CN109872001A - Unmanned vehicle method for allocating tasks based on K-means and discrete particle cluster algorithm - Google Patents
Unmanned vehicle method for allocating tasks based on K-means and discrete particle cluster algorithm Download PDFInfo
- Publication number
- CN109872001A CN109872001A CN201910150788.7A CN201910150788A CN109872001A CN 109872001 A CN109872001 A CN 109872001A CN 201910150788 A CN201910150788 A CN 201910150788A CN 109872001 A CN109872001 A CN 109872001A
- Authority
- CN
- China
- Prior art keywords
- task
- packet
- unmanned vehicle
- algorithm
- packaged
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a kind of more unmanned vehicle method for allocating tasks based on K-means and discrete particle cluster algorithm, comprising the following steps: S1, logistics scene information initializing；S2, logistics task are packaged: determining best be packaged as a result, it is k that it, which is packaged number, using K-means algorithm；S3, k available unmanned vehicle is taken, matches the execution unmanned vehicle of each task packet；S4, to each, nobody automobile-used discrete particle cluster algorithm determines its task sequence.Method combination Clustering and gunz optimization algorithm of the invention answers more unmanned vehicle Task Allocation Problems, it is distributed using the task that the particle swarm algorithm of discretization carries out more unmanned vehicles, the fast convergence rate of particle swarm algorithm, the iterative manner of discretization make algorithm be more suitable for actual logistics scene.And first task is packaged with K-means clustering algorithm before using particle swarm algorithm, greatly reduces the size of solution space, improves task allocative efficiency.
Description
Technical field
The present invention relates to logistics distributions to optimize field, and in particular to the method for allocating tasks in a kind of unmanned logistics field.
Background technique
As the scale of construction of e-commerce industry constantly expands and the continuous improvement of cost of labor, what dispatching was operated as logistics
Basic link, due to the continuous expansion of its coverage rate, data update is more frequent, and dispensing task is increasingly heavy, more has " in delivery
The raising of the service request of door ", wisdom logistics are increasingly prevailing, and Chinese logistics have turned on the unmanned epoch.
The storage efficiency in unmanned storehouse can reach the several times of traditional crossbeam racks store efficiency.According to the letter of certain electric business logistic statistics
Breath, the sorting ability of unmanned sortation hubs can achieve 9000/hour, for 4 times of improved efficiency of packet link, same
Under the premise of place scale and sorting goods amount, each place can save 180 people of manpower.Unmanned plane can be by traditional artificial dispatching
Time shortens several times or even dozens of times, and logistics cost also decreases.But it is due to depositing the inability of delivery in inclement weather, flight
Process not can avoid the disadvantages of artificial destruction, at this stage can not also high-volume it is commercial.And show unmanned vehicle then in Recent experimental
6 express deliveries can be disposably sent, primary 80 kilometers of charging continuation of the journey, unmanned plane dispatching disadvantage can be made up, produce in enormous quantities and use.With
Logistics Modernization and intelligentized development, domestic market for unmanned logistics demand by further expansion.To be objective, nothing
People's vehicle systematic difference field is quite wide, as long as almost there is cargo to need environment that is continuous or repeating transport that can use nothing
People's vehicle system.The scheduling of single unmanned vehicle can no longer meet existing huge logistics capacity, more unmanned vehicles in unmanned logistical applications
Intelligent dispatching system belongs to complicated dynamic, multiple target discrete system, when map density is higher and unmanned vehicle quantity is more, out
The probability that now a large amount of delays wait is also higher.It realizes in mutually coordinated cooperation between unmanned vehicle in uncertain environment (such as on road)
To reduce out the conflict occurred when the operation tasks such as storage execute, become more important realistic objective.
So, it is desirable to reach the hope of " last 1 kilometer " using more unmanned vehicles, the Task Allocation Problem of more unmanned vehicles is ground
Study carefully most important.What more unmanned vehicle distribution aspects mainly used at present has particle swarm algorithm, ant group algorithm, simple scanning method, divides
Branch defines method, simulated annealing, dynamic programming algorithm etc., but these current algorithms all have some disadvantages, and specifically include that
The problems such as low with the compatible degree of practical stream background, convergence rate is slow, and solving result is not necessarily optimal solution, and operation time is long.
Therefore, it is necessary to a kind of more efficient more unmanned vehicle distribution methods.
Summary of the invention
Goal of the invention: in view of the deficiencies of the prior art, the present invention proposes a kind of based on K-means and discrete particle cluster algorithm
More unmanned vehicle method for allocating tasks, realize the distribution of more practical and efficient unmanned vehicle task.
Technical solution: the invention proposes it is a kind of be packaged tactful (K-means) and discrete particle cluster algorithm mostly without
People's vehicle method for allocating tasks, comprising:
S1, logistics scene information initializing；
S2, logistics task are packaged: determining best be packaged as a result, it is k that it, which is packaged number, using K-means algorithm；
S3, k available unmanned vehicle is taken, matches the execution unmanned vehicle of each task packet；
S4, to each, nobody automobile-used discrete particle cluster algorithm determines its task sequence.
Wherein, step S1 includes mission bit stream initialization, warehouse information initialization and unmanned vehicle information initializing.
Step S2 includes:
S21, task point quantity and available unmanned vehicle quantity are obtained, determines the task packet the upper limit of the number for needing to be packaged；
S22, the packing result that k task packet is realized using K-means algorithm；
S23, calculated according to the packing result of k task packet its class external distance, in class away from and similarity, take the smallest similarity
Value corresponding k value and packing scheme, wherein class external distance is distance value the sum of of each task packet center to full task packet center,
Away from the sum of the distance value for task point contained in each task packet to the task packet center in class, similarity is in class external distance and class
The sum of away from.
Particle rapidity and position, the value of calculating target function, so that it is determined that making nobody are updated by discretization in step S4
Vehicle is paid a price the smallest task sequence.
The utility model has the advantages that
1, method combination Clustering of the invention and gunz optimization algorithm solve more unmanned vehicle Task Allocation Problems
It answers, this method is for the purpose of improving task allocation algorithms efficiency, to meet practical logistics scene, is calculated using the population of discretization
Method carries out the task distribution of more unmanned vehicles, and the fast convergence rate of particle swarm algorithm, the iterative manner of discretization keeps algorithm more applicable
In actual stream background.
2, the present invention is first packaged task with K-means clustering algorithm before using particle swarm algorithm, greatly subtracts
The small size of solution space, improves task allocative efficiency.
Detailed description of the invention
Fig. 1 is the overall flow figure of more unmanned vehicle method for allocating tasks based on K-means and discrete particle cluster algorithm；
Fig. 2 is the flow chart that best packing scheme is determined using K-means algorithm；
Fig. 3 is the flow chart that packing once scheme is executed using K-means algorithm；
Fig. 4 is the flow chart of discrete particle cluster algorithm of the invention.
Specific embodiment
Technical solution of the present invention is described further with reference to the accompanying drawing.
More unmanned vehicle method for allocating tasks based on K-means and discrete particle cluster algorithm task distribution before, first with
Task object is packaged by K-means cluster with suitable task packet quantity, to multiple tasks packet discrete particle cluster algorithm
The distribution of more unmanned vehicles is carried out, Fig. 1 gives the overall flow of the method for the present invention.
Step S1 initializes logistics scene information.
It include: each warehouse information Storehouse (x, y, n, id) of load, wherein Storehouse.x indicates the ground in warehouse
Manage dimension, Storehouse.y indicates the geographic logitude in warehouse, Storehouse.n indicates warehouse residue number of items,
The number in Storehouse.id expression warehouse；
Each unmanned vehicle information Car (x, y, v, f, id, visible) is loaded, wherein Car.x indicates the geographical dimension of unmanned vehicle
Degree, Car.y indicate that the geographic logitude of unmanned vehicle, Car.v indicate that the running speed of unmanned vehicle, Car.f indicate the residue of unmanned vehicle
Useable fuel amount, Car.id indicate that the number of unmanned vehicle, Car.visible indicate the working condition of unmanned vehicle, when unmanned vehicle still
In the task of execution, Car.visible=0, when unmanned vehicle is in idle condition, Car.visible=1；
Each task point information Task (x, y, id) is loaded, wherein Task.x indicates the geographic latitude of task point, Task.y table
Show that the geographic logitude of task point, Task.id indicate the number of task point, when the Task (x, y) of two task points is all the same,
Task.id is smaller, and priority is higher.
Object in each warehouse is transported to specified goal task point, each nothing using more unmanned vehicles by logistics scene
People's vehicle returns to warehouse after completing dispatching task, and the geography information in warehouse is for calculating task at a distance from warehouse, and unmanned vehicle is herein
Longitude and latitude, id and speed (speed is used to calculate the running time from one place to a task point) are only used in distribution method,
The priority of task herein it also will be understood that at same position different task, other information is mainly used for the graphical of project
Expression and the operation of specific example.
Step S2 determines best be packaged as a result, it is k that it, which is packaged number, using K-means algorithm.
Mainly according to the relative position of task object point and at a distance from warehouse, task is divided into an appropriate number of task
Packet.Task packet is understood that be packaged task for the set of the task point marked off according to certain condition, this set is appointed
Business specifies a certain vehicle to complete, other vehicles are not involved in.
Referring to Fig. 2, specific steps include:
Step S2.1: obtaining task point quantity n_Task and available unmanned vehicle quantity n_Car, determines the needing to be packaged of the task
Packet the upper limit of the number kmax；
It is packaged number upper limit kmaxIt is determined by both task point quantity and available unmanned vehicle quantity.Rule of thumb Fa Ke get is packaged
Number k meetsTherefore, when available unmanned vehicle quantity is more, being packaged the number upper limit can be obtained with empirical method, when
When can be few with unmanned vehicle quantity, it be packaged the number upper limit and be determined by available unmanned vehicle quantity, be then packaged number upper limit kmaxMeet as follows
Formula:
Wherein, kmaxExpression task is packaged the upper limit value of number, n_Task indicates the total quantity of task point, when the same target
When point has multiple tasks, task point consistent with multiple longitudes and latitudes but different mission number is indicated, n_Car indicates that unmanned vehicle can be used
Sum.
Step S2.2: cyclic variable k=1 is enabled；
Step S2.3: for each k=i, the packing result of k task packet is realized using K-means algorithm, wherein
As shown in figure 3, specific step is as follows for K-means algorithm:
Step 1: the packet center a of k task packet of random initializtion1(x1,y1),a2(x2,y2),...,ak(xk,yk), Bao Zhong
The heart refers herein to the longitude and latitude average value of all task points in this packet.
Step 2: it for each task point Task (x, y, id), is classified as apart from nearest task packet center aj(xj,
yj) where task packet j in, which meets following formula:
Wherein, labelidIt is divided for the packet of task point Task (x, y, id), aj(xj,yj) indicate j-th of task packet center, i
One longitude of a latitude is indicated from 1 to 2, is range formula in min bracket, min is exactly to do one for all task
A such operation, finds one apart from nearest aj, sub, which just refers to, seeks thisjSuch a operation of subscript j.
Step 3: the packet center a of each task packet is updatedj(xj,yj), the attribute value (i.e. longitude and latitude) at the center Xin Bao is to be subordinate to
The attribute value mean value for belonging to all task points of the task packet, meets following formula:
Wherein, aj(xj,yj) the new packet center of expression, cjIt is new task packet division, n_cjRefer to new task packet cjIt is interior
Contained task point number.
Step 4: when the small Mr. Yu's given value of each packet center change rate, or when reaching maximum number of iterations, terminate this dozen
Packet, conversely, returning to step 2.
Packet center change rate is the order of magnitude for describing the variation in each task packet geographical location, updates packet center each time
When, the longitude and latitude at each packet center can accordingly change, for example change to (59.9,129.99) from (60,130), variation
Δ x=0.1, Δ y=0.01, here it is considered that change rate is equal to 0.1, the maximum value for taking the two to change, i.e. formula can be write
At max { Δ x, Δ y }.When this variation is small arrives acceptable, it can think that this packing has been completed.
Step S2.4: judge if (k≤kmax), if logical value is 0, terminate all packings, export it is all be packaged as a result,
Conversely, then k++ and return step S2.3.
Step S2.5: full task packet center a is calculatedtotal(xtotal,ytotal)
Wherein, atotalIndicate the center of all task points, TaskidThe longitude and latitude for indicating the task of the id, that is, pass through longitude and latitude
Degree finds out the center point of all tasks in this packet to be averaged；The quantity of n_Task expression task point.
Step S2.6: result is packaged to the k task packet that each k=i is obtained using K-means algorithm respectively and calculates it
Class external distance, i.e. each task packet center meet following formula to the sum of the distance value at full task packet center:
Wherein, L indicates class external distance, aiIndicate i-th packet center, atotalIndicate full task packet center.Here it takes absolutely
Subtracting each other for value indicates to seek the distance between two points.
Step S2.7: result is packaged to the k task packet that each k=i is obtained using K-means algorithm respectively and calculates it
Away from contained task point meets following formula to the sum of the distance value at the task packet center in that is, each task packet in class:
Wherein, D is indicated in class away from, aiIndicate i-th packet center, CiIndicate i-th of task packet.
Step S2.8: result is packaged to the k task packet that each k=i is obtained using K-means algorithm respectively and calculates it
In Distance conformability degree function, i.e. class external distance and class away from the sum of, meet following formula:
Wherein, F (S, k) indicates this Distance conformability degree functional value being packaged, and the value is smaller, then it is better to be packaged result.
Step S2.9: to save current preferably packing as a result, enabling temp=F (S, kmax), k=kmax-1；
Step S2.10: judging if (temp < F (S, k)), if logical value is 0, illustrates the packing result of k task packet
Than current best result also than get well, then update temp=F (S, k)；
Step S2.11:k=k-1；
Step S2.12: step S2.10, S2.11 is repeated, until k=0；
Step S2.13: record temp corresponding k value and packing scheme.
Step S3 takes k available unmanned vehicle, and the corresponding of each task packet of random fit executes unmanned vehicle, if nobody can be used
When there are many vehicle quantity, k is randomly selected.
Other than random fit, each unmanned vehicle can also be made to possess its unordered claimed by distance with regard near match
Business queue.
Step S4, to each, nobody automobile-used discrete particle cluster algorithm determines its task sequence, then determines by sequence
Sequence executes delivery task.
Confirm that task sequence specific steps include:
Step S4.1: cyclic variable i=1 is enabled；
Step S4.2: for i-th of task packet Ci, determine that task packet correspondence executes nobody using discrete particle cluster algorithm
The task execution sequence of vehicle, wherein as shown in figure 4, specific step is as follows for discrete particle cluster algorithm:
Step 1: a certain number of particle X are generated at randomt, constitute initialization population, XtIllustrate task in task packet
A kind of arrangement mode of point；
Step 2: judge whether the number of iterations is more than its upper limit, if being more than, export the task execution of the execution unmanned vehicle
Sequence；
Step 3: obtaining the task point quantity N of the task packet, updates Pbest according to fitness functiontAnd Gbestt,
PbesttIt is individual extreme value, refers to the desired positions of each particle, GbesttIt is all extreme values, refers to all particles most
Good position, fitness function are exactly following two function, so that the two functions is reached the target that minimum is for we, adapt at this time
Degree can be regarded as executing the cost of unmanned vehicle, then can have following two targets:
Wherein, runi,jIndicate that i-th of stage drives to the duration of task point j, Xi,jIt indicates whether i-th of stage executes to appoint
Be engaged in j, Xi,j=1 is to execute task j, X in i-th of stagei,j=0 i-th of stage of expression did not executed task j, delayi,jIndicate the
I stage executes the time delay of j-th of task point, performi,jIndicate that i-th of stage executes the execution time of j-th of task, mesh
Mark a minJ1It is minimum to can be understood as overall travel time, two minJ of target2It can be understood as always executing task time most short.
To use discrete particle cluster algorithm to solve target, single goal is converted then for above-mentioned multiple target with multiplication and division are as follows:
MinJ=J1J2
Step 4: more new particle goes to step 2, and to guarantee that the update of particle meets actual scene, and task is by whole
It executes, designs new discretization particle update mode, discretization particle update mode meets following expression:
Wherein, VtThe ring shift right vector being randomly generated carries out ring shift right behaviour to task sequence according to the vector
Make, PbesttIt is individual extreme value, GbesttBe global extremum,Circulation needed for finger is converted to individual extreme value is right
Duration set is shifted to,Similarly, R1、R2Respectively indicate randomness selection ring shift right vector.Last each car meeting
Obtain his a task execution sequence.
Step S4.3: judging if (i≤k), if logical value is 0, exports task allocation result, conversely, then i++ and returning
Step S4.2.
The randomness selection of ordinary particle swarm algorithm is shown multiplied by two random decimals, but this method and discomfort
For discrete variable, background of the invention is the unmanned vehicle task execution sequence in logistics, and each number is all integer in sequence,
The case where must assure that after iteration also in integer is not suitable for original particle swarm algorithm at this time, therefore can be execution sequence
Column are arranged new discretization and update, and this update mode is operated based on ring shift right, and the number that ensure that in sequence is done so
Certainty or the feature of integer after iteration, and this itself is a arrangement problems, the arrangement problems present invention uses exchange position
The mode (being described here with local circulation right-shift operation) set solves more to meet daily logic.Additionally due to conventional particle group
Algorithm iteration speed due to solution space is excessive is slow, and the present invention while iterative process discretization of population considering to have used k-
Means algorithm is packaged task, and task is divided into task packing in this way and determines two step of task execution sequence, sees handle on surface
One step becomes two steps, can greatly reduce solution space actually, accelerates calculating speed.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (7)
1. a kind of more unmanned vehicle method for allocating tasks based on K-means and discrete particle cluster algorithm, which is characterized in that the side
Method the following steps are included:
S1, logistics scene information initializing；
S2, logistics task are packaged: determining best be packaged as a result, it is k that it, which is packaged number, using K-means algorithm；
S3, k available unmanned vehicle is taken, matches the execution unmanned vehicle of each task packet；
S4, to each, nobody automobile-used discrete particle cluster algorithm determines its task sequence.
2. more unmanned vehicle method for allocating tasks according to claim 1 based on K-means and discrete particle cluster algorithm,
It is characterized in that, the step S1 includes mission bit stream initialization, warehouse information initialization and unmanned vehicle information initializing.
3. more unmanned vehicle method for allocating tasks according to claim 1 based on K-means and discrete particle cluster algorithm,
It is characterized in that, the step S2 includes:
S21, task point quantity and available unmanned vehicle quantity are obtained, determines the task packet the upper limit of the number for needing to be packaged；
S22, the packing result that k task packet is realized using K-means algorithm；
S23, calculated according to the packing result of k task packet its class external distance, in class away from and similarity, take the smallest similarity value institute
Corresponding k value and packing scheme, wherein class external distance is distance value the sum of of each task packet center to full task packet center, in class
Away from the sum of the distance value for task point contained in each task packet to the task packet center, similarity is in class external distance and class away from it
With.
4. more unmanned vehicle method for allocating tasks according to claim 3 based on K-means and discrete particle cluster algorithm,
It is characterized in that, the step S22 includes:
Step 1: the packet center a of k task packet of random initializtion1(x1,y1),a2(x2,y2),...,ak(xk,yk)；
Step 2: it for each task point Task (x, y, id), is classified as apart from nearest task packet center aj(xj,yj) institute
Task packet j in, which meets following formula:
Wherein, labelidThe geographic latitude of task point, Task.y are indicated for the packet division of task point Task (x, y, id), Task.x
The geographic logitude of expression task point, Task.id indicate the number of task point, aj(xj,yj) indicate j-th of task packet center, i from 1
Refer to 2 expression one latitude one longitude, sub and seeks thisjSubscript j；
Step 3: the packet center a of each task packet is updatedj(xj,yj), the attribute value at the center Xin Bao is the institute for being subordinate to the task packet
There is the attribute value mean value of task point, meet following formula:
Wherein, aj(xj,yj) the new packet center of expression, cjIt is new task packet division, n_cjRefer to new task packet cjIt is interior contained
Task point number；
Step 4: when the small Mr. Yu's given value of each packet center change rate, or when reaching maximum number of iterations, terminate this packing, instead
It, returns to step 2.
5. more unmanned vehicle method for allocating tasks according to claim 1 based on K-means and discrete particle cluster algorithm,
It is characterized in that, the step S4 updates particle rapidity and position, the value of calculating target function, so that it is determined that making nothing by discretization
People's vehicle is paid a price the smallest task sequence.
6. more unmanned vehicle method for allocating tasks according to claim 5 based on K-means and discrete particle cluster algorithm,
It is characterized in that, the objective function includes:
Wherein, runi,jIndicate that i-th of stage drives to the duration of task point j, Xi,jIndicate i-th of stage whether execute task j,
delayi,jIndicate that i-th of stage executes the time delay of j-th of task point, performi,jIndicate that i-th of stage executes j-th of task
The execution time.
7. more unmanned vehicle method for allocating tasks according to claim 5 based on K-means and discrete particle cluster algorithm,
It is characterized in that, the discretization updates particle rapidity and position according to following more new formula:
Wherein, XtFor particle, the current arrangement mode of task point in task packet, V are indicatedtThe ring shift right vector being randomly generated,
Ring shift right operation, Pbest are carried out to task sequence according to the vectortIt is individual extreme value, GbesttIt is global extremum,Ring shift right vector set needed for finger is converted to individual extreme value,Finger is converted to the overall situation
Ring shift right vector set, R needed for extreme value1、R2Respectively indicate randomness selection ring shift right vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910150788.7A CN109872001B (en) | 2019-02-28 | 2019-02-28 | Unmanned vehicle task allocation method based on K-means and discrete particle swarm algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910150788.7A CN109872001B (en) | 2019-02-28 | 2019-02-28 | Unmanned vehicle task allocation method based on K-means and discrete particle swarm algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109872001A true CN109872001A (en) | 2019-06-11 |
| CN109872001B CN109872001B (en) | 2021-03-19 |
Family
ID=66919541
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910150788.7A Active CN109872001B (en) | 2019-02-28 | 2019-02-28 | Unmanned vehicle task allocation method based on K-means and discrete particle swarm algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109872001B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738619A (en) * | 2020-07-06 | 2020-10-02 | 腾讯科技（深圳）有限公司 | Task scheduling method, device, equipment and storage medium |
| CN112035552A (en) * | 2020-09-02 | 2020-12-04 | 国网河南省电力公司电力科学研究院 | Voltage sag severity prediction method and device based on association rule |
| CN112965528A (en) * | 2021-02-22 | 2021-06-15 | 广东电网有限责任公司 | Rescue strategy determination method, device, equipment and storage medium for disaster-affected point |
| CN113269480A (en) * | 2020-02-17 | 2021-08-17 | 百度在线网络技术（北京）有限公司 | Task allocation path determining method and device and electronic equipment |
| CN115065683A (en) * | 2022-07-27 | 2022-09-16 | 多伦科技股份有限公司 | Vehicle edge network task allocation and unloading method based on vehicle clustering |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6952700B2 (en) * | 2001-03-22 | 2005-10-04 | International Business Machines Corporation | Feature weighting in κ-means clustering |
| CN102831474A (en) * | 2012-08-06 | 2012-12-19 | 江南大学 | Improved fuzzy C-mean clustering method based on quantum particle swarm optimization |
| CN106990792A (en) * | 2017-05-23 | 2017-07-28 | 西北工业大学 | Mix the multiple no-manned plane collaboration sequential coupling task distribution method of gravitation search algorithm |
| CN107093046A (en) * | 2017-04-21 | 2017-08-25 | 北京京东尚科信息技术有限公司 | Unmanned dispensing vehicle method for allocating tasks, system and unmanned dispensing vehicle |
| CN108932598A (en) * | 2018-05-31 | 2018-12-04 | 广东工业大学 | A kind of grand scale logistic allocator based on cloud platform |
| CN109063779A (en) * | 2018-08-09 | 2018-12-21 | 河海大学常州校区 | A kind of cloud manufacturing recourses cluster k-means clustering method |
-
2019
- 2019-02-28 CN CN201910150788.7A patent/CN109872001B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6952700B2 (en) * | 2001-03-22 | 2005-10-04 | International Business Machines Corporation | Feature weighting in κ-means clustering |
| CN102831474A (en) * | 2012-08-06 | 2012-12-19 | 江南大学 | Improved fuzzy C-mean clustering method based on quantum particle swarm optimization |
| CN107093046A (en) * | 2017-04-21 | 2017-08-25 | 北京京东尚科信息技术有限公司 | Unmanned dispensing vehicle method for allocating tasks, system and unmanned dispensing vehicle |
| CN106990792A (en) * | 2017-05-23 | 2017-07-28 | 西北工业大学 | Mix the multiple no-manned plane collaboration sequential coupling task distribution method of gravitation search algorithm |
| CN108932598A (en) * | 2018-05-31 | 2018-12-04 | 广东工业大学 | A kind of grand scale logistic allocator based on cloud platform |
| CN109063779A (en) * | 2018-08-09 | 2018-12-21 | 河海大学常州校区 | A kind of cloud manufacturing recourses cluster k-means clustering method |
Non-Patent Citations (2)
| Title |
|---|
| 王强等: "UAV协同任务分配的改进DPSO算法仿真研究", 《系统仿真学报》 * |
| 陈媛媛等: "一种初值优化的K-均值文档聚类算法(英文)", 《江西师范大学学报(自然科学版)》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113269480A (en) * | 2020-02-17 | 2021-08-17 | 百度在线网络技术（北京）有限公司 | Task allocation path determining method and device and electronic equipment |
| CN113269480B (en) * | 2020-02-17 | 2022-06-14 | 百度在线网络技术（北京）有限公司 | Task allocation path determining method and device and electronic equipment |
| CN111738619A (en) * | 2020-07-06 | 2020-10-02 | 腾讯科技（深圳）有限公司 | Task scheduling method, device, equipment and storage medium |
| CN111738619B (en) * | 2020-07-06 | 2023-11-07 | 腾讯科技（深圳）有限公司 | Task scheduling method, device, equipment and storage medium |
| CN112035552A (en) * | 2020-09-02 | 2020-12-04 | 国网河南省电力公司电力科学研究院 | Voltage sag severity prediction method and device based on association rule |
| CN112035552B (en) * | 2020-09-02 | 2023-05-09 | 国网河南省电力公司电力科学研究院 | Method and device for predicting severity of voltage sag based on association rule |
| CN112965528A (en) * | 2021-02-22 | 2021-06-15 | 广东电网有限责任公司 | Rescue strategy determination method, device, equipment and storage medium for disaster-affected point |
| CN112965528B (en) * | 2021-02-22 | 2023-06-30 | 广东电网有限责任公司 | Rescue strategy determination method, device and equipment for disaster points and storage medium |
| CN115065683A (en) * | 2022-07-27 | 2022-09-16 | 多伦科技股份有限公司 | Vehicle edge network task allocation and unloading method based on vehicle clustering |
| CN115065683B (en) * | 2022-07-27 | 2023-12-26 | 多伦互联网技术有限公司 | Vehicle edge network task allocation and unloading method based on vehicle clustering |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109872001B (en) | 2021-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109872001A (en) | Unmanned vehicle method for allocating tasks based on K-means and discrete particle cluster algorithm | |
| CN113011644B (en) | Smart city dynamic cold-chain logistics scheduling method based on ant colony optimization algorithm | |
| CN109214756B (en) | Vehicle logistics scheduling method and device, storage medium and terminal | |
| CN108764634A (en) | A kind of electric automobile charging station dynamic programming method for considering charge requirement and increasing | |
| CN109345091B (en) | Ant colony algorithm-based whole vehicle logistics scheduling method and device, storage medium and terminal | |
| Luo et al. | A last-mile drone-assisted one-to-one pickup and delivery problem with multi-visit drone trips | |
| CN112907103B (en) | Method for sharing dynamic supply and demand balance of single vehicle | |
| CN105631530A (en) | Multiple sequential planning and allocation of time-divisible resources | |
| CN107862403A (en) | A kind of the outbound sequence dispatching method and system of unmanned plane goods to be dispensed | |
| CN106063067A (en) | Electricity-demand prediction device, electricity supply system, electricity-demand prediction method, and program | |
| CN106709680A (en) | Method for optimizing optimal distribution route in dynamic logistics based on historical search information | |
| CN116228089B (en) | Store distribution path planning method based on shortest mileage | |
| CN109191052A (en) | A kind of multi-vehicle-type vehicle routing optimization method, server and system | |
| Li et al. | An agent-based simulation for shared automated electric vehicles with vehicle relocation | |
| CN113706081B (en) | Unmanned aerial vehicle picking and delivering system and method based on automatic express delivery device on city roof | |
| CN114897307A (en) | Logistics centralized distribution network alliance optimization method based on cooperation | |
| CN109615260A (en) | The method for determining the installation addresses of charging pile | |
| CN115062868B (en) | Pre-polymerization type vehicle distribution path planning method and device | |
| CN112785029A (en) | Charging station power consumption prediction method based on deep clustering model | |
| CN106295864A (en) | A kind of method solving single home-delivery center logistics transportation scheduling problem | |
| CN114529241A (en) | Highway fractional freight logistics path planning algorithm based on operational research theory | |
| CN110264140A (en) | A kind of logistics transportation dispatching method, device and equipment with time window | |
| CN107153891A (en) | Zone-transfer dispatching method, device, equipment and storage medium | |
| CN117726040A (en) | Distribution vehicle dynamic scheduling optimization method based on DDQN algorithm | |
| WO2021144577A1 (en) | Vehicle routing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |