A logistics drone task allocation method with priority division

Abstract

The present invention discloses a logistics drone task allocation method with priority division, comprising the following steps: S1, based on the geographical environment, low-altitude airspace structure, and logistics drone performance, extract the three-dimensional operation space information of the logistics drone and obtain the performance parameters of the logistics drone; S2, combining the logistics company and customer needs, obtain the transportation type, flow, flow direction, and flow time data of the logistics drone, and construct the logistics drone operation scene; S3, combining the task priority and the fairness of the logistics company, establish the logistics drone service satisfaction index; S4, with the minimum flight impedance and the maximum service satisfaction as the goal, construct a logistics drone task allocation model; S5, use the NSGA-II algorithm to solve the model, and output the optimal solution for logistics drone resource allocation. The present invention adopts the above-mentioned logistics drone task allocation method with priority division, which can realize the collaborative task allocation of multiple logistics drones, which is helpful to improve the operation efficiency and quality of logistics drones.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle task allocation, and in particular to a logistics unmanned aerial vehicle task allocation method with priority division. Background Art

[0002] The development of e-commerce has intensified the demand for logistics and distribution. Drones have gradually emerged in the field of logistics and distribution with their advantages of strong flexibility and high degree of freedom, which can effectively alleviate traffic congestion and emission pollution. However, drone traffic management is still in its early stages, and there is no unified standard for logistics drone transportation mode. If logistics drones implement point-to-point flights according to flight needs, as the density of drones in the airspace continues to grow, due to the lack of standardized management and scheduling, the overall operation order of drones is unbalanced, which brings great risks to the air and the ground. Therefore, establishing an effective logistics drone task allocation mechanism is key to improving operational safety and efficiency. Existing studies mostly regard the logistics drone task allocation problem as a free airspace path planning problem, and rarely consider the guidance and constraints of the route network on drones, ignoring the attributes and priorities of the delivered goods. Summary of the invention

[0003] The purpose of the present invention is to provide a logistics drone task allocation method with priority division, which can realize the collaborative task allocation of multiple logistics drones, help improve the operating efficiency and quality of logistics drones, and standardize the orderly operation of large-scale logistics drones.

[0004] To achieve the above object, the present invention provides a logistics drone task allocation method with priority division, comprising the following steps:

[0005] S1. Based on the natural geographical environment, low-altitude airspace structure, and logistics UAV performance, extract the three-dimensional operation space information of the logistics UAV and obtain the performance parameters of the logistics UAV;

[0006] S2. Combine the needs of logistics companies and customers to obtain the transportation type, flow, direction, and time data of logistics drones, and build logistics drone operation scenarios;

[0007] S3. Combine task priority with fairness of logistics companies to establish a logistics drone service satisfaction index;

[0008] S4. Construct a logistics UAV task allocation model with the goal of minimizing flight impedance and maximizing service satisfaction;

[0009] S5. Use the NSGA-II algorithm to solve the model and output the optimal solution for logistics drone resource allocation.

[0010] Preferably, in step S1, the three-dimensional operating space information of the logistics UAV is extracted through the geographic information system, including route network boundary data, route data and node data, to obtain the performance parameters of the logistics UAV, including maximum load, full load range, empty load range and full charge capacity.

[0011] Preferably, in step S2, based on the historical delivery data of the logistics company, the express delivery type, flow rate, flow direction, and flow time data of densely populated areas are extracted to obtain the express delivery OD distribution matrix;

[0012] Logistics drones fly along the route network, which can be viewed as a graph G = (N, L), where N is the set of all nodes in the network, N = R ∪ S ∪ K, R is the set of logistics warehouses, S is the set of demand points, K is the set of turning points in the route network, and L is the set of flight segments between two nodes;

[0013] Logistics drones follow the "unified dispatch" management model. Various logistics companies deliver goods to logistics warehouses, and third-party agencies provide drone delivery services in a unified manner.

[0014] Preferably, in step S3, the service satisfaction is quantitatively expressed by the membership function of the fuzzy delivery time, and the satisfaction is in a superlinear power function relationship with the actual delivery time of the goods. F is the set of logistics companies involved, U is the set of logistics drones assigned by the third-party company, and K u The number of times the drone performs delivery tasks;

[0015] The expected delivery time window of the drone delivery company f(f∈F) from warehouse r(r∈R) to the demand point s(s∈S) is [E frs ,L frs ], the maximum tolerable time window is [SE frs ,SL frs ], the actual delivery time is Then the membership function of the fuzzy delivery time of service satisfaction is:

[0016]

[0017] Among them, A ukfrs It means that the UAV u(u∈U) is executing the kth(k∈K u ) is the satisfaction of the delivery company f from warehouse r to demand point s, and θ (θ>1) is the customer's time sensitivity coefficient.

[0018] Preferably, since the goods at the demand point can be delivered by at least one drone, the satisfaction of company f for all goods delivered from warehouse r to demand point s is A frs for:

[0019]

[0020] Among them, w ukfrs is the weight of the cargo from warehouse r to demand point s of delivery company f when drone u performs the kth mission, w frs is the weight of goods that company f needs to deliver from warehouse r to demand point s.

[0021] Preferably, a priority index is constructed to quantitatively express the urgency of delivery. Through fair processing, it is ensured that the average priority of each logistics company is the same under the premise that the weight ratio of the task priority of the logistics company remains unchanged. The task priority after fair processing is used as the weight of service satisfaction. The weighted overall service satisfaction A total for:

[0022]

[0023] Among them, ω frs is the priority of the task of company f to distribute from warehouse r to demand point s after fair processing.

[0024] Preferably, in step S4, when the drone performs a delivery task, there is impedance, which is mainly composed of flight impedance, take-off and landing impedance, and charging impedance. The smaller the impedance, the higher the operating efficiency. The time when the drone u departs from the warehouse r to perform the kth task is The time to return to warehouse r after completing the task is Then the flight impedance of all UAVs is T fly It is expressed as:

[0025]

[0026] A drone can serve at least one demand point in a single mission, and needs to perform take-off and landing operations at least once during the delivery process. The take-off and landing impedance of all drones is T tl It is expressed as:

[0027]

[0028] Among them, y ukfrs is a binary variable, y ukfrs =1 means that drone u is performing the kth mission to deliver the goods of company f from warehouse r to demand point s, y ukfrs =0 means the drone does not deliver the goods, c tl The time it takes for a drone to take off and land once;

[0029] After the drone completes its mission, it returns to the warehouse to recharge. The charging time of the drone is proportional to the power consumption. Ignoring the power consumption of the drone taking off and landing, the total charging time is T e It is expressed as:

[0030]

[0031] Among them, Q e is the total power consumption of all drones, Q ful The full power of the drone is T ful How long does it take to charge the drone from zero to full power?

[0032] The total impedance T of the UAV performing the delivery task total for:

[0033] T total =T fly +T tl +T e

[0034] With the goal of minimizing the total impedance of drones and maximizing the overall service satisfaction, a drone task allocation model is constructed by combining the conservation of route network flow, drone task restrictions, flight path restrictions, maximum load constraints, and maximum range constraints. The objective function is:

[0035]

[0036] During the flight of the drone, in order to meet the mission requirements and performance constraints, the following constraints need to be met:

[0037] (1) Flow conservation

[0038] For any node in the route network, the number of incoming drones is equal to the number of outgoing drones, that is, the flow balance of each node is expressed as:

[0039]

[0040] Among them, x ukij is a binary variable, x ukij =1 means that UAV u flies through the segment between nodes i and j during the kth mission, otherwise x ukij =0.

[0041] (2) Task constraints

[0042] Logistics drones need to complete all delivery tasks, and each delivery point must have at least one drone providing delivery services, which means that:

[0043]

[0044]

[0045] (3) Path constraints

[0046] The starting point and end point of the logistics drone's delivery mission are both in the same warehouse, which can be expressed as:

[0047]

[0048] (4) Load constraints

[0049] The amount of cargo loaded by a logistics drone when taking off must not exceed its maximum load, expressed as:

[0050]

[0051] Among them, W max is the maximum payload of the drone;

[0052] (5) Range constraints

[0053] The range of a logistics drone is limited, and its flight distance must not exceed the maximum range. Since the range of a drone is related to its load, in order to increase the safety margin, the flight distance of a drone must not exceed its maximum range with full load, which can be expressed as:

[0054]

[0055] Among them, V max is the maximum range of the UAV with full load, d ij is the distance between node i and node j.

[0056] Preferably, in step S5, the UAV task allocation model belongs to an integer optimization model. The extracted research data is used as input data, and the NSGA-II algorithm is used for solving. The UAV is encoded according to its load, number of sorties, flight path, and departure time. The fitness is calculated according to the total impedance of the UAV and the overall service satisfaction objective function. The Pareto optimal solution set is obtained through non-dominated sorting and congestion calculation, and the logistics UAV resource allocation plan is output.

[0057] Therefore, the present invention adopts the above-mentioned logistics drone task allocation method with priority division, which can realize the collaborative task allocation of multiple logistics drones, help improve the operating efficiency and quality of logistics drones, and standardize the orderly operation of large-scale logistics drones.

[0058] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a flow chart of a logistics drone task allocation method with priority division according to the present invention;

[0060] Figure 2 This is a schematic diagram of the route network of the logistics UAV of the present invention;

[0061] Figure 3It is a schematic diagram of the NSGA-II algorithm flow of the present invention;

[0062] Figure 4 It is a schematic diagram of the task allocation effect of the logistics drone of the present invention. DETAILED DESCRIPTION

[0063] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0064] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0065] It is obvious to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the main idea or essential features of the present invention. Therefore, from all points of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention, and any reference numerals in the claims should not be regarded as limiting the claims involved.

[0066] Embodiment 1

[0067] like Figure 1-4 As shown, the present invention provides a logistics drone task allocation method with priority division, comprising the following steps:

[0068] S1. Based on the natural geographical environment, low-altitude airspace structure, and logistics UAV performance, extract the three-dimensional operation space information of the logistics UAV and obtain the performance parameters of the logistics UAV.

[0069] The three-dimensional operating space information of the logistics UAV is extracted through the geographic information system, including route network boundary data, route data and node data. The data includes longitude, latitude and altitude information. The performance parameters of the logistics UAV are obtained through the UAV performance manual, including maximum load, full load range, empty load range, full battery power, etc.

[0070] S2. Combine the needs of logistics companies and customers to obtain data on logistics drone transportation types, flow rates, flow directions, flow times, etc., and build logistics drone operation scenarios.

[0071] Based on the historical delivery data of the logistics company, the data on express delivery type, flow, direction, flow time, etc. in densely populated areas such as residential areas and office buildings are extracted to obtain the express delivery OD distribution matrix.

[0072] Logistics drones fly along the route network, which can be viewed as a graph G = (N, L), where N is the set of all nodes in the network, N = R∪S∪K, R is the set of logistics warehouses, S is the set of demand points, K is the set of turning points in the route network, and L is the set of flight segments between two nodes.

[0073] Logistics drones follow the "unified dispatch" management model. Various logistics companies deliver goods to logistics warehouses, and third-party agencies provide unified drone delivery services. One drone can serve a single demand point or multiple demand points continuously. After the drone completes the current delivery task, it returns to the logistics warehouse for charging and can continue to perform new delivery tasks.

[0074] S3. Considering the task priority and fairness of logistics companies, a logistics drone service satisfaction index is established.

[0075] Service satisfaction can be quantitatively expressed through the membership function of fuzzy delivery time. The satisfaction and the actual delivery time of the goods are in a superlinear power function relationship. F is the set of logistics companies involved, U is the set of logistics drones assigned by the third-party company, and K is u The number of times the drone performs delivery tasks.

[0076] The expected delivery time window of the drone delivery company f(f∈F) from warehouse r(r∈R) to the demand point s(s∈S) is [E frs ,L frs ], the maximum tolerable time window is [SE frs ,SL frs ], the actual delivery time is Then the membership function of the fuzzy delivery time of service satisfaction is:

[0077]

[0078] Among them, A ukfrs It means that the UAV u(u∈U) is executing the kth(k∈K u ) is the satisfaction of the delivery company f from warehouse r to demand point s, and θ (θ>1) is the customer's time sensitivity coefficient.

[0079] The goods at the demand point may be delivered by multiple drones, so the satisfaction of company f for all goods delivered from warehouse r to demand point s is A frs for:

[0080]

[0081] Among them, w ukfrs is the weight of the cargo from warehouse r to demand point s of delivery company f when drone u performs the kth mission, w frsis the weight of goods that company f needs to deliver from warehouse r to demand point s.

[0082] Due to the differences in cargo attributes, the urgency of delivery is quantitatively expressed by constructing a priority index. P is a pre-set task priority set. For any task, it is divided into three priorities: low, medium, and high, that is, P = {1, 2, 3}. The larger the value, the higher the priority. In order to prevent logistics companies from assigning their task priorities to a higher level and causing unfair distribution, it is necessary to use fair processing to ensure that the average priority of each logistics company is the same under the premise that the weight ratio of the logistics company's task priority remains unchanged. First, the task priority of each company is normalized to obtain the relative priority, which is expressed as:

[0083]

[0084] Among them, λ frs is the relative priority of the task of company f to distribute from warehouse r to demand point s, p frs is the initial priority of the task of company f to distribute from warehouse r to demand point s, P f is the task priority set of company f.

[0085] For company f, by introducing the fairness factor η f To ensure that all companies have the same average priority, at this time:

[0086]

[0087] The task priority after fair processing can be expressed as:

[0088]

[0089] Among them, ω frs is the absolute priority of the task of company f to distribute from warehouse r to demand point s.

[0090] After fair treatment, the average task priority of company f It can be expressed as:

[0091]

[0092] From the above formula, we can see that the average task priority of a logistics company is equal to the inverse of the sum of all task quantities, which has nothing to do with the task priority. Therefore, no matter how the company sets its task priority, its average task priority remains unchanged. The task priority after fair processing ranges from 0 to 1, which is used as the weight of service satisfaction. The weighted overall service satisfaction A total for:

[0093]

[0094] S4. Construct a logistics UAV task allocation model with the goal of minimizing flight impedance and maximizing service satisfaction.

[0095] When a drone performs a delivery task, there is impedance, which is mainly composed of flight impedance, take-off and landing impedance, and charging impedance. The smaller the impedance, the higher the operating efficiency. The time when drone u departs from warehouse r to perform the kth task is The time to return to warehouse r after completing the task is Then the flight impedance of all UAVs is T fly It can be expressed as:

[0096]

[0097] A drone can serve multiple demand points in a single mission. During the delivery process, it needs to perform take-off and landing operations multiple times. The take-off and landing impedance of all drones is T tl It can be expressed as:

[0098]

[0099] Among them, y ukfrs is a binary variable, y ukfrs =1 means that drone u is performing the kth mission to deliver the goods of company f from warehouse r to demand point s, y ukfrs =0 means the drone does not deliver the goods, c tl It is the time for a single take-off and landing of the drone.

[0100] After the drone completes its mission, it returns to the warehouse to recharge. If the drone is fully charged, the power level is Q ful , the maximum load is w max , full load range is l ful , the empty-load range is l emp , assuming that the UAV range and load are in linear relationship, the power consumption per unit range of the UAV is c w for:

[0101]

[0102] Among them, w ukij is the load of UAV u when it flies through segment ij during the kth mission;

[0103] At this time, the total power consumption of all drones is Q e It can be expressed as:

[0104]

[0105] Among them, x ukij is a binary variable, x ukij=1 means that UAV u flies through the segment between nodes i and j during the kth mission, d ij is the distance between node i and node j.

[0106] The charging time of the drone is proportional to the power consumption. If the charging time of the drone from zero power to full power is T ful , ignoring the power consumption of the drone taking off and landing, the total charging time is T e It can be expressed as:

[0107]

[0108] The total impedance T of the UAV performing the delivery task total for:

[0109] T total =T fly +T tl +T e

[0110] With the goal of minimizing the total impedance of drones and maximizing the overall service satisfaction, a drone task allocation model is constructed, and the objective function is:

[0111]

[0112] minT total =T fly +T tl +T e

[0113] During the flight of the drone, in order to meet the mission requirements and performance constraints, the following constraints need to be met:

[0114] (1) Flow conservation

[0115] For any node in the route network, the number of incoming drones is equal to the number of outgoing drones, that is, the flow balance of each node is expressed as:

[0116]

[0117] (2) Task constraints

[0118] Logistics drones need to complete all delivery tasks, and each delivery point must have at least one drone providing delivery services, which means that:

[0119]

[0120]

[0121] (3) Path constraints

[0122] The starting point and end point of the logistics drone's delivery mission are both in the same warehouse, which can be expressed as:

[0123]

[0124] (4) Load constraints

[0125] The amount of cargo loaded by a logistics drone when taking off must not exceed its maximum load, expressed as:

[0126]

[0127] Among them, W max The maximum payload of the drone.

[0128] (5) Range constraints

[0129] The range of a logistics drone is limited, and its flight distance must not exceed the maximum range. Since the range of a drone is related to its load, in order to increase the safety margin, the flight distance of a drone must not exceed its maximum range with full load, which can be expressed as:

[0130]

[0131] Among them, V max The maximum flight range of the drone when fully loaded.

[0132] S5. Use the NSGA-II algorithm to solve the model and output the optimal solution for logistics drone resource allocation.

[0133] The UAV task allocation model belongs to an integer optimization model. The extracted research data is used as input data and solved by the NSGA-II algorithm. The UAV load, sortie, flight path, and departure time are encoded to generate an initial population. The initial population generates a child population through non-dominated sorting, selection, crossover, and mutation. The two populations are merged to generate a new generation of parent population through non-dominated sorting and crowding calculation. A new generation of child population is generated through selection, crossover, and mutation. After the two populations are merged, the above operations are repeated until the maximum number of iterations is reached to obtain the Pareto optimal solution set and output the logistics UAV resource allocation plan.

[0134] The present invention provides a logistics UAV task allocation method with priority division. Considering the guiding role of the route network in structured airspace, combined with the performance constraints of UAVs, based on the spatiotemporal distribution requirements of cargo, with the goal of minimizing the total impedance of UAVs and maximizing the overall service satisfaction, a UAV task allocation model is established, and the UAV task allocation result is obtained by solving the NSGA-II algorithm. On the one hand, the priority index considering the fairness of logistics companies is introduced into the service satisfaction, making the UAV resource allocation more fair; on the other hand, the UAV take-off and landing and charging impedance are considered in the UAV task allocation model, making the task allocation closer to the actual situation.

[0135] Therefore, the present invention adopts the above-mentioned logistics drone task allocation method with priority division, which can realize the collaborative task allocation of multiple logistics drones, help improve the operating efficiency and quality of logistics drones, and standardize the orderly operation of large-scale logistics drones.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A logistics drone task allocation method with priority classification, characterized in that: The following steps are involved: S1. Based on the natural geographical environment, low-altitude airspace structure, and logistics UAV performance, extract the three-dimensional operation space information of the logistics UAV and obtain the performance parameters of the logistics UAV; S2. Combine the needs of logistics companies and customers to obtain the transportation type, flow, direction, and time data of logistics drones, and build logistics drone operation scenarios; S3. Combine task priority with fairness of logistics companies to establish a logistics drone service satisfaction index; In step S3, service satisfaction is quantitatively expressed through the membership function of fuzzy delivery time. The satisfaction and the actual delivery time of the goods are in a superlinear power function relationship. F is the set of logistics companies involved, U is the set of logistics drones assigned by the third-party company, and K is the service satisfaction of the logistics company. u The number of times the drone performs delivery tasks; The expected delivery time window of the drone delivery company f(f∈F) from warehouse r(r∈R) to the demand point s(s∈S) is [E frs ,L frs ], the maximum tolerable time window is [SE frs ,SL frs ], the actual delivery time is Then the membership function of the fuzzy delivery time of service satisfaction is: Among them, R is the set of logistics warehouses, S is the set of demand points, and A ukfrs It means that the UAV u(u∈U) is executing the kth(k∈K u ) is the satisfaction of the delivery company f from warehouse r to demand point s, θ (θ>1) is the customer time sensitivity coefficient; Since the goods at the demand point can be delivered by at least one drone, the satisfaction of all goods delivered by company f from warehouse r to demand point s is A frs for: Among them, w ukfrs is the weight of the cargo from warehouse r to demand point s of delivery company f when drone u performs the kth mission, w frs is the weight of goods that company f needs to deliver from warehouse r to demand point s; Priority indicators are constructed to quantitatively describe the urgency of delivery. Fairness processing is used to ensure that the average priority of each logistics company is the same under the premise that the weight ratio of the task priority of the logistics company remains unchanged. The task priority after fairness processing is used as the weight of service satisfaction. The weighted overall service satisfaction A total for: Among them, ω frs is the priority of the task of company f to distribute from warehouse r to demand point s after fair processing; S4. Construct a logistics UAV task allocation model with the goal of minimizing flight impedance and maximizing service satisfaction; S5. Use the NSGA-II algorithm to solve the model and output the optimal solution for logistics drone resource allocation.

2. A method for allocating logistics drone tasks with priority according to claim 1, characterized in that: In step S1, the three-dimensional operation space information of the logistics UAV is extracted through the geographic information system, including route network boundary data, route data and node data, and the performance parameters of the logistics UAV are obtained, including maximum load, full load range, empty load range and full power.

3. A logistics drone task allocation method with priority classification according to claim 1, characterized in that: In step S2, based on the historical delivery data of the logistics company, the express delivery type, flow, direction, and time data of the densely populated areas are extracted to obtain the express delivery OD distribution matrix; The logistics drone flies along the route network, which is regarded as a graph G = (N, L), where N is the set of all nodes in the network, N = R ∪ S ∪ K, K is the set of turning points in the route network, and L is the set of flight segments between two nodes; Logistics drones follow the "unified dispatch" management model. Various logistics companies deliver goods to logistics warehouses, and third-party agencies provide drone delivery services.

4. The method for allocating logistics drone tasks with priority according to claim 1 is characterized by: In step S4, when the drone performs the delivery task, there is impedance, which is mainly composed of flight impedance, take-off and landing impedance, and charging impedance. The smaller the impedance, the higher the operating efficiency. The time when the drone u departs from the warehouse r to perform the kth task is The time to return to warehouse r after completing the task is Then the flight impedance of all UAVs is T fly It is expressed as: A drone can serve at least one demand point in a single mission, and needs to perform take-off and landing operations at least once during the delivery process. The take-off and landing impedance of all drones is T tl It is expressed as: Among them, y ukfrs is a binary variable, y ukfrs =1 means that drone u is performing the kth mission to deliver the goods of company f from warehouse r to demand point s, y ukfrs =0 means the drone does not deliver the goods, c tl The time it takes for a drone to take off and land once; After completing the mission, the drone returns to the warehouse to charge. The charging time of the drone is proportional to the power consumption. Ignoring the power consumption of the drone taking off and landing, the total charging time is T e It is expressed as: Among them, Q e is the total power consumption of all drones, Q ful The full power of the drone is T ful How long does it take to charge the drone from zero to full power? The total impedance T of the UAV performing the delivery task total for: T total =T fly +T tl +T e With the goal of minimizing the total impedance of drones and maximizing the overall service satisfaction, a drone task allocation model is constructed by combining the conservation of route network flow, drone task restrictions, flight path restrictions, maximum load constraints, and maximum range constraints. The objective function is:

5. The method for allocating logistics drone tasks with priority according to claim 1 is characterized by: In step S5, the UAV task allocation model belongs to an integer optimization model. The extracted research data is used as input data and solved by the NSGA-II algorithm. The UAV load, sortie, flight path, and departure time are encoded. The fitness is calculated according to the total impedance of the UAV and the overall service satisfaction objective function. The Pareto optimal solution set is obtained through non-dominated sorting and congestion calculation, and the logistics UAV resource allocation plan is output.

Images