time (for example, one day), that is, the objective function satisfies the following conditions:

[0039] T=min((∑ n b＝1 (T b end -T b begin )) / n);

[0040] T is the objective function, T b end is the time when the bth aircraft arrives at the parking stand or runway, T b begin It is the time when the bth aircraft leaves the parking stand or runway. The value of b ranges from 1 to n, and n is the total number of aircraft take-offs and landings in the set time period.

[0041] The arrival or departure time of aircraft arriving or departing from the airport varies, so the time when each aircraft starts taxiing is also different. Although the taxiway path of aircraft can be planned according to the sequence of the flight schedule, this simple and direct planning method may not achieve the shortest average taxiing time. Figure 2 As shown in the figure, nodes M1-M8 are all taxiway nodes. Aircraft P1 and P2 will depart from parking nodes A1 and A2 at time t+1 and time t respectively to the target locations (runway entrance nodes) G1 and G2. Assume that the taxiing time of the two aircraft on each taxiway is 1. There are two planning sequences for the paths of the two aircraft:

[0042] (1) First, plan P2. At time t+1, P2 glides to M8. At this time, P2 has two sliding action options. Suppose that it chooses to glide to M7 at time t+2, and then reaches the target node after 4 time steps;

[0043] Then plan P1. At time t+1 and time t+2, P1 and P2 will have vertex conflict and following conflict respectively. P1 needs to wait in place and glide to M7 after the conflict is resolved. The average glide time obtained under this planning sequence is 6.

[0044] (2) Prioritize planning P1. P1 reaches the target location via the optimal path A1→M7→M6→M5→G1. Then plan P2. To avoid possible conflicts with P1, the optimal path of P2 is A2→M8→M4→M3→M2→M1→G2. Under this planning sequence, the aircraft does not wait and the average taxiing time is 5. Compared with prioritizing planning P2, prioritizing planning P1 results in a shorter average taxiing time.

[0045] The above examples show that adjusting the path planning order of take-off and landing aircraft will produce a shorter average taxiing time. However, busy large airports have a large number of take-off and landing aircraft in a day. If the planning order of all aircraft is randomly adjusted, the convergence speed of the algorithm may decrease. In the embodiment of the present invention, all aircraft are divided into several planning combinations, and the aircraft within each planning combination are randomly planned and sorted, and path planning is performed in sequence. Specifically, assuming that there are n incoming and outgoing aircraft at an airport in a day, all aircraft are numbered from 1 to n according to the flight schedule. The aircraft in the flight sequence to be planned are divided into multiple groups to be planned, with y aircraft in each group. Random planning and sorting the aircraft in each group, and performing path planning in sequence, can solve the problem of slow algorithm convergence speed that may be caused by completely random sorting, while avoiding the constraints caused by the fixed order of aircraft before and after the time period interval.

[0046] In the embodiment of the present invention, when generating a taxi path for an aircraft, it should be ensured that any taxi action of the aircraft will not conflict with the taxi paths of aircraft in the planned flight sequence. Specifically, the preset constraint condition satisfies the following conditions:

[0047]

[0048] Where x represents the xth taxiing action of the ith aircraft in the unplanned flight sequence, π i [x] represents the position of the i-th aircraft after performing the x-th taxiing action from the corresponding source node, u i [x] represents the time after the i-th aircraft starts to perform the x-th taxiing action from the corresponding source node, i ranges from 1 to z, z is the number of aircraft in the unplanned flight sequence, and H is the node information table formed by the node information occupied by the aircraft in the planned flight sequence. H = {h u} u＝1……m ,h u is the node information occupied by the u-th aircraft in the planned flight sequence, h u =(N uv ) u＝1……f(u) , N uv h u The vth node information in N uv =(Ncuv , tc uv ), Nc uv is the ID of the vth node occupied by the uth aircraft in the planned flight sequence, tc uv is the time of the vth node occupied by the uth aircraft in the planned flight sequence, u ranges from 1 to m, m is the number of aircraft in the planned flight sequence, and v ranges from 1 to f(u), f(u) is the number of nodes occupied by the uth aircraft in the planned flight sequence.

[0049] Furthermore, in S400, the taxiing path of any aircraft is obtained through the following steps:

[0050] S401, add the node characteristic information corresponding to the source node of any aircraft to the corresponding list of node information to be visited; the node characteristic information includes the node ID, the node access time, the node access weight, the node's parent node ID and the parent node access weight.

[0051] In the embodiment of the present invention, the access weight of the source node is 0, and the ID of the parent node and the access weight of the parent node are null values.

[0052] S402, determine whether there is node feature information to be visited in the current list of node information to be visited. If not, execute S200. If so, take out the node feature information of the head node in the current list of node information to be visited and store it in the visited node information list corresponding to any aircraft, and determine whether the head node is the target node of any aircraft. If yes, execute S408. If not, execute S403.

[0053] In the embodiment of the present invention, the initial value of the visited node information list corresponding to any aircraft is empty. If the node feature information to be visited does not exist in the current node information list to be visited, it means that the target node is unreachable, that is, there is no taxi path to the target node, and the return value does not exist, which means that the taxi path planning has failed and needs to be re-planned.

[0054] In the embodiment of the present invention, the head node is the node located at the first position in the current list of node information to be visited.

[0055] S403, based on the current node access time record information table and preset constraints, determine the adjacent nodes accessible to the head node and the access time to form a current accessible node information set; the current accessible node information set includes multiple accessible node information, each accessible node information includes the ID of the corresponding accessible adjacent node and the corresponding access time, and the current node access time record information table includes the ID of each node in G and the corresponding access time information; execute S404.

[0056] In the embodiment of the invention, the node ID may be the number of the node. The node ID and the corresponding access time in the current node access time record information table may be represented by a two-tuple sequence, that is, the current node access time record information table may be represented as {(a1, ts1), ..., (a y ,ts y ),……,(a Q ,ts Q )},a y represents the ID of the yth node in G, ts y is the access time information of the yth node. The access time information is used to represent the time when the node is occupied. For example, the access time information of node 1 is 12:00 and 13:30, which means that node 1 will be occupied at 12:00 and 13:30.

[0057] In the embodiment of the present invention, the access time of the accessible adjacent node is the arrival time to the accessible adjacent node. The accessible adjacent nodes of the head node can be obtained based on the following steps:

[0058] S4031, obtain the arrival time of the head node to each adjacent node, and form a candidate adjacent node access information table I = {(a1, t1), ..., (a i1 , t i1 ),……,(a A , t A )},a i1 is the ID of the i1th adjacent node of the head node, t i1 is the arrival time of the head node to the i1th adjacent node, i1 ranges from 1 to A, and A is the number of adjacent nodes of the head node;

[0059] S4032, obtaining nodes whose access time does not conflict with the arrival time in I from the current node access time record information table as candidate accessible adjacent nodes of the head node, to form a candidate accessible adjacent node set;

[0060] S4033, if the tuple formed by a candidate accessible adjacent node and the corresponding access time in the candidate accessible adjacent node set belongs to H, delete the candidate accessible adjacent node, and use the adjacent nodes in the candidate accessible adjacent node set from which the candidate accessible adjacent node has been deleted as accessible adjacent nodes.

[0061] S404, obtaining the access weight w of any accessible adjacent node in the currently accessible node information set n =w f +t fn , w f is the access weight corresponding to the head node, tfn The time required for any aircraft to depart from the head node and reach any accessible adjacent node; execute S405.

[0062] S405, if the ID and node arrival time of a node in a certain node characteristic information in the current list of node information to be visited are respectively the same as the ID and access time of a node in a certain accessible node information in the current accessible node information set, execute S406, otherwise, that is, if the ID and node arrival time of a node in any node characteristic information in the current list of node information to be visited are different from the ID and access time of a node in any accessible node information in the current accessible node information set, execute S407.

[0063] S406, if the access weight in the certain node characteristic information is greater than the access weight of the certain accessible node information, delete the certain node characteristic information from the current list of node information to be accessed; if the access weight in the certain node characteristic information is less than the access weight of the certain accessible node information, delete the certain accessible node information from the current accessible node information set; execute S407.

[0064] For example, if the current accessible node information set includes node 2 and node 3, and node 3 exists in the current list of node information to be visited, if the access weight of node 3 in the current accessible node information set is greater than the access weight of node 3 in the current list of node information to be visited, then node 3 in the current accessible node information set will be deleted; if it is less than the access weight of node 3 in the current list of node information to be visited, then node 3 in the current list of node information to be visited will be deleted. This will minimize the gliding time of the generated gliding path.

[0065] S407, insert each accessible node information in the current accessible node information set and the corresponding access weight, the ID of the parent node and the access weight of the parent node as the corresponding node feature information into the current list of node information to be accessed, and execute S402.

[0066] In an embodiment of the present invention, the node feature information corresponding to each node in the currently accessible node information set can be inserted after the last node feature information in the current list of node information to be visited. If there are multiple node feature information that need to be inserted, they can be inserted in random order.

[0067] S408: Generate a path in reverse order according to the parent node of the target node in the current visited node information list as the taxiing path of any aircraft.

[0068] In an embodiment of the present invention, when generating the taxi path of an aircraft: (1) the impact caused by different starting times of the aircraft is taken into account to reduce conflicts between aircraft with different take-off and landing times; (2) the impact of the loss caused by the aircraft reaching the target node on the subsequent aircraft is ignored; (3) the impact of the conflict caused by the occurrence of each taxiway is added, that is, the access weight of each reachable node needs to be calculated.

[0069] S500, based on the current taxi path generation information, determine whether there is a path conflict between all aircrafts, if there is a path conflict, execute S600; if there is no path conflict, execute S700.

[0070] In the embodiment of the present invention, possible path conflicts include vertex conflict, following conflict and collision conflict. Among them: (1) Vertex conflict: When a pair of aircraft are planned to occupy the same vertex at the same time, a vertex conflict will occur between them. Formally, when aircraft i and aircraft j have a taxi sequence π respectively i =(a1, ..., a x ), π j =(a1, ..., a y ), i and j slide forward one position each, so that π i [x+1]=π j [y+1]andu i [x+1]=u j [y+1], π i and π j There is a vertex conflict between them. (2) Following conflict: When an aircraft plans to taxi to the node where another aircraft is currently located, a following conflict may occur because j may have to stay at the current location. Formally, aircraft i and j each have a taxi sequence π i and π j , so that π i [x+1]=π j [y]andu i [x+1]=u j [y+1]. (3) Collision conflict: When a pair of aircraft are scheduled to pass through the same edge in opposite directions at the same time, a collision conflict will occur between them. Formally, when aircraft i and j have a taxiing sequence π i and π j , i and j slide forward one position each, so that π i [x] = π j [y+1]、π i [x+1]=π j [y]andu i [x] = u j [y], π iand π j There are vertex conflicts.

[0071] Further, in S500, the determining whether there is a path conflict between all aircraft based on the current taxi path generation information specifically includes the following steps:

[0072] S501, obtaining the path conflict mark C corresponding to the a-th taxiing action of the d-th aircraft in the current unplanned flight sequence d a , the value of a ranges from 1 to g(d), where g(d) is the total number of taxiing actions of the d-th aircraft.

[0073] Among them, C d a The following conditions are met: If C d a =0, if (π d [a],u d [a])∈S o , C d a =1; where π d [a] represents the position of the d-th aircraft after performing the a-th taxiing action from the corresponding source node, u d [a] represents the time after the d-th aircraft starts to perform the a-th taxiing action from the corresponding source node, S o The node information table is formed by the node information occupied by other aircraft except the dth aircraft in the current unplanned flight sequence. o The node information table is formed by the node information occupied by other aircraft except the dth aircraft in the current unplanned flight sequence; S o ={s o j} j＝1……(w-1) ,s o j is the node information occupied by the jth aircraft among other aircraft, j ranges from 1 to (w-1), w is the number of aircraft in the current unplanned flight sequence, s o j =(R jp ) p＝1……h(j) , R jp For o j The pth node information in R jp =(Nr jp , tr jp ), Nr jp is the ID of the pth node occupied by the jth aircraft among other aircraft, tr jpis the time when the jth aircraft among other aircraft occupies the pth node, p ranges from 1 to h(j), and h(j) is the number of nodes occupied by the jth aircraft among other aircraft.

[0074] S502, if ∑ w d＝1 ∑ g(d) a＝1 C d a ≠0, it is determined that there is a path conflict between all aircraft. If ∑ w d＝1 ∑ g(d) a＝1 C d a =0, it is determined that there is no path conflict between all aircraft.

[0075] S600, update the current unplanned flight sequence and the current planned flight sequence, obtain the updated unplanned flight sequence and the updated planned flight sequence, and use the updated unplanned flight sequence and the updated planned flight sequence as the current unplanned flight sequence and the current planned flight sequence respectively, and execute S400.

[0076] S700, based on the current taxi path generation information, obtain the current average taxi time of all aircraft, and add the obtained current average taxi time of all aircraft to the current average taxi time list; set C=C+1, and execute S300; wherein, the initial value of the current average taxi time list is empty.

[0077] S800: Use the glide path generation information corresponding to the minimum average glide time in the current average glide time list as the target glide path generation information.

[0078] Furthermore, S500 also includes:

[0079] Obtain the number of conflict pairs with path conflicts and add them to the current conflict pair record table, and update the current weight of the current flight sequence selection method to be planned based on the current conflict pair record table, and execute S600; if there is no path conflict, execute S700; the initial value of the current conflict pair record table is empty.

[0080] In the embodiment of the present invention, a conflict pair is formed by any two aircraft having a path conflict, and the conflict pair may be determined based on the taxiing paths of the aircraft.

[0081] If there is a conflict in the aircraft path, it is necessary to start a repair algorithm to select some aircraft to form the flight sequence to be planned, and the remaining aircraft to form the planned flight sequence, so as to repair the aircraft conflict. Further, S600 may specifically include:

[0082] S601, setting the current unplanned flight sequence and the current planned flight sequence to empty sequences.

[0083] S602, select the selection method with the largest weight among the k methods for selecting the flight sequence to be planned as the current method for selecting the flight sequence to be planned and use the current method for selecting the flight sequence to be planned to select f aircraft from all aircraft to add to the current unplanned flight sequence, add the unselected aircraft to the current planned flight sequence, and execute S400; the initial value of the current weight of each method for selecting the flight sequence to be planned is 1 / k.

[0084] In the embodiment of the present invention, f is an integer greater than 1 and can be set based on actual needs.

[0085] In the embodiment of the present invention, if the weights of the flight sequences to be planned are the same, then one is randomly selected as the current method for selecting the flight sequence to be planned.

[0086] Further, in S500, the updated current weight of the current method for selecting the flight sequence to be planned satisfies the following conditions:

[0087] w r =γ×max{0,n - -n +}+(1-γ)×w r ;

[0088] Among them, w r is the current weight of the method for selecting the flight sequence to be planned, and γ is a preset coefficient used to control the sensitivity of the weight to successfully reducing the conflict pairs, which can be an empirical value. In an exemplary embodiment, γ can be set to 0.1. + is the number of the last path conflict pair in the current conflict pair record table, n - The current conflict pair record table is located at n + The number of conflicting paths before the replanning, that is, n- and n+ are the number of conflicting paths before and after the replanning. max{} means taking the maximum value.

[0089] Furthermore, in the embodiment of the present invention, three methods for selecting the flight sequence to be planned may be included, namely, a random selection method, a collision selection method and a failure selection method, wherein the random selection method includes the following steps:

[0090] Set the selection weights for all aircraft based on whether they are in conflict or not.

[0091] In the embodiment of the present invention, for an aircraft that has conflicted among the n aircraft, the selection weight may be set to 2 / n, otherwise, it may be set to 1 / n.

[0092] Randomly select f aircraft from all aircraft with set selection weights and add them to the unplanned flight sequence;

[0093] The collision selection method comprises the following steps:

[0094] Selecting one aircraft from the conflicting aircraft as a reference aircraft, and adding all aircraft that collide with the reference aircraft to the unplanned flight sequence;

[0095] If the number of aircraft in the unplanned flight sequence e<f, (fe) aircraft are randomly selected from the remaining aircraft to be added to the current unplanned flight sequence.

[0096] The failure selection method comprises the following steps:

[0097] Randomly select a node where an aircraft collision occurs from the nodes where the aircraft collision occurs as a reference node, and add the aircraft including the reference node in the taxiing path to the unplanned flight sequence;

[0098] If the number of aircraft in the unplanned flight sequence is e<f, (fe) aircraft are randomly selected from the remaining aircraft to be added to the unplanned flight sequence.

[0099] In the embodiment of the present invention, the flight sequence to be planned is continuously extracted through the flight sequence selection method to be planned, and its path is replanned until there is no conflict, and then the flight sequence selection algorithm to be planned is exited. In addition, the advantage of the flight sequence selection method to be planned provided by the present invention is that it can dynamically adjust the search strategy according to previous successful experience, so as to find a higher quality solution, and intelligently combine multiple selection methods to improve the search efficiency.

[0100] In the embodiment of the present invention, the T2 terminal of Beijing Capital Airport is taken as the research object. The present invention selects the taxiing data of 90 aircraft during the busy period from 6:00 to 11:00 on a certain day, and conducts case simulation analysis. The average speed of the aircraft when taxiing on the ground is usually between 10 and 30 knots, which is about 18 to 55 kilometers per hour (11 to 34 miles). Therefore, the present invention assumes that it taxis on the scene at a speed of 18.52 kilometers per hour. According to the guidance documents of the FAA, the recommended safety interval between aircraft on the taxiway is usually at the minute level. For this reason, the safety interval set by the present invention is 1 minute. The terminal has a total of 66 parking spaces and 9 arrival and departure nodes, as well as 112 intersection nodes. The setting of multiple arrival and departure nodes can better simulate the multi-runway operation in the real airport scene.

[0101] In order to comprehensively evaluate the performance of the airport surface taxiway path generation method proposed in the embodiment of the present invention in taxiway path planning, the present invention conducted comparative experiments with the improved ant colony algorithm, the improved bee colony algorithm and the A* algorithm. Figure 3a to Figure 3d The results of the average taxiing time of aircraft obtained by four optimization algorithms when planning 10-40 aircraft respectively are shown, among which MAPF-ATP-LNS represents the airport surface taxiing path generation method of the present invention. Figure 3a to Figure 3d It can be seen that when processing the taxiway path optimization of 10 to 40 aircraft, the taxiing efficiency of the method of the present invention is improved by at least five times compared with the A* algorithm, and the performance is improved by more than 40% compared with the improved ant colony algorithm. At the same time, compared with the improved bee colony algorithm, the method of the present invention also shows better performance improvement. In addition, the improved bee colony algorithm has a stronger optimization ability than the improved ant colony algorithm and the A* algorithm in reducing the average taxiing time of aircraft.

[0102] In order to explore the influence of selecting different repair sorties under different planned flight sorties, that is, the f value in the selection of the flight sequence to be planned, on the average taxiing time of aircraft, the average number of repairs and the computer execution time in the conflict repair process of the method provided by the present invention, this paper selects 40-90 aircraft and 10-30 repair sorties for experiments, and the experimental results are as follows: Figures 4a to 4c shown.

[0103] Depend on Figure 4a It can be seen that in the experiment of average taxiing time of 40 to 90 flights, when the number of planned aircraft is constant, as the number of selected aircraft repairs increases, the average taxiing time shows an overall downward trend, indicating that the more flights are replanned, the more conducive it is to find a better taxiing path. Figure 4bIt can be seen that in the experiment of the average number of repairs from 40 to 90 sorties, as the number of planned aircraft increases, the number of repairs also gradually increases. In addition, when the number of planned aircraft is fixed, as the number of selected aircraft repairs increases, the average number of repairs required for aircraft conflicts also increases; Figure 4c It can be seen that as the number of aircraft sorties increases, the overall computer execution time shows an increasing trend. When the number of planned aircraft is constant, the computer execution time also increases with the increase in the number of selected aircraft repairs. It is worth noting that when the number of planned aircraft reaches 90 sorties, the execution time corresponding to the number of repairs for different aircraft has increased significantly.

[0104] In summary, the present invention aims at solving the problem that under realistic conditions, flights are easily disturbed by various factors and need to rearrange the taxiing path, and the conflict between runway resources and growing demand, and the need to reduce fuel costs. A method for generating taxiing paths on the airport surface is proposed. Under the condition of ensuring zero conflict, the method uses the average taxiing time as the objective function, selects the optimal planning sequence in a local random manner, and plans the path for each aircraft through an improved path generation algorithm, and finally repairs possible conflicts through a large-area search. Experiments show that the method provided by the present invention can obtain a higher-quality solution in a shorter time compared to the improved ant colony, improved bee colony and A* algorithms, can help deal with the need to re-plan the path of flights in realistic situations due to various emergencies, and can increase taxiing efficiency, provide services for the growing aviation needs, and reduce fuel consumption.

[0105] An embodiment of the present invention also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are configured to execute the method described in the embodiment of the present invention.

[0106] The embodiment of the present invention further provides a computer-readable storage medium storing computer-executable instructions, wherein the computer instructions are used to execute the method described in the embodiment of the present invention.

[0107] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0108] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for generating a taxiing path on an airport surface, characterized in that: The method comprises the following steps: S100, constructing a taxi network quadruple (G, q, o, p) based on the airport map and the information of arriving and departing aircraft within a set time period, wherein G is the airport taxiway graph structure, G = (V, E, W), V is a node set formed by nodes on the airport surface, E is an edge set, the edges in E represent the taxiways connecting the nodes, W is a taxi time set formed by the taxi times between adjacent nodes, q is a source node set formed by the source nodes of the aircraft within the set time period, o is a target node set formed by the target nodes of the aircraft within the set time period, and p is a departure time set formed by the timestamps of the aircraft departing from the source nodes within the set time period, and the nodes represent the intersections between the parking stands and taxiways on the airport surface or the intersections between the runways and taxiways; S200, setting the iteration counter C=1; S300, if C≤C0, add all aircraft within the set time period to the current unplanned flight sequence and set the current planned flight sequence to empty, and execute S400; otherwise, execute S800; C0 is the maximum number of iterations; S400, grouping the aircraft in the current unplanned flight sequence to obtain a plurality of to-be-planned groups, and randomly initializing the path planning order of the aircraft in each to-be-planned group, and generating a taxi path for each aircraft in each to-be-planned group based on a preset objective function and preset constraints, to obtain current taxi path generation information; S500, based on the current taxi path generation information, determining whether there is a path conflict between all aircrafts, if there is a path conflict, executing S600; if there is no path conflict, executing S700; S600, updating the current unplanned flight sequence and the current planned flight sequence, obtaining an updated unplanned flight sequence and an updated planned flight sequence, and using the updated unplanned flight sequence and the updated planned flight sequence as the current unplanned flight sequence and the current planned flight sequence, respectively, and executing S400; S700, based on the current taxi path generation information, obtain the current average taxi time of all aircraft, and add the obtained current average taxi time of all aircraft to the current average taxi time list; set C=C+1, and execute S300; wherein the initial value of the current average taxi time list is empty; S800: Use the glide path generation information corresponding to the minimum average glide time in the current average glide time list as the target glide path generation information.

2. The method according to claim 1, characterized in that The objective function satisfies the following conditions: T=min((∑ n b＝1 (T b end -T b begin )) / n); T is the objective function, T b end is the time when the bth aircraft arrives at the parking stand or runway, T b begin It is the time when the bth aircraft leaves the parking stand or runway. The value of b ranges from 1 to n, and n is the total number of aircraft take-offs and landings in the set time period.

3. The method according to claim 1, characterized in that The preset constraints meet the following conditions: Where x represents the xth taxiing action of the ith aircraft in the unplanned flight sequence, π i [x] represents the position of the i-th aircraft after performing the x-th taxiing action from the corresponding source node, u i [x] represents the time after the i-th aircraft performs the x-th taxiing action from the corresponding source node, the value of i ranges from 1 to z, z is the number of aircraft in the unplanned flight sequence, and H is the node information table formed by the node information occupied by the aircraft in the planned flight sequence.

4. The method according to claim 1, characterized in that: In S500, the process of determining whether there is a path conflict between all aircraft based on the current taxi path generation information specifically includes the following steps: S501, obtaining the path conflict mark C corresponding to the a-th taxiing action of the d-th aircraft in the current unplanned flight sequence d a , a ranges from 1 to g(d), g(d) is the total number of taxiing actions of the dth aircraft, where C d a The following conditions are met: If C d a =0, if (π d [a],u d [a])∈S o , C d a =1; where π d [a] represents the position of the d-th aircraft after performing the a-th taxiing action from the corresponding source node, u d [a] represents the time after the d-th aircraft starts to perform the a-th taxiing action from the corresponding source node, S o A node information table formed by the node information occupied by other aircraft except the dth aircraft in the current unplanned flight sequence; S502, if ∑ w d＝1 ∑ g(d) a＝1 C d a ≠0, it is determined that there is a path conflict between all aircraft. If ∑ w d＝1 ∑ g(d) a＝1 C d a =0, it is determined that there is no path conflict between all aircraft.

5. The method according to claim 1, characterized in that In S400, the taxiing path of any aircraft is obtained through the following steps: S401, adding node characteristic information corresponding to the source node of any aircraft to the corresponding list of node information to be visited; the node characteristic information includes the ID of the node, the access time of the node, the access weight of the node, the ID of the parent node of the node, and the access weight of the parent node; S402, determine whether there is node feature information to be visited in the current list of node information to be visited, if not, execute S200, if yes, take out the node feature information of the head node in the current list of node information to be visited and store it in the visited node information list corresponding to any aircraft, and determine whether the head node is the target node of any aircraft, if yes, execute S408, if not, execute S403; S403, based on the current node access time record information table and preset constraints, determine the adjacent nodes accessible to the head node and the access time to form a current accessible node information set; The currently accessible node information set includes multiple accessible node information, each accessible node information includes the ID of the corresponding accessible adjacent node and the corresponding access time, and the current node access time record information table includes the ID of each node in G and the corresponding access time information; Execute S404; S404, obtaining the access weight w of any accessible adjacent node in the currently accessible node information set n =w f +t fn , w f is the access weight corresponding to the head node, t fn The time required for any aircraft to depart from the head node and reach any accessible adjacent node; executing S405; S405, if the ID and the node arrival time of a node in the node characteristic information in the current to-be-visited node information list are respectively the same as the ID and the access time of a node in the accessible node information in the current accessible node information set, execute S406, otherwise, execute S407; S406, if the access weight in the certain node feature information is greater than the access weight of the certain accessible node information, delete the certain node feature information from the current list of node information to be accessed; if the access weight in the certain node feature information is less than the access weight of the certain accessible node information, delete the certain accessible node information from the current accessible node information set; Execute S407; S407, insert each accessible node information and the corresponding access weight, the ID of the parent node and the access weight of the parent node in the current accessible node information set as the corresponding node feature information into the current list of to-be-accessed node information, and execute S402; S408: Generate a path in reverse order according to the parent node of the target node in the current visited node information list as the taxiing path of any aircraft.

6. The method according to claim 1, characterized in that The S500 also includes: The number of conflict pairs with path conflicts is obtained and added to the current conflict pair record table, and the current weight of the current flight sequence selection method to be planned is updated based on the current conflict pair record table, and S600 is executed; if there is no path conflict, S700 is executed; the initial value of the current conflict pair record table is empty; S600 specifically includes: S601, setting the current unplanned flight sequence and the current planned flight sequence to empty sequences; S602, select the selection method with the largest weight among the k methods for selecting the flight sequence to be planned as the current method for selecting the flight sequence to be planned and use the current method for selecting the flight sequence to be planned to select f aircraft from all aircraft to add to the current unplanned flight sequence, add the unselected aircraft to the current planned flight sequence, and execute S400; the initial value of the current weight of each method for selecting the flight sequence to be planned is 1 / k.

7. The method according to claim 6, characterized in that In S500, the updated current weight of the current method for selecting the flight sequence to be planned satisfies the following conditions: w r =γ×max{0,n - -n + }+(1-γ)×w r ; Among them, w r is the current weight of the method for selecting the flight sequence to be planned, γ is the preset coefficient, n + is the number of the last path conflict pair in the current conflict pair record table, n - The current conflict pair record table is located at n + The number of conflicting paths in the previous one. max{} indicates the maximum value.

8. The method according to claim 5, characterized in that There are three methods for selecting flight sequences to be planned, namely, random selection method, collision selection method and failure selection method. The random selection method includes the following steps: Set selection weights for all aircraft based on whether they are in conflict; Randomly select f aircraft from all aircraft with set selection weights and add them to the unplanned flight sequence; The collision selection method comprises the following steps: Selecting one aircraft from the conflicting aircraft as a reference aircraft, and adding all aircraft that collide with the reference aircraft to the unplanned flight sequence; If the number of aircraft in the unplanned flight sequence is e<f, randomly select (fe) aircraft from the remaining aircraft and add them to the current unplanned flight sequence; The failure selection method comprises the following steps: Randomly select a node where an aircraft collision occurs from the nodes where the aircraft collision occurs as a reference node, and add the aircraft including the reference node in the taxiing path to the unplanned flight sequence; If the number of aircraft in the unplanned flight sequence e<f, (fe) aircraft are randomly selected from the remaining aircraft to be added to the unplanned flight sequence.

9. An electronic device, characterized in that: including a processor and a memory; The processor is used to execute the steps of the method according to any one of claims 1 to 8 by calling the program or instruction stored in the memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a program or an instruction, wherein the program or the instruction enables a computer to execute the steps of the method according to any one of claims 1 to 8.

Images