A method for generating UAV mission formations for urban logistics based on spatiotemporal paths

Abstract

The present invention discloses a method for generating a city logistics UAV task formation based on a time-space path, comprising the following steps: (1) searching for potential logistics task cooperation objects based on flight information, aircraft models and other information, constructing a candidate UAV sequence set, and giving priority to matching according to the length of matching waiting time; (2) for each travel object in the candidate UAV sequence set, by sharing its own flight path and departure time and obtaining the flight path and departure time of other logistics UAVs, predicting based on the average UAV flight speed, judging whether there are overlapping time-space paths, planning and determining the flight path of the task formation; (3) determining the queue order of the UAV task formation according to the arrival time, and planning and generating the UAV task formation for executing logistics delivery in advance. The present invention is simple and fast in calculation, and by generating a UAV task formation, a safe and efficient logistics UAV delivery method can be achieved at a relatively low cost.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicles (UAVs) and logistics, and in particular to a method for generating a task formation of urban logistics UAVs based on a space-time path. Background Art

[0002] With the development of modern computer technology and the advancement of Internet technology, logistics drone delivery driving technology has become a reality. Compared with traditional delivery methods, drone delivery can avoid traffic jams and ground obstacles and deliver goods at a faster speed. This can play an important role in emergency situations or items that need to be delivered quickly. On the other hand, drone delivery can reduce labor and transportation costs because they do not require drivers or fuel costs. The carbon emissions generated by drones are also much less than fleet delivery, which helps to reduce the impact of logistics activities on the environment and promote sustainable development. In addition, by using sensors and monitoring technology, drone delivery can monitor the status and location of goods in real time. The delivery party can deal with any problems in a timely manner to ensure the safety and integrity of the delivery.

[0003] However, drone delivery also faces a series of problems. How to plan the flight paths of multiple logistics drones to minimize the risk of collision while maintaining an efficient delivery speed is the key to the widespread application of this delivery model. Existing drone delivery is often a single drone delivery model. A single drone may find it difficult to cope with emergencies, while a formation system can better deal with problems and risks. At the same time, in high-traffic cities or densely populated areas, a single drone may face air traffic congestion and conflicts, while a formation system can better plan flight paths. Compared with single drone transportation, the key issues that need to be solved in drone mission formation transportation are the generation of formation members and the decision-making of formation transportation paths. In order to realize drone mission formation transportation, it is necessary to construct a formation generation decision method that can realize drone delivery information interaction and common transportation path identification. Summary of the invention

[0004] Purpose of the invention: In view of the shortcomings of the existing technology, this patent proposes a method for generating urban logistics drone task formations based on space-time paths. By mining the space-time path information of logistics drone flights, drone task formations are matched and generated in advance to achieve self-organized flight control. By applying this patent, safe and efficient drone logistics distribution can be achieved at a lower cost, avoiding the chaos of drone distribution.

[0005] Technical solution: A method for generating a task formation of urban logistics drones based on time-space paths, comprising the following steps:

[0006] (1) Search for potential logistics task partners based on flight information, aircraft models, and other information, and construct a candidate UAV sequence set S1;

[0007] (2) For each travel object in the candidate UAV sequence set S1, by sharing its own flight path and departure time and obtaining the flight paths and departure times of other logistics UAVs, a prediction is made based on the average UAV flight speed, and the flight path of the task formation is planned and determined;

[0008] (3) Determine the queue order of the drone task formation according to the arrival time, and plan and generate the drone task formation for logistics delivery in advance.

[0009] Furthermore, the potential task partners in step 1 need to meet the constraints: the UAVs participate in the flight formation generation and have a certain matching waiting time T wait ≥3s. Logistics drones with long matching waiting time will be given priority.

[0010] Furthermore, the step 2 comprises the following steps:

[0011] 2.1. Determine the minimum number threshold Num for task formation generation based on the number of logistics drone delivery requirements j , when the number of matching task formations is greater than the threshold, the task formation generation can be completed:

[0012]

[0013] Among them, A j is the minimum number of drones to match unmanned mission formation j, A t is the total number of unmatched drones at time t. j ≥2, that is, in any case, the number of drone task formation matches must reach at least 2 to generate a task formation.

[0014] 2.2, predict the time and space paths of each transport object;

[0015] For each travel object in the candidate drone sequence set S1, by sharing and obtaining the start and end paths and departure times of the logistics drones, combined with the shortest path planning of the drone route network map, the historical mean, time series, neural network or decision tree algorithm is used to predict the drone flight speed, thereby predicting the spatiotemporal paths of each logistics drone that needs to be matched:

[0016] Except_lon it =Where it _lon

[0017] Except_lat it =Where it _lat

[0018] Except_h it =Where it _h

[0019] Among them, Except_lon it 、Except_lat it 、Except_h it Where are the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the position of drone i at time t. it _lon、Where it _lat、Where it _h are the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the location at time t using the average UAV flight speed, based on the starting and ending points and departure time of UAV i, the shortest path planned using the UAV route network map.

[0020] 2.3. For the logistics drones with the same spatiotemporal paths in the candidate drone sequence set S1, they are matched according to the spatiotemporal path overlap of the transport objects:

[0021]

[0022] Joint ij =Route i ∩Route j

[0023] i, j∈Queue i , Joint ij ≥c*(S i +S j )

[0024] Among them, Dis i 、Dis j Respectively represent the distances between the transportation starting point and the transportation end point of logistics drone i and logistics drone j; A ix , A iy , A iz They represent the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the transport starting point of logistics drone i; B ix , B iy , B iz They represent the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the transport destination of logistics drone i respectively; ij represents the spatiotemporal overlapping paths of logistics drone i and logistics drone j; Route i 、Route j Respectively represent the space-time paths of logistics drone i and logistics drone j; Queue irepresents the task formation to which UAV i belongs; c represents the spatiotemporal overlapping path coefficient, and its value range is (0,0.5].

[0025] 2.4. For the logistics drones that do not have the same space-time path in the candidate drone sequence set S1, they are matched according to the proximity between the terminal path of the required logistics drone and the existing task formation:

[0026]

[0027] i, j∈Queue i , D ijt <=pri

[0028] Where D ijt Indicates the distance between the planned path of logistics drone i and the transportation destination of logistics drone j at time t; Except_lon it 、Except_lat it 、Except_h it are the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the position of drone i at time t; B ix , B iy , B iz A represents the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the transport destination of logistics drone i respectively; it represents the departure time of logistics drone i, B jt Table Expected arrival time of logistics drone j: Queue i represents the task formation to which the logistics UAV i belongs, pri represents the distance threshold for evaluating the path proximity, and its value range is (0,200km].

[0029] Beneficial effects: The present invention requires drones for logistics distribution, and by sharing information such as transportation paths and departure times between logistics drones, combined with the path prediction of the drone route network map, by calculating the overlap of space-time paths or the degree of space-time interweaving between the transportation paths of logistics drones and their destinations, logistics drones with similar space-time path planning are generated into task formations, providing a decision-making solution for logistics drone task formation delivery. The present invention can effectively identify the overlapping space-time transportation paths of each logistics drone, reduce the time loss of each logistics drone in the task formation transportation, and realize the self-organized delivery of logistics drones through formation transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 The present invention is a flow chart of a method for generating a city logistics UAV task formation based on a spatiotemporal path. DETAILED DESCRIPTION

[0031] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0032] Figure 1 The flowchart of the present invention for making a decision on task formation generation based on logistics drone information is shown. The present invention determines the members of the task formation based on the logistics drone transportation plan information, and mainly includes the following steps:

[0033] Step (1) searches for potential task partners based on the transportation plan information and constructs a candidate drone sequence set S1. Assume that the number of logistics drones is K, and the departure information of the kth logistics drone includes eight fields: logistics drone number, departure time, arrival time, departure longitude, departure latitude, arrival longitude, arrival latitude, and matching waiting time. The departure longitude and departure latitude constitute the transportation starting point, and the arrival latitude and arrival longitude constitute the transportation destination.

[0034] The transport objects that can participate in the formation of the task force are determined based on the matching waiting time of the logistics drones. The logistics drones participating in the transportation need to meet the constraints: the drones participate in the formation of the flight and have a certain matching waiting time. The logistics drones with long matching waiting time will be given priority.

[0035] Step (2) is to share its own transportation path and departure time and obtain the transportation paths and transportation times of other logistics drones, make predictions based on the average drone flight speed, determine whether there are overlapping space-time paths, and plan and determine the following path.

[0036] The specific steps include:

[0037] (21) Determine the minimum number threshold Num for generating a task formation based on the number of logistics drone delivery requirements j , when the number of matching task formations is greater than the threshold, the task formation generation can be completed:

[0038]

[0039] Among them, A j is the minimum number of drones to match unmanned mission formation j, A t is the total number of unmatched drones at time t. j ≥2, that is, in any case, the number of drone task formation matches must reach at least 2 to generate a task formation.

[0040] (22) Predict the time and space paths of each transport object;

[0041] For each travel object in the candidate drone sequence set S1, by sharing and obtaining the start and end paths and departure times of other logistics drones, combined with the shortest path planning of the drone route network map, the historical mean, time series, neural network or decision tree algorithm is used to predict the flight speed of the logistics drone, thereby predicting the spatiotemporal paths of each logistics drone that needs to be matched:

[0042] Except_lon it =Where it _lon

[0043] Except_lat it =Where it _lat

[0044] Except_h it =Where it _h

[0045] Among them, Except_lon it 、Except_lat it 、Except_h it Where are the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the position of drone i at time t. it _lon、Where it _lat、Where it _h are the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the location at time t using the average UAV flight speed, based on the starting and ending points and departure time of UAV i, the shortest path planned using the UAV route network map.

[0046] The total number of logistics drones is N, and the transportation starting point of the i-th logistics drone is A i , departure time is A it The transport destination is B i , based on the shortest path in the route network, the predicted arrival time is B it At time t, the predicted position based on the shortest path of the route network is S it The transportation starting point of the jth logistics drone is A j , departure time is B jt The transport destination is B j , based on the shortest path in the route network, the predicted arrival time is B jt At time t, the predicted position based on the shortest path of the route network is S jt If the planned paths of logistics drone i and logistics drone j are at the same or close locations at time t, it is considered that there are spatiotemporal overlapping paths between logistics drone i and logistics drone j. Otherwise, it is considered that there are no spatiotemporal overlapping paths.

[0047] (23) For the logistics drones with overlapping spatiotemporal paths in the candidate drone sequence set S1, matching is performed based on the spatiotemporal path overlap of the transport object:

[0048]

[0049] Joint ij =Route i ∩Route j

[0050] i, j∈Queue i , Joint ij ≥c*(S i +S j )

[0051] Among them, Dis i A represents the distance between the transportation starting point and the transportation end point of logistics drone i; ix , A iy , A iz They represent the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the transport starting point of logistics drone i; B ix , B iy , B iz They represent the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the transport destination of logistics drone i respectively; ij represents the spatiotemporal overlapping paths of logistics drone i and logistics drone j; Route i 、Route j Respectively represent the space-time paths of logistics drone i and logistics drone j; Queue i represents the task formation to which UAV i belongs; c represents the spatiotemporal overlapping path coefficient, and its value range is (0,0.5].

[0052] (24) For the logistics drones that do not have the same spatiotemporal path in the candidate drone sequence set S1, the destination path of the logistics drones that need to form a formation is matched according to the proximity between the destination path of the logistics drones that need to form a formation and the existing task formation:

[0053]

[0054] i, j∈Queue i , D ijt <=pri

[0055] Where D ijt Indicates the distance between the planned path of logistics drone i and the transportation destination of logistics drone j at time t; Except_lon it 、Except_lat it、Except_h it are the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the position of drone i at time t; B ix , B iy , B iz A represents the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the transport destination of logistics drone i respectively; it represents the departure time of logistics drone i, B jt Table Expected arrival time of logistics drone j: Queue i represents the task formation to which the logistics UAV i belongs, pri represents the distance threshold for evaluating the path proximity, and its value range is (0,200km].

[0056] Step (3), determine the first aircraft of the task formation and the queue sequence according to the aircraft type and arrival time, and plan and generate the task formation in advance.

[0057] Assume that there are N logistics drones that form a task formation for formation transportation. The one with the highest departure time is the lead drone. The logistics drones that arrive later are arranged in front and back according to the same size and drive adjacent to each other, with large drones in front and small drones in the back, and are inserted into the task formation. All logistics drones rotate the lead drone regularly and exchange positions with logistics drones of similar models. If there are no similar models, they will be moved to the end of the task formation.

[0058] The technical solution of the present invention is further illustrated by using specific embodiments below.

[0059] The research period T is set to one day (i.e. 24 hours), and there are 8 logistics drones participating in the task formation generation decision. The transportation data of each logistics drone is shown in Table 1.

[0060] Table 1 Logistics drone transportation data on September 20, 2022

[0061]

[0062]

[0063] 1. Search for potential task partners based on transportation information and determine the matching priority of logistics drones to build a candidate drone sequence set S1. The logistics drones numbered 1-8 all meet the requirements for drones to participate in flight formation generation and have a certain matching waiting time. Therefore, they can all be matched for transportation. The matching priority is determined based on the length of the matching waiting time. That is, the matching order is 8-7-5-6-3-2-4-1.

[0064] (2) For each travel object in the candidate drone sequence set S1, by sharing its own transportation path and obtaining the transportation paths and transportation times of other logistics drones, a prediction is made based on the average flight speed of the logistics drones to determine whether there are overlapping space-time paths, and the flight path of the task formation is planned and determined.

[0065] (21) Determine the minimum number threshold Num for generating a task formation based on the number of logistics drone transportation requirements j ,When the number of matching task formations is greater than the threshold, the task formation generation can be completed.

[0066] According to the minimum number formula and constraints of the mission formation, it is calculated that the minimum number of drones transported by this mission formation is 4.

[0067] (22) Predict the spatiotemporal path of each transport object.

[0068] Based on the path planning of the drone route network map, combined with the start and end point information and departure time of each logistics drone, the historical mean algorithm is used to predict the speed, so as to calculate the spatiotemporal path of each logistics drone, and compare them to determine whether there are logistics drones with overlapping spatiotemporal paths. Based on the matching order, the logistics drones with overlapping paths of each transport logistics drone are matched with the logistics drones with overlapping spatiotemporal paths as numbered 8, 6, 3, 2, and 1. The logistics drones numbered 4 and 5 do not have spatiotemporal overlapping paths with any logistics drones.

[0069] Table 2: Time-space path matching of logistics drone transportation on September 20, 2022

[0070]

[0071] (23) For the logistics drones with overlapping spatiotemporal paths in the candidate drone sequence set S1, they are matched according to the overlapping of the spatiotemporal paths of transportation.

[0072] According to the above formula and constraints, task formation is generated for logistics UAVs with overlapping space-time paths. The results are shown in Table 3.

[0073] Table 3 Task formation matching of logistics drones with overlapping paths on September 20, 2022

[0074]

[0075] (24) For the logistics drones in the candidate drone sequence set S1 that do not have the same spatiotemporal path, they are matched according to the proximity between the terminal path of the required logistics drone and the existing task formation.

[0076] According to the logistics drone matching priority, formula and constraints, the logistics drones that do not meet the overlapping path constraints are matched. After meeting the constraints of the minimum number of task formations and matching priority, the resulting team composition is shown in Table 4.

[0077] Table 4 Formation matching decision results for the formation transport mission on September 20, 2022

[0078]

[0079] 3. Determine the first aircraft and queue order of the task formation according to the aircraft type and arrival time, plan and generate the task formation in advance. After the task formation is generated, the first aircraft to arrive is the first aircraft, aircraft of the same size travel next to each other, and large aircraft are in front and small aircraft are in the back. The queue order of the logistics drone task formation generated is 4-3-1-6 and 8-2-5.

[0080] The spatiotemporal path information of logistics drones can be effectively used as decision-making information for optimizing transportation methods. Especially for situations where the spatiotemporal paths of logistics drones overlap or the destinations are intertwined in time and space, the mining and analysis of logistics drone transportation information data can help provide decision-making solutions for logistics drone task formation delivery.

[0081] The above content is only a further detailed description of the present invention through a specific embodiment, and the present invention cannot be considered to be limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A method for generating urban logistics UAV task formation based on spatiotemporal paths, characterized in that: The method comprises the following steps: (1) Search for potential logistics mission partners based on flight information and aircraft model information, and construct a candidate UAV sequence set S1 for subsequent logistics formation; (2) For each travel object in the candidate UAV sequence set S1, by sharing its own flight path and departure time and obtaining the flight paths and departure times of other logistics UAVs, the average flight speed of the potential logistics task cooperation formation is predicted, and the flight path of the task formation is planned and determined, including the following steps: 2.

1. Determine the minimum number threshold A for generating a drone task formation based on the number of logistics drone transportation requirements j , when the number of drones matching the task formation is greater than the threshold, the task formation generation can be completed: Among them, A j is the minimum number of logistics drones to match unmanned mission formation j, A t is the total number of logistics drones that are not successfully matched in the candidate drone sequence set S1 at time t, satisfying the constraint condition A j ≥2, that is, in any case, the number of logistics drone task formation matches must reach at least 2 to generate a task formation; 2.2, predict the time and space paths of each transport object; By sharing and obtaining the start-end paths and departure times of other logistics drones, combined with the shortest path planning of the logistics drone route network map, the average flight speed of the potential logistics task cooperation formation is used to predict the spatiotemporal path of each drone that needs to be matched: Except_lon it =Where it _lon Except_lat it =Where it _lat Except_h it =Where it _h Among them, Except_lon it 、Except_lat it 、Except_h it are the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the location of logistics drone i at time t, respectively. it _lon、Where it _lat、Where it _h are the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the location at time t, respectively, of the shortest path planned by the logistics drone route network map using the potential logistics task cooperation formation according to the starting and ending points and departure time of the logistics drone i; 2.

3. For the logistics drones with the same spatiotemporal paths in the candidate drone sequence set S1, they are matched according to the spatiotemporal path overlap of the transport objects: Joint ij =Route i ∩Route j i、j∈Queue i ,Joint ij ≥c*(S i +S j ) Among them, Dis i A represents the distance between the transportation starting point and the transportation end point of logistics drone i; ix , A iy , A iz They represent the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the transport starting point of logistics drone i; B ix , B iy , B iz They represent the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the transport destination of logistics drone i respectively; ij represents the spatiotemporal overlapping paths of logistics drone i and logistics drone j; Route i 、Route j Respectively represent the space-time paths of logistics drone i and logistics drone j; Queue i represents the task formation to which UAV i belongs; c represents the spatiotemporal overlapping path coefficient, and its value range is (0, 0.5]; 2.

4. For the logistics drones that do not have the same space-time path in the candidate drone sequence set S1, they are matched according to the proximity between the destination path of the drone to be transported and the existing task formation: i、j∈Queue i ,D ijt <=pri Where D ijt Indicates the distance between the planned path of logistics drone i and the transportation destination of logistics drone j at time t; Except_lon it 、Except_lat it 、Except_h it are the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the location of logistics drone i at time t; B ix , B iy , B iz A represents the longitude x-coordinate, latitude y-coordinate and altitude z-coordinate of the transport destination of logistics drone i respectively; it represents the departure time of logistics drone i, B jt Table Expected arrival time of logistics drone j: Queue i represents the task formation to which the logistics UAV i belongs, pri represents the distance threshold for evaluating the path proximity, and its value range is (0, 200km]; (3) Determine the queue order of the logistics drone task formation according to the arrival time, and plan and generate the logistics drone task formation for logistics delivery in advance.

2. The method for generating urban logistics UAV task formation based on time-space path according to claim 1 is characterized in that: The potential logistics task cooperation object in step 1 needs to meet the constraint conditions: the logistics drone participates in the flight formation generation and meets the matching waiting time T wait ≥3s, logistics drones with long matching waiting time will be given priority.

3. The method for generating urban logistics UAV task formation based on time-space path according to claim 1 is characterized in that: The matching algorithms used in the search for potential logistics task partners in step 1 include: first-come-first-served and maximum-weight matching algorithms.

4. The method for generating urban logistics UAV task formation based on time-space path according to claim 1 is characterized in that: The criteria for determining the first aircraft of the task formation and the queue sorting in step 3 are: the first aircraft is the one with priority in arrival time, and aircraft of the same size try to travel next to each other in a front-to-back sorting order. All logistics drones rotate the first aircraft regularly and exchange positions with similar models. If there are no similar models, they will be moved to the end of the task formation.

Images