WO2019114153A1

WO2019114153A1 - Order picking path planning method and device

Info

Publication number: WO2019114153A1
Application number: PCT/CN2018/081019
Authority: WO
Inventors: 王天文; 董红宇; 莫泽
Original assignee: 北京京东尚科信息技术有限公司; 北京京东世纪贸易有限公司
Priority date: 2017-12-12
Filing date: 2018-03-29
Publication date: 2019-06-20
Also published as: CN109919345A; CN109919345B

Abstract

The present invention provides an order picking path planning method and device based on an ant colony algorithm. The order picking path planning method comprises: using a plurality of nodes as start points, and starting path traversal of a plurality of ants, the nodes comprising a plurality of order picking nodes, order picking start points, and order picking end points; calculating the transition probability of the ants to other nodes according to pheromone values of paths among the plurality of nodes, and moving the ants to the node with the maximum transition probability; updating the pheromone values of the paths after determining that an ant traverses all the nodes, and recording a traversal path; and determining an optimal path according to the plurality of traversal paths. The order picking path planning method can improve the order picking efficiency and save the order picking path planning time.

Description

Picking path planning method and device

Technical field

The present disclosure relates to the field of machine learning technology, and in particular to a method and apparatus for picking path planning based on an elite ant colony algorithm.

Background technique

With the development of e-commerce, path planning in the process of picking goods in warehouses has become an important part of improving logistics efficiency. In the warehouse, due to the large area of the logical area, the area of the selected area is wide, and there are usually dozens of storage points in a collection list. Finding a reasonable picking path can reduce the picking distance, improve the picking efficiency and save. Labor costs. Especially during the promotion period, by optimizing the picking process, it is possible to significantly speed up the delivery, improve the utilization of the warehouse and improve the user experience.

Picking path optimization refers to reducing the walking distance of pickers in warehouses by optimizing, picking out shipments in the shortest time, and improving picking efficiency. Related pick path optimizations often pass a policy-based approach, such as a picking method based on a U-type strategy or an S-type strategy. Since the warehouse storage layout is different, the existing picking path strategy can give a picking path plan with a fixed starting point and an ending point. However, due to the fixed constraints of the starting point and the ending point, the generated traversing path is difficult to ensure the shortest path. Therefore, the problem of picking personnel detours and low picking efficiency; in addition, the picking path optimization problem belongs to the NP-Hard problem. As the scale of the problem increases, the original picking path algorithm still has solution time and optimization effect. Must increase the space.

Therefore, there is a need for a picking path planning method that can quickly plan an optimal picking path.

It should be noted that the information disclosed in the Background section above is only for enhancement of understanding of the background of the present disclosure, and thus may include information that does not constitute prior art known to those of ordinary skill in the art.

Summary of the invention

The purpose of the present disclosure is to provide a picking path planning method and a path planning device for providing a better picking path planning scheme while improving the efficiency of the picking path planning.

According to a first aspect of the embodiments of the present disclosure, a method for planning a picking path includes: starting a path traversal of a plurality of ants starting from a plurality of nodes, the node including a plurality of picking nodes, a picking starting point, a picking end point; calculating, according to a pheromone value of a path between the plurality of nodes, a transition probability of the ant transferring to the other node, moving the ant to the node with the highest transition probability; determining one After the ant traverses all the nodes, the pheromone value of the path is updated, and a traversal path is recorded; the optimal path is determined according to the plurality of traversal paths.

According to a second aspect of the embodiments of the present disclosure, a path planning apparatus is provided, including: an initialization module, configured to start path traversal of multiple ants starting from a plurality of nodes, the node including multiple picking nodes, and picking a starting point of the goods, a picking end point; the path generating module is configured to calculate a transition probability of the ant transferring to the other node according to a pheromone value of the path between the plurality of nodes, and move the ant to the transition probability The pheromone update module is configured to: after determining that an ant traverses all the nodes, update the pheromone value of the path, and record a traversal path; and the optimal selection module is set to determine the most according to the multiple traversal paths. Excellent path.

According to a third aspect of the present disclosure, there is provided a path planning apparatus comprising: a memory; and a processor coupled to the associated memory, the processor being configured to perform any one of the above, based on an instruction stored in the memory The method described in the item.

According to a fourth aspect of the present disclosure, there is provided a computer readable storage medium having stored thereon a program, the program being executed by a processor to implement the picking path planning method according to any of the above.

The picking path planning method provided by the present disclosure uses the optimal picking path scheme generated based on the existing strategy as a reference elite scheme of the asynchronous parallel elite ant colony optimization algorithm, and asynchronously calculates the optimal picking path scheme in parallel, and improves the picking The efficiency of the cargo path planning provides a better picking path planning solution. The asynchronous parallel elite ant colony optimization algorithm guarantees the solution time and optimization effect of the algorithm, and solves the problem that the optimization strategy of the existing pure strategy-based picking path planning scheme is not ideal and the time difference of the algorithm such as machine learning is poor.

The above general description and the following detailed description are intended to be illustrative and not restrictive.

DRAWINGS

The accompanying drawings, which are incorporated in the specification It is apparent that the drawings in the following description are only some of the embodiments of the present disclosure, and other drawings may be obtained from those skilled in the art without departing from the drawings.

Fig. 1 schematically shows a flow chart of a method of planning a picking path of the present disclosure.

2 is a flow chart schematically showing a method of picking a path in an embodiment of the present disclosure.

FIG. 3 schematically shows a flow chart of a picking path planning method in one embodiment of the present disclosure.

4 is a flow chart schematically showing a method of picking a path in an embodiment of the present disclosure.

Fig. 5 schematically shows a block diagram of a picking path planning device in an exemplary embodiment of the present disclosure.

Fig. 6 schematically shows a block diagram of another picking path planning device in an exemplary embodiment of the present disclosure.

Detailed ways

Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be embodied in a variety of forms and should not be construed as being limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be more complete and complete, To those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are set forth However, one skilled in the art will appreciate that one or more of the specific details may be omitted or other methods, components, devices, steps, etc. may be employed. In other instances, various aspects of the present disclosure are not obscured by the details of the invention.

Further, the drawings are only schematic illustrations of the present disclosure, and the same reference numerals are used to refer to the same or like parts in the drawings, and the repeated description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily have to correspond to physically or logically separate entities. These functional entities may be implemented in software, or implemented in one or more hardware modules or integrated circuits, or implemented in different network and/or processor devices and/or microcontroller devices.

The exemplary embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

FIG. 1 schematically shows a flow chart of a picking path planning method in an exemplary embodiment of the present disclosure. Referring to Figure 1, the picking path planning method 100 can include:

Step S102, starting a path traversal of multiple ants starting from multiple nodes, where the node includes multiple picking nodes, a picking starting point, and a picking end point;

Step S104: Calculate, according to a pheromone value of a path between the plurality of nodes, a transition probability of the ant transferring to another node, and move the ant to the node with the highest transition probability;

Step S106, determining that an ant traverses all the nodes, updating the pheromone value of the path, and recording a traversal path;

Step S108, determining an optimal path according to the plurality of traversal paths.

The picking path optimization method based on the asynchronous parallel elite ant colony algorithm provided by the present disclosure is based on the classic ACO (Ant colony optimization algorithm, a biomimetic intelligent optimization method proposed by Marco Dorigo in his doctoral thesis in 1992). On the premise of ensuring the optimization rate index, the algorithm running time is reduced and the optimization efficiency of the picking path is improved.

First, in this method, the starting route of the ant is not limited to N calibration starting points, and the algorithm calculation phase ensures that each node can be used as a starting point (asynchronous parallel computing); secondly, a historical optimal scheme is introduced to algorithm calculation. In the process of ensuring the quality of the solution, the running time of the algorithm (elite strategy) is reduced.

Next, each step of the picking path planning method 100 will be described in detail.

In step S102, a path traversal of a plurality of ants is started starting from a plurality of nodes, and the node includes a plurality of picking nodes, a picking starting point, and a picking end point.

The model can be initialized first before the calculation begins. Set the time t=0, the number of path traversal cycles Nc=0, the total number of cycles of all ants does not exceed Ncmax, the initialization information amount and information amount increment of the path (i, j) between the picking nodes i and j It is τ _ij (0)=const, Δτ _ij (0)=0.

At the beginning of the calculation, first place multiple ants on multiple nodes (including picking node, picking starting point, picking end point), start path traversal from multiple nodes, and set Nc=Nc+1.

Different from the related path planning method, this method improves the optimization degree and calculation speed of the path planning scheme by setting multiple starting nodes.

Step S104: Calculate a transition probability of the ant to transfer to another node according to a pheromone value of a path between the plurality of nodes, and move the ant to the node with the highest transition probability.

The probability that the ant k selects the city j, j ∈ {C-tabu _k }, can be calculated according to the following state transition probability formula (1). among them,

For the transition probability of ant k moving to the jth node on the i-th node, t is the number of times the path travels, C is the set of all nodes, tabu _k is the node that ant k has passed, and d _ij is i and j The distance between two nodes, α is a pheromone index with a value range of [0, 5], β is a heuristic factor index with a value range of [0, 5], and τ _ij is a node i and j The current pheromone value of the path.

After calculating the transition probability of the ant to other nodes on the node, the ant k is moved to the node with the highest transition probability. In one embodiment, each ant has a taboo table corresponding thereto for recording the nodes it passes through to exclude these nodes during calculation, preventing repeated walking and ensuring ergodic efficiency. Therefore, after transferring the ant k to the new node, it is necessary to record the previous node into the taboo table tabu _k .

Step S106, after determining that an ant traverses all the nodes, updating the pheromone value of the path and recording a traversal path.

The method for judging that the ant traverses all the nodes may be, for example, a view of whether the union of the tabu table and the current node is equal to the set of all nodes. If the ant does not traverse all the nodes, continue to select the next node according to step S104. If the ant has traversed all the nodes, record the path parameters such as the path sequence and the total path length of the current traversal, and update the path through the path. Pheromone.

In one embodiment, updating the pheromone can be performed according to the following formula:

Where τ _ij is the current pheromone value of the path between nodes i and j, Δτ _ij represents the updated pheromone increment; L _k represents the total length of the path taken by the kth ant in this cycle; Q represents ant The total amount of pheromone released on the path in one cycle, affecting the convergence speed of the algorithm; γ represents the weight parameter; T ^bs represents the current optimal path, L _bs represents the length of the current optimal path; ρ represents the evaporation coefficient.

The recommended values for each parameter of this method are shown in Table 1.

Table 1

序号Serial number	参数parameter	建议值suggested value	解释Explanation
11	0≤α≤50 ≤ α ≤ 5	11	信息素指数Pheromone index
22	0≤β≤50 ≤ β ≤ 5	55	启发式因子指数Heuristic factor index
33	0.1≤ρ≤0.990.1≤ρ≤0.99	0.10.1	蒸发系数Evaporation coefficient
44	10≤Q≤1000010≤Q≤10000	100100	信息素总量Total pheromone
55	初始化信息素constInitialization pheromone const	11	初始化信息素Initialization pheromone
66	γγ	100100	权重系数Weight coefficient
77	蚂蚁数/节点数≈1.1Number of ants / number of nodes ≈ 1.1	1.1节点数(向上取整)1.1 number of nodes (rounded up)	蚂蚁数Ant number
88	循环计数或重复计数Loop count or repeat count	Ncmax＝4或K＝10Ncmax=4 or K=10	停止条件Stop condition

The plurality of traversal paths described in the present disclosure include not only a plurality of traversal paths provided by a plurality of ants, but also traversal paths provided by one ant in some cases.

For example, ant 1 and ant 2 start path traversal from two nodes at the same time. Since the path selected by ant 1 is short, the traversal process is completed first. At this time, the traversal path of ant 1 is recorded as the first traversal path, and restarted. The path of ant 1 is traversed. If the second completion path traverses the ant 2, the traversal path of the ant 2 is recorded as the second traversal path, and the path traversal of the ant 2 is restarted. However, in some cases, the second completion path traverses the ant 1, then the traversal path of the second traversal of the ant 1 is recorded as the second traversal path, and the path traversal of the ant 1 is restarted.

To ensure that the algorithm converges, the maximum total number of loops can be set. When it is judged that the traversal path of the record reaches the maximum total number of loops, the calculation result is output according to the immediate situation or the calculation is restarted. The maximum total number of cycles only constrains the total number of traversal paths, and does not constrain which ant is provided by each ant.

2 is a flow chart of a method for improving the determination of an optimal path of the present disclosure.

Referring to FIG. 2, step S108 may include:

Step S1082: Acquire a historical optimal path and a traversal path;

Step S1084: selecting the historical optimal path and the path length in the traversal path to be the current optimal path;

Step S1086: Acquire a traversal path multiple times, and select a scheme with a small path length in both the traversal path and the current optimal path as the current optimal path;

In step S1088, the previous step is repeated n times. When it is determined that the current optimal path is selected k consecutive times, the current optimal path is determined as an optimal path, where n is a preset maximum total number of cycles.

Wherein, in step S1082, the historical optimal path includes acquiring a historical optimal path traversing the picking node according to the S-type policy method. That is, the existing optimal path obtained by the S-type policy method is taken as the ACO elite path (elite strategy). In addition, in the selection of the historical optimal path, it is recommended to select the traversal path scheme of the same lane picking order from the outside to the inside (the main channel is the reference object).

In steps S1084 to S1088, the termination condition of the algorithm may be set to a maximum number of constraints in the outer loop (the total number of traversal times of all ants), and one traversal path scheme has been selected as the current optimal path for K consecutive times. It can be judged that the algorithm has converged and the calculation is terminated.

By setting the historical optimal path as the elite path, the solution obtained by the method can be at least inferior to the solution calculated by the current algorithm, and the quality of the solution is guaranteed. At the same time, by introducing the elite path as the initial solution, the calculation amount of the method is reduced, the calculation time is reduced, and the calculation efficiency is improved. Therefore, the method improves the computational efficiency and improves the optimization quality of the scheme.

FIG. 3 schematically shows a flow chart of a picking path planning method in an exemplary embodiment of the present disclosure. Referring to FIG. 3, the picking path planning method 100 may further include:

Step S110, determining a starting point of the warehouse picking path, and modifying the optimal path according to the starting point to generate an optimal picking path.

The start and end points of the traversal path calculated by the ant colony algorithm described above may be different from the actual picking start and end points in the warehouse. Therefore, after obtaining the optimal convenience path, the actual picking start point and end point can be fine-tuned to obtain the optimal picking path.

For example, when the actual picking starting point is in the non-head-to-end node in the optimal path, the scheme with the shortest path in the two schemes may be selected with the starting point as the origin to generate a final traversal scheme. The end point, the picking of the final composite station, is related to the picking end point, so the final compounding station can be set according to the picking end point in the above traversing path, ie the end point changes as the picking path changes.

The above method 100 will be described in detail below by way of specific embodiments.

4 is a flow chart of one embodiment of the present disclosure.

Referring to FIG. 4, after receiving the picking task list, a picking path optimization scheme is obtained by the method 100 based on the distance table between the nodes. In FIG. 4, step S402 represents an asynchronous parallel strategy, step S403 represents an elite strategy, and numbers represent a picking order.

Step S401: Initialization, setting time t=0, total number of cycles Nc=0, total maximum number of cycles Ncmax=4, initial information amount of path (i, j) τ _ij (0)=1, Δτ _ij (0)= 0, the distance L _bs = 391 based on the historical optimal path formed by the original S-type strategy.

Step S402: Place 13 ants on 11 starting nodes in C (the extra 2 ants are randomly placed on 11 nodes), Nc=Nc+1.

Step S403: Perform parameter setting according to Formula 1 and Table 1, and calculate transition probability

Such as:

Calculate the probability that ant k selects node j, j∈{C-tabu ₁ }, the starting node selected for ant k.

Step S404: Select the point having the maximum state transition probability, move the ant k to the point, and record the point in the tab row tabu k of the ant _k .

Step S405: It is judged whether 11 nodes in the set C are accessed, if yes, the process goes to step S406, otherwise the process goes to step S403.

Step S406: Update the amount of information on each path that an ant passes after the end of one traversal according to formula (2) and generate a current optimal path. Such as:

In step S407, it is determined whether a traversal path scheme is consecutively selected as the current optimal path, and if yes, the calculation result is output, otherwise the taboo table is cleared and the process proceeds to step S102.

According to the total length of the optimal picking path formed by the S-type strategy of 391 m and the total length of the optimal picking path formed by the method of 297 m, the optimization rate a is calculated:

a=(391-297)/391=24%

After a large number of actual data tests, the overall optimization rate (lowering the cost of the picking path) has exceeded 8%, or even 13%, and the algorithm running time meets the actual business time requirements. A warehouse test data shows that the average optimization rate is 9%-13%, the path optimization rate is over 8%, and the solution time is up to 100ms.

The performance comparison of the algorithm is shown in Table 2. This method is generally superior to the standard ant colony algorithm in various indicators, and the average optimization rate (average optimization rate = (elite ACO path cost - original path distance cost) / original path distance cost) is higher than 6%, the algorithm is applied The rate (algorithm application rate = the total number of programs obtained by this algorithm is better than the original program / the total number of programs) is higher than 30%, the response time of the algorithm is significantly lower, and the picking efficiency is obviously improved, at 65%-72%. The picking path calculation plays a role in reducing costs.

Table 2

指标项Indicator item	本方法The method	标准蚁群算法Standard ant colony algorithm
平均优化率Average optimization rate	9％-13％9%-13%	3％-6％3%-6%
算法投用率Algorithmic rate	65％-72％65%-72%	30％-40％30%-40%
算法平均响应时间Algorithm average response time	约100msAbout 100ms	约150ms以上About 150ms or more

The picking path planning method provided by the present disclosure improves the classical ant colony algorithm, introduces an asynchronous parallel strategy into the ant colony algorithm, and solves the problem in the picking path optimization problem, thereby greatly improving the optimization effect. In addition, by introducing the elite strategy to the initial solution and process solution of the ant colony algorithm, the optimization effect is greatly improved, and the method has great practical application value.

Corresponding to the above method embodiment, the present disclosure also provides a picking path planning device, which can be used to execute the above method embodiment.

Referring to FIG. 5, the picking path planning apparatus 500 may include:

The initialization module 502 is configured to place a plurality of ants on a plurality of nodes to initiate a path traversal cycle of the ants.

The path generation module 508 is configured to calculate a transition probability of the ants transferring to other picking nodes according to the pheromone value of the path between the picking nodes, and move the ants to the node with the highest transition probability.

The pheromone update module 506 is configured to update the pheromone value of the path after the ant traverses all the nodes, and record a traversal path.

An optimal selection module 508 is arranged to determine an optimal path from the plurality of traversal paths.

In an embodiment, the method may further include:

The solution adjustment module 510 is configured to determine a starting point and an ending point of the warehouse picking path, and modify the starting point and the ending point of the optimal path according to the starting point and the ending point to generate an optimal picking path.

Since the functions of the device 500 have been described in detail in their corresponding method embodiments, the present disclosure will not be described herein.

According to an aspect of the present disclosure, a path planning apparatus is provided, including:

Memory;

A processor coupled to the associated memory, the processor being configured to perform the method of any of the preceding ones based on instructions stored in the memory.

The specific manner in which the processor of the apparatus in this embodiment performs the operation has been described in detail in the embodiment relating to the picking path planning method, and will not be explained in detail herein.

FIG. 6 is a block diagram of an apparatus 600, according to an exemplary embodiment. The device 600 may be a mobile terminal such as a smartphone or a tablet.

Referring to FIG. 6, device 600 can include one or more of the following components: processing component 602, memory 604, power component 606, multimedia component 608, audio component 610, sensor component 614, and communication component 616.

Processing component 602 typically controls the overall operation of device 600, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. Processing component 602 can include one or more processors 618 to execute instructions to perform all or part of the steps described above. Moreover, processing component 602 can include one or more modules to facilitate interaction between component 602 and other components. For example, processing component 602 can include a multimedia module to facilitate interaction between multimedia component 608 and processing component 602.

Memory 604 is configured to store various types of data to support operation at device 600. Examples of such data include instructions for any application or method operating on device 600. The memory 604 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Disk or Optical Disk. Also stored in memory 604 is one or more modules that are configured to be executed by the one or more processors 618 to perform all or part of the steps of any of the methods described above.

Power component 606 provides power to various components of device 600. Power component 606 can include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for device 600.

The multimedia component 608 includes a screen between the device 600 and the user that provides an output interface. In some embodiments, the screen can include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, slides, and gestures on the touch panel. The touch sensor may sense not only the boundary of the touch or sliding action, but also the duration and pressure associated with the touch or slide operation.

The audio component 610 is configured to output and/or input an audio signal. For example, audio component 610 includes a microphone (MIC) that is configured to receive an external audio signal when device 600 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal may be further stored in memory 604 or transmitted via communication component 616. In some embodiments, audio component 610 also includes a speaker for outputting an audio signal.

Sensor assembly 614 includes one or more sensors for providing device 600 with a status assessment of various aspects. For example, sensor assembly 614 can detect an open/closed state of device 600, relative positioning of components, and sensor assembly 614 can also detect changes in position of one component of device 600 or device 600 and temperature changes of device 600. In some embodiments, the sensor assembly 614 can also include a magnetic sensor, a pressure sensor, or a temperature sensor.

Communication component 616 is configured to facilitate wired or wireless communication between device 600 and other devices. The device 600 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, communication component 616 receives broadcast signals or broadcast associated information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 616 also includes a near field communication (NFC) module to facilitate short range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, device 600 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable A gate array (FPGA), controller, microcontroller, microprocessor, or other electronic component implementation for performing the above methods.

In an exemplary embodiment of the present disclosure, there is also provided a computer readable storage medium having stored thereon a program, the program being executed by a processor to implement the picking path planning method according to any one of the above . The computer readable storage medium can be, for example, a temporary and non-transitory computer readable storage medium including instructions.

Other embodiments of the present disclosure will be apparent to those skilled in the <RTIgt; The present application is intended to cover any variations, uses, or adaptations of the present disclosure, which are in accordance with the general principles of the disclosure and include common general knowledge or common technical means in the art that are not disclosed in the present disclosure. . The specification and examples are to be considered as illustrative only,

Industrial applicability

The picking path planning method provided by the embodiment of the present disclosure asynchronously calculates the optimal picking path solution by using the optimal picking path scheme generated based on the existing strategy as the reference elite scheme of the asynchronous parallel elite ant colony optimization algorithm. Improve the efficiency of the picking path planning while providing a better picking path planning solution. The asynchronous parallel elite ant colony optimization algorithm guarantees the solution time and optimization effect of the algorithm, and solves the problem that the optimization strategy of the existing pure strategy-based picking path planning scheme is not ideal and the time difference of the algorithm such as machine learning is poor.

Claims

A method for singular path planning based on ant colony algorithm, which comprises:

Starting with a plurality of nodes as a starting point, starting a path traversal of a plurality of ants, the node includes a plurality of picking nodes, a picking starting point, and a picking end point;

Calculating a transition probability of the ant transferring to the other node according to a pheromone value of a path between the plurality of nodes, and moving the ant to the node with the highest transition probability;

After determining that an ant traverses all the nodes, updating the pheromone value of the path and recording a traversal path;

The optimal path is determined according to a plurality of traversal paths.
The method of claim 1 according to claim 1, wherein the determining the optimal path according to the plurality of traversal paths comprises:

Obtain a historical optimal path and a traversal path;

Selecting the historical optimal path and the path length in the traversal path as the current optimal path;

Obtaining a traversal path a plurality of times, and selecting a scheme having a small path length in both the traversal path and the current optimal path as the current optimal path;

The previous step is repeated n times, and when it is determined that the current optimal path is selected consecutively k times, the current optimal path is determined as an optimal path, where n is a preset maximum total number of cycles.
The picking path planning method according to claim 2, wherein the obtaining the historical optimal path comprises acquiring a historical optimal path traversing the node according to the S-type policy method.
The method of planning a routing path according to claim 2, further comprising:

Counting the node through which the ant passes into the taboo table corresponding to the ant;

When it is determined that no path is selected as the current optimal path for consecutive k times, the tabu table and the current optimal path are cleared, and the path traversal cycle of the ant is restarted.
The method of planning a routing path according to claim 1, further comprising:

Determining the picking starting point of the warehouse picking path, and modifying the optimal path according to the picking starting point to generate an optimal picking path.
The picking path planning method according to claim 1, wherein said calculating a transition probability of said ant transferring to said other node comprises calculating by the following formula:

among them,
The transition probability of the ant k moving to the jth node on the i-th node, t is the number of times the path travels, and C is a set of all nodes.
For the node where ant k has passed, d ij is the distance between two nodes i and j, α is the pheromone index with the value range [0, 5], and β is the value range [0, 5] The heuristic factor index, τ ij is the current pheromone value of the path between nodes i and j.
The picking path planning method according to claim 1, wherein the pheromone value of the update path is updated by the following formula:

τ ij (t+n)=(1-ρ)·τ ij (t)+Δτ ij

Where τ ij is the current pheromone value of the path between nodes i and j, and Δτ ij represents the updated pheromone increment,
Indicates the total length of the path taken by the kth ant in this cycle, Q represents the total amount of pheromone released by the ant cycle over the path, γ represents the weight parameter, T bs represents the current optimal path; L bs represents The length of the current optimal path, ρ represents the evaporation coefficient.
A picking path planning device, comprising:

The initialization module is configured to start a path traversal of multiple ants starting from a plurality of nodes, where the node includes multiple picking nodes, a picking starting point, and a picking end point;

a path generating module, configured to calculate, according to a pheromone value of a path between the plurality of nodes, a transition probability of the ant to transfer to another node, and move the ant to the node with the highest transition probability;

a pheromone update module, configured to: after an ant traverses all the nodes, update a pheromone value of the path, and record a traversal path;

An optimal selection module is arranged to determine an optimal path based on the plurality of traversal paths.
A picking path planning device, comprising:

Memory;

A processor coupled to the associated memory, the processor being configured to perform the picking path planning method of any of claims 1-7 based on instructions stored in the memory.
A computer readable storage medium having stored thereon a program, the program being executed by a processor to implement the picking path planning method according to any one of claims 1-7.