Optimization method and system for collaborative delivery of swarm drones

Through the clustered drone collaborative delivery solution, the path is optimized using the ant colony algorithm, the problems of battery life and fault recovery in drone distribution are solved, and the distribution efficiency and distribution accuracy are improved.

CN115018412BActive Publication Date: 2024-11-15HEFEI UNIV OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210606364.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-31
Publication Date
2024-11-15
Estimated Expiration
2042-05-31

AI Technical Summary

Technical Problem

In the existing drone distribution technology, the battery life is poor and the speed cannot be accurately planned, making it difficult to determine the return time, and the fault recovery is difficult, and the dynamic path leads to the complex path.

Method used

Adopt clustered drone collaborative distribution scheme, randomly select truck distribution points, calculate transfer probability, generate truck and drone distribution schemes, and combine ant colony algorithm to optimize paths to ensure that drones complete nearby delivery when trucks complete delivery tasks. Task.

Benefits of technology

It improves the efficiency of collaborative distribution of drones and trucks, solves the problem of distribution point allocation, realizes rapid iterative algorithm updates, and reduces the difficulty of drone failure recovery.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115018412B_ABST
    Figure CN115018412B_ABST
Patent Text Reader

Abstract

The embodiment of the present invention provides an optimization method and system for a swarm drone collaborative delivery solution, which belongs to the technical field of logistics distribution scheduling. Through this technical solution, the optimization method and system for a swarm drone collaborative delivery solution provided by the present invention solves the technical problem of the difficulty in determining the allocation of truck and drone customer points caused by the dynamic delivery of trucks and drones in the prior art by designing a truck to go to the corresponding truck delivery point to complete the delivery task, and at the same time combining a model in which the drone goes to a nearby delivery point to complete the delivery task when the truck completes the delivery task. On the other hand, the optimization method provided by the present invention also realizes rapid iteration of the algorithm by continuously updating the number and location of the points to be delivered in combination with the selected delivery points when determining the truck delivery points and the drone delivery points.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of logistics distribution scheduling, and in particular to an optimization method and system for a clustered unmanned aerial vehicle collaborative distribution solution. Background Art

[0002] In recent years, as drone technology has gradually matured, the use of drones for short-distance delivery tasks has gradually emerged in the field of logistics and distribution. In the prior art, there has been a technical solution that uses trucks to transport multiple drones, and the drones fly out in real time to complete the delivery task while the truck is moving, and return to the truck at a predetermined location according to the speed of the truck. However, this solution has the following technical defects: First, the battery life of drones is relatively poor, and it is difficult to achieve accurate speed planning during the actual dispatch process; second, drone delivery cannot guarantee the response enthusiasm of the personnel receiving the express delivery, so it is impossible to predict the waiting time of the drone, so it is difficult to determine the return time of the drone; third, once the drone fails, since the path itself is dynamic, it will also be difficult to recover the drone. Summary of the invention

[0003] The purpose of the embodiments of the present invention is to provide an optimization method and system for a swarm drone collaborative delivery solution, which can improve the efficiency of collaborative delivery between drones and trucks.

[0004] In order to achieve the above object, an embodiment of the present invention provides an optimization method for a swarm drone collaborative delivery solution, comprising:

[0005] Randomly select an unselected delivery point in the delivery area as the truck delivery point;

[0006] Selecting a delivery point where the drone can complete the delivery task when the truck stops at the current truck delivery point, with the selected truck delivery point as the center;

[0007] Determine whether there are any unselected delivery points in the to-be-delivered area;

[0008] In the case where it is determined that there are unselected delivery points in the to-be-delivered area, taking the currently selected truck delivery point as the origin, respectively calculating the transfer probabilities of the unselected delivery points in the to-be-delivered area;

[0009] Selecting an unselected delivery point in the to-be-delivered area as a new truck delivery point according to the transition probability, and returning to the step of selecting a delivery point where the drone can complete the delivery task when the truck stops at the current truck delivery point with the selected truck delivery point as the center;

[0010] In the case where it is determined that there is no unselected delivery point in the to-be-delivered area, generating a truck route delivery plan;

[0011] Determine whether the current first iteration number is greater than or equal to a preset first threshold;

[0012] If it is determined that the first iteration number is less than the first threshold, returning to the step of randomly selecting an unselected delivery point in the to-be-delivered area as a truck delivery point;

[0013] When it is determined that the first iteration number is greater than or equal to the first threshold, calculating the fitness of the currently generated truck path distribution solution;

[0014] updating the pheromone concentration according to the fitness;

[0015] Determine whether the current number of iterations is greater than or equal to a preset second threshold;

[0016] When it is determined that the current second iteration number is greater than or equal to the second threshold, generating a drone delivery plan according to the truck path delivery plan with the minimum fitness;

[0017] The truck path delivery plan and the drone delivery plan are combined to form a truck and drone collaborative delivery plan.

[0018] Optionally, respectively calculating the transfer probabilities of the unselected delivery points in the to-be-delivered area includes:

[0019] The transition probability is calculated according to formula (1):

[0020]

[0021] Among them, P ij k is the transition probability of selecting the jth delivery point from the ith delivery point as the next truck delivery point in the kth first iteration, τ ij (t) is the pheromone function between the ith distribution point and the jth distribution point in the tth second iteration, α is the pheromone importance factor, β is the heuristic importance, η ij (t) is the heuristic function between the i-th distribution point and the j-th distribution point in the t-th second iteration, allow k is the set of delivery points to be selected for the first k iterations.

[0022] Optionally, calculating the fitness of the currently generated truck route distribution solution includes:

[0023] The fitness is calculated according to formula (2):

[0024]

[0025] Among them, obj k is the fitness of the first iteration of the kth time, route k The truck routing solution generated for the kth first iteration, d ij is the truck delivery path between the i-th delivery point and the j-th delivery point, c T is the delivery energy consumption per unit distance of the truck, tabu k is the flight path of the drone, is the total path that the drone needs to fly when the truck stops at the jth delivery point, c M is the delivery energy consumption per unit distance of the drone.

[0026] Optionally, updating the pheromone concentration according to the fitness includes:

[0027] The pheromone concentration is updated according to formula (3) and formula (4),

[0028]

[0029]

[0030] Among them, τ ij (t+1) is the pheromone function between the i-th distribution point and the j-th distribution point in the t+1-th second iteration, ρ(t) is the pheromone volatility factor in the t+1-th second iteration, τ ij (t) is the pheromone function between the ith distribution point and the jth distribution point in the tth second iteration, Δτ ij is the sum of the pheromone concentrations between the i-th distribution point and the j-th distribution point in the k-th first iteration, n is the first threshold, Q is the total pheromone volatilization amount, obj k is the fitness of the first k-th iteration.

[0031] Optionally, updating the pheromone concentration according to the fitness includes:

[0032] Determine whether the third iteration number is greater than or equal to a third threshold;

[0033] In the case where it is determined that the third iteration number is greater than or equal to the third threshold, determining whether the minimum value of the fitness is updated in the iteration process up to the current third iteration number;

[0034] When it is determined that the minimum value of fitness has not been updated in the iteration process up to the third iteration number, the pheromone volatilization factor is updated using formula (5):

[0035]

[0036] Among them, ρ(t-1) is the pheromone volatility factor of the second iteration number t-1, ρ min It is the minimum value of the preset pheromone volatility factor.

[0037] Optionally, generating a drone delivery solution according to the truck path delivery solution with the minimum fitness includes:

[0038] Taking each of the delivery points as the origin, calculate the distance to each truck delivery point;

[0039] The truck delivery points corresponding to the minimum distance are selected and associated with each other to form the drone delivery plan.

[0040] Optionally, selecting truck delivery points corresponding to minimum distances to be associated with each other to form the drone delivery solution includes:

[0041] For each delivery point cluster formed after being associated with each other and consisting of a truck delivery point as the center and a corresponding drone delivery point, determine whether the number of delivery points in each delivery point cluster is less than or equal to 3;

[0042] When it is determined that the number of the delivery points is greater than 3, a preset ant colony algorithm is used to generate a corresponding drone path for the delivery point cluster.

[0043] Optionally, using a preset ant colony algorithm to generate a corresponding drone path for the delivery point cluster includes:

[0044] Taking the current position of the drone as the starting point, calculating the transfer probability of each delivery point in the delivery point cluster that has not been delivered by the drone, wherein the initial position is the vehicle delivery point of the delivery point cluster;

[0045] A roulette wheel method is used to select a delivery point as a proposed delivery point of the UAV according to the transfer probability;

[0046] Determine whether the drone is currently capable of completing the delivery task of the intended delivery point;

[0047] In the case where it is determined that the drone cannot complete the delivery task of the intended delivery point, a complete drone path is generated, and the vehicle delivery point is selected as the current position of the drone, and the step of returning to the current position of the drone as the starting point to calculate the transfer probability of each delivery point in the delivery point cluster that has not been delivered by the drone;

[0048] If it is determined that the drone can complete the delivery task of the intended delivery point, the intended delivery point is added to the current drone path to determine whether there are any undelivered delivery points;

[0049] In the case where it is determined that there are still undelivered delivery points, returning to the step of calculating the transfer probability of each undelivered delivery point of the delivery point cluster by taking the current position of the drone as the starting point;

[0050] When it is determined that there are no undelivered delivery points at present, a drone delivery plan for the current delivery point cluster is formed;

[0051] Determine whether the current fourth iteration number is greater than or equal to a preset fourth threshold;

[0052] When it is determined that the current fourth iteration number is less than the fourth threshold, returning to the step of calculating the transfer probability of each delivery point not delivered by the drone in the delivery point cluster with the current position of the drone as the starting point;

[0053] When it is determined that the current fourth iteration number is greater than the fourth threshold, the fitness value of each of the drone delivery solutions is calculated, and the drone delivery solution with the smallest fitness value is selected as the optimal solution in the current fourth iteration number;

[0054] updating the pheromone concentration according to the optimal solution;

[0055] Determine whether the current fifth iteration number is greater than or equal to a fifth threshold;

[0056] When it is determined that the fifth iteration number is less than the fifth threshold, returning to the step of calculating the transfer probability of each delivery point in the delivery point cluster that has not been delivered by the drone, starting from the current position of the drone;

[0057] When it is determined that the current fifth iteration number is greater than or equal to the fifth threshold, the drone delivery solution with a small fitness value is selected as the optimal solution for the current delivery point cluster.

[0058] On the other hand, the present invention also provides an optimization system for a clustered drone collaborative delivery solution, the optimization system comprising a processor, and the processor is used to execute any of the optimization methods described above.

[0059] On the other hand, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, wherein the instructions are used to be read by a machine so that the machine executes any of the optimization methods described above.

[0060] Through the above technical solution, the optimization method and system of the swarm drone collaborative delivery solution provided by the present invention solves the technical problem of the difficulty in determining the allocation of truck and drone customer points caused by the dynamic delivery of trucks and drones in the prior art by designing a truck to go to the corresponding truck delivery point to complete the delivery task, and at the same time combining a model in which the drone goes to a nearby delivery point to complete the delivery task when the truck completes the delivery task. On the other hand, the optimization method provided by the present invention also realizes the rapid iteration of the algorithm by continuously updating the number and location of the points to be delivered in combination with the selected delivery points when determining the truck delivery points and the drone delivery points.

[0061] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0063] Figure 1 is a flow chart of an optimization method for a swarm drone collaborative delivery solution according to an embodiment of the present invention;

[0064] Figure 2 is a partial flow chart of a method for optimizing a swarm drone collaborative delivery solution according to an embodiment of the present invention;

[0065] Figure 3 is a partial flow chart of a method for optimizing a swarm drone collaborative delivery solution according to an embodiment of the present invention;

[0066] Figure 4 It is a partial flow chart of an optimization method for a swarm drone collaborative delivery solution according to an embodiment of the present invention. DETAILED DESCRIPTION

[0067] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0068] like Figure 1 FIG. 1 is a flow chart of an optimization method for a swarm drone collaborative delivery solution according to an embodiment of the present invention. Figure 1 In the optimization method, the optimization method may include:

[0069] In step S10, a non-selected delivery point is randomly selected in the to-be-delivered area as a truck delivery point;

[0070] In step S11, a delivery point where the drone can complete the delivery task when the truck stops at the current truck delivery point is selected with the selected truck delivery point as the center;

[0071] In step S12, it is determined whether there are any unselected delivery points in the delivery area;

[0072] In step S13, when it is determined that there are unselected delivery points in the delivery area, the transfer probabilities of the unselected delivery points in the delivery area are calculated respectively with the currently selected truck delivery point as the origin;

[0073] In step S14, an unselected delivery point is selected in the delivery area as a new truck delivery point according to the transition probability, and the step of returning to the selected truck delivery point as the center to select a delivery point where the drone can complete the delivery task when the truck stops at the current truck delivery point;

[0074] In step S15, if it is determined that there is no unselected delivery point in the delivery area, a truck route delivery plan is generated;

[0075] In step S16, it is determined whether the current first iteration number is greater than or equal to a preset first threshold;

[0076] When it is determined that the first iteration number is less than the first threshold, returning to the step of randomly selecting an unselected delivery point in the to-be-delivered area as a truck delivery point;

[0077] In step S17, when it is determined that the first iteration number is greater than or equal to the first threshold, the fitness of the currently generated truck path distribution plan is calculated;

[0078] In step S18, the pheromone concentration is updated according to the fitness;

[0079] In step S19, it is determined whether the current number of iterations is greater than or equal to a preset second threshold;

[0080] In step S20, when it is determined that the current second iteration number is greater than or equal to the second threshold, a drone delivery plan is generated according to the truck path delivery plan with the minimum fitness;

[0081] In step S21, the truck path delivery plan and the drone delivery plan are combined to form a truck and drone collaborative delivery plan.

[0082] In this Figure 1In the method shown, steps S10 to S16 can be used to generate a truck path distribution plan. Specifically, step S10 adds a truck distribution point to the truck path distribution plan by selecting an unselected distribution point in the area to be distributed as a truck distribution point. When the truck is located at the current truck distribution point, the drone needs to go to a nearby distribution point to complete the distribution task, so step S11 can be used to find the distribution point of the drone with the truck distribution point as the center. In this embodiment, the distribution point of the drone can be associated with the corresponding truck distribution point to form a cluster (distribution point cluster). The method for obtaining the distribution point of the drone in step S11 can be a variety of methods known to those skilled in the art. In one example of the present invention, the method can be to first determine the flight range of the drone to screen the distribution point, that is, after going to the distribution point to complete the distribution task, whether the flight distance returning to the truck distribution point again is greater than or equal to the maximum flight distance of the drone; secondly, it can be judged by the cargo quality of the distribution task of each distribution point in the screened distribution points, that is, whether the cargo quality is greater than the maximum load-bearing capacity of the drone. More specifically, the method can include the following steps: Figure 2 In the steps shown in Figure 2 In, the method may include:

[0083] In step S30, the current flyable area of ​​the drone is drawn within the delivery area, with the truck delivery point as the center and half of the maximum flight distance of the drone as the radius;

[0084] In step S31, a non-selected delivery point is randomly selected from the delivery points within the current flyable area of ​​the drone;

[0085] In step S32, it is determined whether the mass of the goods at the selected delivery point is greater than the maximum load-bearing capacity of the drone;

[0086] In step S33, when it is determined that the cargo mass of the selected delivery point is greater than the maximum load-bearing capacity of the drone, it is determined whether there are any unselected delivery points in the current flyable area;

[0087] In step S34, if it is determined that the cargo mass of the selected delivery point is less than or equal to the maximum load-bearing capacity of the drone, the delivery point is added to the current cluster and step S33 is executed;

[0088] In step S35, when it is determined that there are no unselected delivery points in the current flyable area, it is determined to generate the current cluster; otherwise, when it is determined that there are still unselected delivery points in the current flyable area, it returns to execute step S31.

[0089] Step S12 can be used to determine whether there are any unselected delivery points in the current area to be delivered, that is, whether all the current delivery points have been selected, so as to determine whether a complete truck path delivery plan can be generated at present. In the case where it is determined that there are no unselected delivery points in the current area to be delivered, step S15 can be executed at this time, that is, a truck path delivery plan is generated. On the contrary, if there are still unselected delivery points, the currently selected truck delivery point can be used as the origin to calculate the transfer probability of the unselected delivery points in the area to be delivered, and then the new truck delivery point can be selected in combination with the transfer probability, that is, step S13 and step S14 are executed. Among them, the specific calculation method of the transfer probability can be a variety of methods known to those skilled in the art. However, in a preferred example of the present invention, the calculation method can be to calculate the transfer probability according to formula (1),

[0090]

[0091] Among them, P ij k is the transition probability of selecting the jth delivery point from the ith delivery point as the next truck delivery point in the kth first iteration, τ ij (t) is the pheromone function between the ith distribution point and the jth distribution point in the tth second iteration, α is the pheromone importance factor, β is the heuristic importance, η ij (t) is the heuristic function between the i-th distribution point and the j-th distribution point in the t-th second iteration, allow k is the set of delivery points to be selected for the first k iterations.

[0092] Although the method of selecting a new truck delivery point according to the transition probability in step S14 may be a variety of methods known to those skilled in the art, in a preferred embodiment of the present invention, the method may be to select a new truck delivery point by using a roulette wheel method.

[0093] After a truck route distribution plan is generated in step S15, considering that the method of selecting truck distribution points based on transition probability cannot generate the optimal solution in one round of iteration, multiple truck route distribution plans can be generated through the first iteration judgment in step S16, that is, judging whether the first iteration number is greater than or equal to a preset first threshold. The first threshold is the number of truck route distribution plans generated after the first iteration judgment in one round.

[0094] In step S16, when it is determined that the first iteration number is greater than or equal to the first threshold, in order to select the optimal solution from the generated multiple truck path distribution solutions, it is necessary to calculate the fitness of each truck path distribution solution. The fitness can be calculated in various forms known to those skilled in the art. However, in a preferred example of the present invention, the fitness can be calculated according to formula (2):

[0095]

[0096] Among them, obj k is the fitness of the first iteration of the kth time, route k The truck routing solution generated for the kth first iteration, d ij is the truck delivery path between the i-th delivery point and the j-th delivery point, c T is the delivery energy consumption per unit distance of the truck, tabu k is the flight path of the drone, is the total path that the drone needs to fly when the truck stops at the jth delivery point, c M is the delivery energy consumption per unit distance of the drone.

[0097] After calculating the fitness of each truck path distribution plan, the pheromone concentration can be further updated according to each truck path distribution plan, thereby updating the transition probability of the next first iteration process. Although the specific method for updating the pheromone concentration can be a variety of methods known to those skilled in the art, in a preferred example of the present invention, the method for updating the pheromone concentration can be to update the pheromone concentration according to formula (3) and formula (4),

[0098]

[0099]

[0100] Among them, τ ij (t+1) is the pheromone function between the i-th distribution point and the j-th distribution point in the t+1-th second iteration, ρ(t) is the pheromone volatility factor in the t+1-th second iteration, τ ij (t) is the pheromone function between the ith distribution point and the jth distribution point in the tth second iteration, Δτ ij is the sum of the pheromone concentrations between the i-th distribution point and the j-th distribution point in the k-th first iteration, n is the first threshold, Q is the total pheromone volatilization amount, obj kis the fitness of the first iteration number of the kth time. Furthermore, in order to avoid the situation that the pheromone concentration update fails to optimize the transfer probability during the iteration process, the third iteration can be used to determine whether the function of updating the pheromone concentration itself needs to be updated, that is, first determine whether the third iteration number is greater than or equal to the third threshold; if the third iteration number is greater than or equal to the third threshold, then determine whether the minimum value of the fitness is updated in the iteration process up to the current third iteration number; if the minimum value of the fitness is not updated in the iteration process up to the current third iteration number, the pheromone volatility factor can be updated using formula (5),

[0101]

[0102] Among them, ρ(t-1) is the pheromone volatility factor of the second iteration number t-1, ρ min It is the minimum value of the preset pheromone volatility factor.

[0103] After the pheromone concentration is updated in step S18, the transfer probability calculated in step S13 will also change accordingly. In order to determine whether the change can optimize the truck delivery path, the second iteration judgment can be used to return to generate multiple new truck path delivery plans, that is, to determine whether the current second iteration number is greater than or equal to the second threshold, and if the current iteration number is greater than or equal to the second threshold, return to execute step S10. On the contrary, if the current iteration number is greater than or equal to the second threshold, it means that the optimal truck path delivery plan can be generated. Then step S20 can be further executed, that is, the UAV delivery plan is generated according to the truck path delivery plan with the smallest fitness (that is, the optimal truck path delivery plan), and finally, the truck path delivery plan and the UAV delivery plan are combined to form the optimal truck and UAV collaborative delivery plan.

[0104] In step S20, before generating the drone delivery plan, it is necessary to first ensure that each delivery point cluster in the truck route delivery plan can ensure that the distance from the drone delivery point to the truck delivery point is the shortest. Although the truck route delivery plan generated by steps S10 to S18 can ensure that the distance from the delivery point to the truck delivery point is the shortest within the same delivery point cluster, it cannot ensure that the distance from the delivery point within the cluster to the truck delivery point outside the cluster is the shortest. Therefore, it is necessary to first ensure that each delivery point cluster in the truck route delivery plan can ensure that the distance from the drone delivery point to the truck delivery point is the shortest before generating the drone delivery plan. Figure 3 The steps shown in are used to update the cluster of truck routing solutions. Figure 3 In, the updating method may include:

[0105] In step S40, taking each delivery point as the origin, the distance to each truck delivery point is calculated;

[0106] In step S41, the truck delivery points corresponding to the minimum distance are selected and associated with each other to form a drone delivery plan.

[0107] The method for generating the drone delivery plan may be a variety of methods known to those skilled in the art. For example, the truck delivery points in the cluster may be directly used as the initial delivery points, and the method may be similar to the following: Figure 1 The method in . Figure 1 The difference between the method shown in is that, within a distribution point cluster, there may be only one or only two distribution points. In this case, if there is only one distribution point, the drone path can be directly obtained; and for the case of only two distribution points, it is only necessary to determine whether a drone can complete the delivery tasks of two distribution points in one flight. If it can, the drone path is to start from the truck distribution point, pass through the two distribution points in sequence, and then return to the truck distribution point. On the contrary, if the drone cannot complete the delivery tasks of the two distribution points in one flight, the drone path is to start from the truck distribution point, go to the first distribution point, return to the truck distribution point, then go to the second distribution point, and return to the truck distribution point.

[0108] For the case where the number of delivery points in a delivery point cluster is greater than or equal to 3 (the truck delivery point itself is also a delivery point), a method similar to Figure 1 The method in the embodiment of the present invention is to generate the drone path, that is, to use the preset ant colony algorithm to generate the corresponding drone path for the distribution point cluster. Although the ant colony algorithm can be a variety of methods known to those skilled in the art, in a preferred example of the present invention, the method may include: Figure 4 In the steps shown in Figure 4 In, the method may include:

[0109] In step S50, the current position of the drone is used as the starting point to calculate the transition probability of each delivery point in the delivery point cluster that has not been delivered by the drone. The initial position is the vehicle delivery point in the delivery point cluster. In step S50, the method for calculating the transition probability can be the same as Figure 1 The method is the same as that in , that is, formula (1) is used for calculation.

[0110] In step S51, a roulette wheel method is used to select a delivery point as a proposed delivery point of the drone according to the transfer probability;

[0111] In step S52, it is determined whether the drone can currently complete the delivery task of the intended delivery point. In step S52, since the load capacity of the drone is different from that of the truck itself, the drone is more likely to be delivered to the destination point than the truck. Figure 1 For the method in , when generating the drone path, it is necessary to consider whether the drone itself can complete the delivery task. Specifically, the method for determining the delivery task can be the same as Figure 2 The method is similar to that in , that is, first determine whether the maximum flight distance of the drone can complete the delivery task, and secondly determine whether the current maximum load-bearing capacity of the drone can complete the delivery task of the delivery point. It is worth noting here that the drone may be performing a delivery task for a single delivery point or for multiple delivery points during the delivery task. Then, considering the case of performing multiple delivery tasks, when determining whether the maximum flight distance of the drone can complete the delivery task, the maximum flight distance should be the remaining flight distance of the drone; similarly, when determining whether the maximum load-bearing capacity of the drone can complete the delivery task, the maximum load-bearing capacity should be the remaining load-bearing capacity of the drone.

[0112] In step S53, if it is determined that the drone cannot complete the delivery task of the intended delivery point, a complete drone path is generated, and the vehicle delivery point is selected as the current position of the drone, and the process returns to the step of calculating the transfer probability of each delivery point in the delivery point cluster that has not been delivered by the drone, starting from the current position of the drone;

[0113] In step S54, if it is determined that the drone can complete the delivery task of the intended delivery point, the intended delivery point is added to the current drone path to determine whether there are any undelivered delivery points;

[0114] In step S55, when it is determined that there are still undelivered delivery points, the process returns to the step of calculating the transfer probability of each undelivered delivery point in the delivery point cluster with the current position of the drone as the starting point;

[0115] In step S56, when it is determined that there is no undelivered delivery point at present, a drone delivery plan for the current delivery point cluster is formed;

[0116] In step S57, it is determined whether the current fourth iteration number is greater than or equal to a preset fourth threshold;

[0117] When it is determined that the current fourth iteration number is less than the fourth threshold, returning to the step of calculating the transfer probability of each delivery point in the delivery point cluster that has not been delivered by the drone, taking the current position of the drone as the starting point;

[0118] In step S58, when it is determined that the current fourth iteration number is greater than or equal to the fourth threshold, the fitness value of each drone delivery solution is calculated, and the drone delivery solution with the smallest fitness value is selected as the optimal solution in the current fourth iteration number. Figure 1 The method in is similar to that in which the fitness value is determined by calculating the product of the travel distance of the drone in the drone delivery solution and the delivery energy consumption per unit distance.

[0119] In step S59, the pheromone concentration is updated according to the optimal solution. The method for updating the pheromone concentration can also be the same as Figure 1 The method is the same as in , that is, formula (3) to formula (5) are used to update.

[0120] In step S60, it is determined whether the current fifth iteration number is greater than or equal to a fifth threshold;

[0121] When it is determined that the fifth iteration number is less than the fifth threshold, returning to the step of calculating the transfer probability of each delivery point in the delivery point cluster that has not been delivered by the drone, taking the current position of the drone as the starting point;

[0122] In step S61, when it is determined that the current fifth iteration number is greater than or equal to the fifth threshold, the drone delivery solution with a small fitness value is selected as the optimal solution for the current delivery point cluster.

[0123] After the truck and drone collaborative delivery solution is generated in step S21, in this embodiment, in order to further reflect the total energy consumption of the truck and drone collaborative delivery solution, the total energy consumption can be calculated and displayed according to formula (6):

[0124]

[0125] Among them, total cost is the total energy consumption, lengt best is the set of straight-line paths for trucks in the truck routing solution, dronedis i is the i-th straight line path of the UAV in the UAV path, and u is the number of straight line paths.

[0126] On the other hand, the present invention also provides an optimization system for a clustered drone collaborative delivery solution, the optimization system comprising a processor, and the processor is used to execute any of the optimization methods described above.

[0127] On the other hand, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, wherein the instructions are used to be read by a machine so that the machine executes any of the optimization methods described above.

[0128] Through the above technical solution, the optimization method and system of the swarm drone collaborative delivery solution provided by the present invention solves the technical problem of difficulty in determining the delivery time caused by the dynamic delivery of trucks and drones in the prior art by designing a truck to go to the corresponding truck delivery point to complete the delivery task, and at the same time combining a model in which the drone goes to a nearby delivery point to complete the delivery task when the truck completes the delivery task. On the other hand, the optimization method provided by the present invention also realizes the rapid iteration of the algorithm by continuously updating the number and location of the delivery points to be delivered in combination with the selected delivery points when determining the truck delivery points and the drone delivery points.

[0129] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0130] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0131] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0132] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0133] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0134] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0135] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0136] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0137] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. An optimization method for clustered drone collaborative delivery solutions, It is characterized in that The optimization method comprises: Randomly select an unselected delivery point in the delivery area as the truck delivery point; Selecting a delivery point where the drone can complete the delivery task when the truck stops at the current truck delivery point, with the selected truck delivery point as the center; Determine whether there are any unselected delivery points in the to-be-delivered area; In the case where it is determined that there are unselected delivery points in the to-be-delivered area, taking the currently selected truck delivery point as the origin, respectively calculating the transfer probabilities of the unselected delivery points in the to-be-delivered area; Selecting an unselected delivery point in the to-be-delivered area as a new truck delivery point according to the transition probability, and returning to the step of selecting a delivery point where the drone can complete the delivery task when the truck stops at the current truck delivery point with the selected truck delivery point as the center; In the case where it is determined that there is no unselected delivery point in the to-be-delivered area, generating a truck route delivery plan; Determine whether the current first iteration number is greater than or equal to a preset first threshold; If it is determined that the first iteration number is less than the first threshold, returning to the step of randomly selecting an unselected delivery point in the to-be-delivered area as a truck delivery point; When it is determined that the first iteration number is greater than or equal to the first threshold, calculating the fitness of the currently generated truck path distribution solution; updating the pheromone concentration according to the fitness; Determine whether the current number of iterations is greater than or equal to a preset second threshold; When it is determined that the current second iteration number is greater than or equal to the second threshold, generating a drone delivery plan according to the truck path delivery plan with the minimum fitness; Combining the truck path delivery solution and the drone delivery solution to form a truck and drone collaborative delivery solution; Calculating the fitness of the currently generated truck routing solution includes: The fitness is calculated according to formula (2): ,(2) in, For the The fitness of the first iteration, For the The truck routing solution generated by the first iteration is For the The delivery point and Truck delivery routes for each delivery point, is the delivery energy consumption per unit distance of the truck, is the flight path of the drone, For the truck The total route that the drone needs to fly when stopping at multiple delivery points is: is the delivery energy consumption per unit distance of the drone.

2. The optimization method according to claim 1, It is characterized in that Calculating the transfer probabilities of the unselected delivery points in the delivery area includes: The transition probability is calculated according to formula (1): ,(1) in, For the The first iteration number is from Select the delivery point The transfer probability of a delivery point as the next truck delivery point, For the The second iteration number The distribution point and The pheromone function between the distribution points, is the pheromone importance factor, is the heuristic importance, For the The second iteration number The distribution point and The heuristic function between the delivery points is for The set of delivery points to be selected for the first iteration.

3. The optimization method according to claim 1, It is characterized in that Updating the pheromone concentration according to the fitness includes: The pheromone concentration is updated according to formula (3) and formula (4), ,(3) ,(4) in, For the The second iteration number The distribution point and The pheromone function between the distribution points, For the The pheromone volatility factor in the second iteration, For the The second iteration number The distribution point and The pheromone function between the distribution points, For the In the first iteration, The distribution point and The sum of the pheromone concentrations between the distribution points, is the first threshold, is the total amount of pheromone volatilization, For the The fitness of the first iteration.

4. The optimization method according to claim 3, It is characterized in that Updating the pheromone concentration according to the fitness includes: Determine whether the third iteration number is greater than or equal to a third threshold; In the case where it is determined that the third iteration number is greater than or equal to the third threshold, determining whether the minimum value of the fitness is updated in the iteration process up to the current third iteration number; When it is determined that the minimum value of fitness has not been updated in the iteration process up to the third iteration, the pheromone volatilization factor is updated using formula (5): ,(5) in, For the The pheromone volatility factor of the second iteration, It is the minimum value of the preset pheromone volatility factor.

5. The optimization method according to claim 1, It is characterized in that Generating a drone delivery plan based on the truck path delivery plan with the minimum fitness includes: Taking each of the delivery points as the origin, calculate the distance to each truck delivery point; The truck delivery points corresponding to the minimum distance are selected and associated with each other to form the drone delivery plan.

6. The optimization method according to claim 5, It is characterized in that Selecting the truck delivery points corresponding to the minimum distances to be associated with each other to form the drone delivery solution includes: For each delivery point cluster formed after being associated with each other and consisting of a truck delivery point as the center and a corresponding drone delivery point, determine whether the number of delivery points in each delivery point cluster is less than or equal to 3; When it is determined that the number of the delivery points is greater than 3, a preset ant colony algorithm is used to generate a corresponding drone path for the delivery point cluster.

7. The optimization method according to claim 5, It is characterized in that The preset ant colony algorithm is used to generate the corresponding drone path for the distribution point cluster, including: Taking the current position of the drone as the starting point, calculating the transfer probability of each delivery point in the delivery point cluster that has not been delivered by the drone, wherein the initial position is the vehicle delivery point of the delivery point cluster; A roulette wheel method is used to select a delivery point as a proposed delivery point of the UAV according to the transfer probability; Determine whether the drone is currently capable of completing the delivery task of the intended delivery point; In the case where it is determined that the drone cannot complete the delivery task of the intended delivery point, a complete drone path is generated, and the vehicle delivery point is selected as the current position of the drone, and the step of returning to the current position of the drone as the starting point to calculate the transfer probability of each delivery point in the delivery point cluster that has not been delivered by the drone; If it is determined that the drone can complete the delivery task of the intended delivery point, the intended delivery point is added to the current drone path to determine whether there are any undelivered delivery points; In the case where it is determined that there are still undelivered delivery points, returning to the step of calculating the transfer probability of each undelivered delivery point of the delivery point cluster by taking the current position of the drone as the starting point; When it is determined that there are no undelivered delivery points at present, a drone delivery plan for the current delivery point cluster is formed; Determine whether the current fourth iteration number is greater than or equal to a preset fourth threshold; When it is determined that the current fourth iteration number is less than the fourth threshold, returning to the step of calculating the transfer probability of each delivery point not delivered by the drone in the delivery point cluster with the current position of the drone as the starting point; When it is determined that the current fourth iteration number is greater than the fourth threshold, the fitness value of each of the drone delivery solutions is calculated, and the drone delivery solution with the smallest fitness value is selected as the optimal solution in the current fourth iteration number; updating the pheromone concentration according to the optimal solution; Determine whether the current fifth iteration number is greater than or equal to a fifth threshold; When it is determined that the fifth iteration number is less than the fifth threshold, returning to the step of calculating the transfer probability of each delivery point in the delivery point cluster that has not been delivered by the drone, starting from the current position of the drone; When it is determined that the current fifth iteration number is greater than or equal to the fifth threshold, the drone delivery solution with a small fitness value is selected as the optimal solution for the current delivery point cluster.

8. An optimization system for clustered drone collaborative delivery solutions, It is characterized in that The optimization system comprises a processor, and the processor is used to execute the optimization method according to any one of claims 1 to 7.

9. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores instructions, and the instructions are used to be read by a machine so that the machine executes the optimization method according to any one of claims 1 to 7.