CN109711624A - Packing method, equipment and computer readable storage medium - Google Patents
Packing method, equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN109711624A CN109711624A CN201811626116.0A CN201811626116A CN109711624A CN 109711624 A CN109711624 A CN 109711624A CN 201811626116 A CN201811626116 A CN 201811626116A CN 109711624 A CN109711624 A CN 109711624A
- Authority
- CN
- China
- Prior art keywords
- cargo
- information
- load
- packing method
- layer
- 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)
Abstract
The present invention discloses a kind of packing method, equipment and computer readable storage medium, and the packing method includes: the receiving information for obtaining container and the bin information of cargo to be installed；According to the receiving information and the bin information, and is searched for according to Monte Carlo tree and obtain vanning scheme；During searching for acquisition vanning scheme according to Monte Carlo tree, reward function is obtained according to space utilization rate.The present invention has the effect of improving vanning versatility and computational efficiency.
Description
Technical field
The present invention relates to logistics to sort field, in particular to packing method, equipment and computer readable storage medium.
Background technique
Vanning carries out man power encasement usually in accordance with experience at present.In order to improve vanning reasonability and efficiently
Property, it can assist casing using intelligent algorithm.
Bin packing is a classical academic problem, while possessing extensive commercial application value.In logistics field
In, it usually will appear and need a series of specified cargos being packed into specified containers, such as compartment, the problem of transport.
Existing most algorithms can be divided into two classes, and (one) uses the heuristic search algorithm (two) artificially to lay down a regulation
The optimization algorithm such as genetic algorithm that nonlinear optimization is regarded as to solve, with deep learning algorithm.
But the essential defect based on heuristic search of the first kind is its result dependent on artificially specified inspiration
Formula rule.It turns for the better in regular where applicable result, otherwise rule is difficult to obtain available scheme.However most of vanning scenes itself
There is complicated restrictive condition, this makes us be difficult to find out a set of applicable rule.Secondly, whenever main change occurs for scene
When, rule must reformulate adjustment, this affects the versatility of algorithm itself.
Second class method, nonlinear optimization-genetic algorithm and deep learning algorithm, major defect have two: (one) lacks
Support to complex space physical constraint.These restrictive conditions make traditional algorithm be difficult to handle；(2) this kind of non-thread
There has been no enough theories integrations for the mechanism of property optimization algorithm.The person of often requiring to use consumes plenty of time adjusting parameter to obtain
Qualified result.Therefore percent of automatization is lower, constrains popularization and use of such algorithm in industry vanning scene.
Summary of the invention
The main object of the present invention is to provide packing method, equipment and computer readable storage medium, it is intended to improve vanning
Versatility and computational efficiency.
To achieve the above object, a kind of packing method proposed by the present invention sorts, the packing method packet for logistics
It includes:
Obtain the receiving information of container and the bin information of cargo to be installed；
According to the receiving information and the bin information, and is searched for according to Monte Carlo tree and obtain vanning scheme；
During searching for acquisition vanning scheme according to Monte Carlo tree, reward function is obtained according to space utilization rate.
Optionally, described to accommodate the shape and position letter that information includes the shape of the container and the object accommodated
Breath；
The bin information includes the shape, weight and load-bearing information of the cargo to be installed；
The vanning scheme includes each cabinet position finally in a reservoir and orientation information.
Optionally, the packing method further include:
In carrying out the simulation process in the search of Monte Carlo tree, the expansion of a tree node of first layer meets load-bearing limitation
In the case where, carry out the expansion of the relevant second layer tree node of the tree node；Otherwise the exhibition of another tree node of first layer is executed
It opens.
Optionally, the packing method further include:
During searching for acquisition vanning scheme according to Monte Carlo tree, pass through the stacking pass that AABB tree records cargo
System, and judge whether each cargo meets load-bearing limitation.
Optionally, the layered relationship that cargo is recorded by AABB tree, and judge whether each cargo meets load-bearing
Limitation, comprising:
In newly-increased cargo, the supporter that the bottom of newly-increased cargo directly contacts is obtained；
According to the contact relation of the weight of newly-increased cargo and supporter, the load-bearing information of each supporter is updated.
Optionally, the basis increases the weight of cargo and the contact relation of supporter newly, updates the load-bearing of each supporter
Information, comprising:
The newly-increased cargo is connected to upper one layer of above support, obtains the newly-increased cargo and each support
The contact area of object, and the gross contact area with all above supports；
In the weight or load-bearing information change of upper one layer of cargo, the supporter of this layer updates load-bearing according to the variation
Information；Also,
This layer of supporter according to the contact area with upper one layer of cargo, the gross contact area for the cargo that mono- layer of Zhan Shang
Ratio shares the variation；Wherein,
The lowest level supporter is the supporting surface of the container.
Optionally, the packing method further include:
In carrying out the simulation process in the search of Monte Carlo tree, the expansion of a tree node of first layer meets space limitation
In the case where, carry out the expansion of the relevant second layer tree node of the tree node；Otherwise the exhibition of another tree node of first layer is executed
It opens；
In simulation process, by placing cargo in the space of discretization, and judge whether newly-increased cargo meets sky
Between limit.
The present invention also provides a kind of packing methods, sort for logistics, and the packing method includes:
The receiving information of container is obtained, and waits the bin information of the cargo loaded；
When the volume of the container is greater than preset value, existing container is divided into multiple regions；
It is obtained one by one according to the receiving information and the bin information, and according to the search of Monte Carlo tree, Ge Gekong
Between each vanning scheme；
When searching for the vanning scheme in a space, according to the appearance of the vanning scheme more new container in other spaces obtained before
Receive information.
The present invention also provides a kind of boxing apparatus, the boxing apparatus includes reservoir and processor；The reservoir
Store vanning program；
The processor is for executing the vanning program, to the step of executing above-mentioned packing method.
The present invention also provides a kind of computer readable storage medium, dress is stored on the computer readable storage medium
Case program is realized when the vanning program is executed by processor such as the step of above-mentioned packing method.
Packing method provided by the present invention, it is at work, then defeated by pre-set space utilization rate as reward function
Enter the receiving information of container and the bin information of cargo to be installed.Then Monte Carlo tree search algorithm executes simulation vanning, i.e.,
Constantly by carrying out the iteration of Action-State, the direction of simulation vanning is obtained.It is arranged finally by according to space utilization rate
Reward function formula calculate the simulation vanning direction reward function, according to it is different vanning directions multiple reward functions, come
Optimal vanning direction is selected, then carries out iteration next time again, so that optimal vanning direction next time is obtained, until
Simulation installs all chests, to obtain optimal vanning scheme.Therefore, the present invention, can by Monte Carlo tree search algorithm
Universally and automatically to solve, the more excellent solution of vanning between cargo and container to be installed is solved, to instruct artificial or machine
People cases.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
The structure shown according to these attached drawings obtains other attached drawings.
Fig. 1 is the flow chart of packing method first embodiment of the present invention；
Fig. 2 is packing method calculating process schematic diagram shown in FIG. 1；
Fig. 3 is the flow chart of packing method second embodiment of the present invention；
Fig. 4 is the flow chart of packing method 3rd embodiment of the present invention；
Fig. 5 is the flow chart of packing method fourth embodiment of the present invention；
Fig. 6 is the flow chart of the 5th embodiment of packing method of the present invention；
Fig. 7 is the flow chart of packing method sixth embodiment of the present invention；
Fig. 8 is packing method calculating process schematic diagram shown in Fig. 7.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
In present embodiment, vanning simulation is carried out by Monte Carlo tree searching method and is calculated, to by a system
Arranging, there is the cargo of different shape and physical characteristic to be put into specified containers, and meet certain space and limitation requirement physically.
Such as: space requirement may include the restrictive condition of charging ratio and other spaces, such as 10 lis of container of distance vanning
The barrier of the similar such as built-in corner fittings of container is had existed in rice or container.Desired physical considerations include chest load-bearing rule.
Load-bearing control may include that the practical load-bearing of a cargo in the structure must not exceed the maximum capacity of its calibration.
Or 90% etc. of practical load-bearing not higher than the maximum capacity that it is demarcated of control cargo.
Include: by the process that Monte Carlo tree searching method calculates vanning
Firstly, the calculating process of Monte Carlo tree searching method is set.
Then, the input of algorithm can be with are as follows:
It a. include the length, width and height of target container,
It b. can also include the shape and location information of already present object inside target container；
It c. can also include the weight and load-bearing information of existing object；
It d. include the length, width and height information to cased goods；
It e. can also include the weight information and load-bearing information to cased goods；
Finally, the output of algorithm can be with are as follows:
It f. include the location information of each cargo finally in a reservoir；
It g. can also include each cargo towards Orientation information.
It is searched for after calculation method by having set Monte Carlo tree, it is only necessary to the information of input pod and the letter of vanning
Breath, then can calculate, which kind of vanning scheme is best.It should be noted that the preferred plan calculated and Monte Carlo
Pre-set reward function is related in tree search algorithm.
For example, the space utilization rate of setting container as reward function, is then, full by calculating obtained vanning scheme
When under sufficient preset condition (space or load-bearing/space and load-bearing), the highest scheme of space utilization rate.It is, of course, also possible to be arranged
The load-bearing utilization rate of container is as reward function.Then scheme obtained is at this time, and meeting preset condition, (space is held
Weight/space and load-bearing) under when, the highest scheme of load-bearing utilization rate.It is, of course, also possible to multiple reward functions be arranged, such as together
When setting container space utilization rate and load-bearing utilization rate as reward function, the two elements respectively account for fifty percent weight.
It may include: emulation Action and state State about the calculating process of Monte Carlo tree search；
In the search of Monte Carlo tree, each state State corresponding is already present structure in current spatial, comprising
Cargo through placing, and the cargo that do not place also.
Emulation Action in the search of Monte Carlo tree may include the information of the cargo to be placed, wherein goods information
It may include the material of cargo, the density of cargo or the shape of cargo.Or goods information can be in one embodiment
Direction including cargo, the position that will be placed.
I.e. during carrying out the search of Monte Carlo tree, Action is executed in original State state, thus full
Emulation is put into chest in the case where the limitation of sufficient space and load-bearing, is often put into a chest and is known as expanding one level.Firstly, emulation is arrived
Predetermined expansion level, such as emulation are put into 10 chests；Or when emulation to expansion the last layer, i.e. expansion is put into the last one
Chest to be put.Then, it returns and calculates original State should select to execute which chest be put into, is i.e. selection executes the first floor
In the sub- state of which State1.The process of the selection are as follows: carry out for the sub- state of each State1 about specified reward letter
Several calculating, then with the calculated result of reward function, to choose an optimal sub- state of State1.It finally determines and executes
To the sub- state of the State1, so that the sub- state of the State1 becomes new State state.Then it recycles and executes again
Action, then obtain another new State.It moves in circles, constantly carries out emulation and iteration, until it is put into last chest,
Or virtually it is put into the desired physical considerations that any chest is all unable to satisfy setting.
In summary, the iterative operation of Monte Carlo tree algorithm is divided into four steps:
Step 1: choosing, an original state state is chosen.Choose an executable tree node, tree node lower section
It can have the child node of several second layers.
Step 2: tree node is unfolded, i.e., the tree node S1 chosen from the first step carries out the expansion of tree node, to produce
A raw new child node S11.Child node S11 is one in the second layer node of link node S1.
Step 3: emulation carries out the subsequent operation for being put into chest at random of exploratory execution since child node S11, from
And obtain subsequent deeper node.Such as the node S111, S112, S113 etc. of emulation expansion third layer；Such as from
The 4th layer of node S1111, S1112, S1113 etc. of node S111 emulation expansion, from the 4th layer of node S112 emulation expansion
Node S1121, S1122, S1123 etc..
When executing these simulation operations, the result of each step emulation is needed all to meet set physical constraint condition.
Remaining chest is randomly selected in the exploratory emulation, and is put into container one by one, and the quantity of chest is put into, can be until container
Until space cannot place into any chest, it can not perhaps meet physical constraint condition again or system resource can support
Until the maximum emulation expansion number of plies, etc..
Step 4: evaluation, the reward function of this route of node S11 is calculated according to reward function formula.
Above four steps are the process of an iteration.
While presetting secondary iteration, the reward for obtaining several child node S11 of the second layer of original state state is calculated
Function, the reward function of S12, reward function of S13 etc..According to above-mentioned reward function, an optimal child node is executed,
Form new state.
If being judged using space utilization rate as reward function, in same container, the total volume of the cargo of loading is accounted for
Iteration more maximum than the volume of container is as optimal solution.
When executing above-mentioned Action, about the calculating of space limitation, can be expressed by discrete space representation
The space of container, the then position that cabinet can be put will select in the space of discretization.For example, discrete space representation can wrap
It includes: extreme point method Extreme Points or gridding method Grids etc..By carrying out discretization expression to space, so as to
Continuous space is become into the array by sequence, from calculation amount can be reduced, increases computational efficiency.
When executing above-mentioned Action, about the calculating of load-bearing limitation, can be recorded by AABB-Tree structure every
The relativeness up and down of one cargo.Then when every Action is used, the Action cargo increased newly is increased into AABB-
In Tree structure, and calculate new load-bearing.
In order to further decrease calculation amount, can also be realized by adjusting the volume of container.For example, by dividing and ruling
The strategy of method Divide-and-Conquer solves the problems, such as this, it may be assumed that
Point -- by PROBLEM DECOMPOSITION be the smaller subproblem of scale；
Control -- the smaller subproblem of these scales is broken up one by one；
Close -- settled subproblem is merged, finally obtains the solution of " mother " problem.
When implementing, multiple regions can be divided the container into, the optimal solution in each region is then calculated separately, finally will
All optimal solutions merge, to obtain the vanning optimal solution of whole container.
The thought of vanning is realized based on above-mentioned Monte Carlo tree searching method, proposes each embodiment of the method for the present invention.
Embodiment one
Present embodiments provide packing method first embodiment.
Fig. 1 is please referred to, a kind of packing method is sorted for logistics, and the packing method includes:
Step S101 obtains the receiving information of container and the bin information of cargo to be installed；
Step S102 is filled according to the receiving information and the bin information, and according to the search of Monte Carlo tree
Case scheme；
Step S103 is obtained during searching for acquisition vanning scheme according to Monte Carlo tree according to space utilization rate
Reward function.
In the present embodiment, the receiving information of container and the bin information of cargo to be installed are obtained first.Wherein, container
Receiving information include container length, width and height, and internal already existing Item Information.For example, leaning on angle information, or
Internally placed chest information etc..The bin information of cargo to be installed may include the ID of cargo to be installed, weight, volume, be
It is no to have the requirement placed upward, if to have load-bearing limitation etc..
In the present embodiment, in the acquisition receiving information and bin information and then according to the receiving information and institute
Bin information is stated, and is searched for according to Monte Carlo tree and obtains vanning scheme.Wherein, information and bin information are accommodated as input
Information.The calculation method of Monte Carlo tree search, which carries out constantly simulating according to random number, places chest.Pass through Monte Carlo tree
The calculating process of search, the final and optimal state finally obtained, as vanning scheme.
In the present embodiment, during searching for acquisition vanning scheme according to Monte Carlo tree, according to space utilization rate
Obtain reward function.Wherein, the reward function of Monte Carlo tree search is obtained according to space utilization rate, is used to Simulation and is cased
Iteration each time in the process, finally using scheme composed by a succession of optimal iteration as vanning scheme.For example, passing through meter
The ratio between the total volume of chest and the volume of container being packed into is calculated, to obtain the space of the different node of same layer
Utilization rate, for example, 80%~98%, 70%~95%, 85%~92%, 70%~80% etc.；Wherein, space utilization rate
Up to 98%, then it can be using the node of corresponding 80%~98% space utilization rate as virtual vanning direction, to obtain one
New State；Again from the new State, the highest node of space utilization rate in next layer is obtained, the node is as another
New State；Finally in a series of optimal selection, vanning scheme is obtained.
After obtaining vanning scheme, artificial or robot can control according to vanning scheme, to be one by one packed into chest
In container.
Packing method provided by the present embodiment obtains reward function by pre-set space utilization rate, then at work
The receiving information of input pod and the bin information of cargo to be installed.Then Monte Carlo tree search algorithm carries out simulation vanning,
The direction of simulation vanning constantly is obtained by carrying out the iteration of Action-State.It is set finally by according to space utilization rate
Set the reward function that reward function formula calculates simulation vanning direction.According to multiple reward functions in different vanning directions, come
Optimal vanning direction is selected, then carries out iteration next time again, so that optimal vanning direction next time is obtained, until
Simulation installs all chests, to obtain optimal vanning scheme.Therefore, the present embodiment passes through Monte Carlo tree search algorithm,
It can universally and automatically solve, solve the more excellent solution of vanning between cargo and container to be installed, to instruct artificial or machine
Device people cases.
Further, in the present embodiment, the object for accommodating information and including the shape of the container and having accommodated
Shape and location information.To packing method provided in this embodiment, part Very Important Cargo manually first can be put into container,
Then it is calculated again.Or carry out relaying calculating etc..
Further, in the present embodiment, the bin information includes shape, weight and the load-bearing letter of the cargo to be installed
Breath.It, can be according to the load-bearing information of each cargo, thus to each step Action to packing method provided by the present embodiment
Carry out the calculating of load-bearing limitation.
Further, in the present embodiment, the vanning scheme includes each cabinet position finally in a reservoir and court
To information.It, can be when carrying out each step Action to packing method provided by embodiment, it can be to cargo towards not
It is simulated when equidirectional, increases the diversity of vanning, so that it is truer to simulate obtained result.
Further, in the present embodiment, the packing method further include:
Step S104, in carrying out the simulation process in the search of Monte Carlo tree, the expansion of a tree node of first layer is accorded with
In the case where closing load-bearing limitation, the expansion of the relevant second layer tree node of the tree node is carried out；Otherwise the another of first layer is executed
The expansion of tree node.Wherein, the first layer and the second layer are merely representative of overlying relation, and do not limit specifically in entirety
Specific layer.
As shown in Fig. 2, there are 5 chests at this time, need to be put into container.Wherein:
In the state of State00000, container is sky, needs for cargo to be installed to be fitted into container.At this point, cargo needs to fill
The first position entered, the first position are a preset positions, such as the specified corner in inside of container.Have at present
5 cargos to be installed.Since each chest can have the direction of six kinds of postures.Therefore, when being packed into first position, Ke Yicong
State00000 root node sets out, and is arbitrarily unfolded one in 5 × 6=30 child node.It puts in the corner formulated into container
Set chest 1, chest 2, chest 3, chest 4 or chest 5, every kind of chest can be put with any one in six kinds of postures
It sets.
Herein, in order to make explanation more succinct, each cargo is set and is only capable of with a kind of placement of posture.To
When being packed into first position, the child node that can be unfolded is reduced to 5.
From State00000 root node, start first time iteration: selecting any child node exhibition in 5 child nodes at random
The case where opening, and emulating the subsequent node of expansion, load-bearing limitation is carried out in one node of every expansion judges.If emulating the son of expansion
Node meets the limitation of load-bearing, then can continue any sub- child node that the child node is unfolded, to constantly execute emulation.?
Emulating no any child node can satisfy load-bearing limitation, or when discharging all chests, calculates and return to first
The reward function of the child node of first expansion in secondary iteration.Then it is set out, is started second with State00000 root node again
Iteration, from remaining 4 child nodes, optional child node expansion, then return reward function is emulated and is calculated, thus
Complete second of iteration.Until reaching the iteration of preset times, or 5 iteration are carried out, that is, State00000 root has been calculated
The reward function of 5 child nodes of node.
If the value for calculating the reward function that first position is put into chest 4 at this time is maximum, first position is chosen to be put into chest
4, and form new State --- State40000.
Under State40000 state, chest 1, chest 2, chest 3 and chest 5 can be packed into container at this time, it can be with
The position of loading includes the second position, the third place, the 4th position and the 5th position, at this time since variable quantity will be excessively difficult to
It is bright.In order to concise explanation, the position of vanning is fixed at this time.I.e. subsequent Action all will according to the second position,
The third place, the 4th position and the 5th position progress sequence are put.
Then, in the state of State40000,4 iteration can be executed, if executing the iterative process for being put into chest 1
In, the expansion emulated to a certain layer tree node is unsatisfactory for load-bearing limitation.The iteration about chest 1 is then interrupted, then executes and is put into case
The iteration of son 2.If execution is put into the iterative process of chest 2, the expansion emulated to a certain layer tree node is unsatisfactory for load-bearing limitation.
The iteration about chest 2 is then interrupted, then executes the iteration for being put into chest 3.And it is put into the iteration of chest 3 and is put into changing for chest 5
During generation, emulates all meet load-bearing limitation each time, then obtain and be put into chest 3 and be put into the reward function of chest 5.Lead at this time
Cross comparison, it is believed that the value for being put into the reward function of chest 3 is bigger.It then chooses the second position to be put into chest 3, and is formed newly
State——State43000。
And so on, in subsequent emulation and iterative process, choose the third place to be put into chest 2, and formed newly
State——State43200.The 4th position is chosen to be put into chest 5, the 5th position is put into chest 1, and ultimately forms new
State——State43251。
Packing method provided by the present embodiment carries out load-bearing limit by each step emulation in iterative process each time
The calculating of system thereby reduces the search of Monte Carlo tree so that subsequent simulation can just be continued by meeting the emulation of load-bearing limitation
Calculation amount, improve the efficiency of algorithm, reduce the waiting time.
Certainly, in other embodiments, can also be when an iteration in progress Monte Carlo tree search finish, then sentence
Whether the emulation in the iteration of breaking meets load-bearing limitation.For example, being put into chest 4 about first position for above-mentioned example
Iteration finish after, building is sequentially placed into the overlay structure of chest 4, chest 3, chest 2, chest 5 and chest 1, then carries out
The judgement of load-bearing limitation, then obtaining judging result is to meet load-bearing limitation.And after another an iteration finishes, building is sequentially placed into
Chest 4, chest 2, chest 3, chest 5 and chest 1 overlay structure, carry out load-bearing limitation judgement, then obtaining judging result is
Load-bearing limitation is not met.The emulation for then meeting load-bearing limitation participates in obtaining reward function, and does not meet the emulation of load-bearing limitation then
It is not involved in acquisition reward function.
Embodiment two
Fig. 3 is please referred to, a kind of packing method is present embodiments provided.The present embodiment is based on the above embodiment, additional to increase
Step.It is specific as follows:
The packing method further include:
Step S205 passes through AABB tree and records cargo during searching for acquisition vanning scheme according to Monte Carlo tree
Layered relationship, and judge each cargo whether meet load-bearing limitation.
Other steps of the present embodiment are same as the previously described embodiments, specifically may refer to above-described embodiment, no longer superfluous herein
It states.
In the present embodiment, during searching for acquisition vanning scheme according to Monte Carlo tree, pass through AABB tree and record
The layered relationship of cargo, and judge whether each cargo meets load-bearing limitation.Wherein, AABB (Axis Aligned
Bounding Box, the parallel bounding box of axis) structure be parallel to reference axis encirclement object minimum cuboid.AABB tree is base
In the hierarchical structure binary tree of AABB structure building.The layered relationship of chest is recorded by AABB tree, is then executing emulation
When, newly-increased cargo is increased in AABB tree, a new node is generated.Then it is closed by the stacking between cargo in AABB tree
It is to calculate whether each cargo meets load-bearing limitation.
In the present embodiment, AABB tree can be introduced, and carried out sentencing for load-bearing during being emulated
It is disconnected.It will be interrupted in time to not meet the emulation of load-bearing limitation, to reduce the amount of calculating, and the iteration result obtained is not
In the presence of the risk for not meeting load-bearing limitation.Specifically, when can be each tree node of expansion, all to corresponding to the tree node
State carries out load-bearing judgement；Alternatively, being also possible in the tree node expansion for executing two steps, then to the tree node of second step expansion
Corresponding state carries out load-bearing judgement；Alternatively, can also be in the tree node expansion for executing N step, then to N step exhibition
State corresponding to the tree node opened carries out load-bearing judgement.
The present embodiment by using the layered relationship of AABB tree record cargo, then judges each cargo in simulation vanning
Whether load-bearing limitation is met.To which the description method of AABB tree be utilized, so that cargo is more convenient for, mathematicization is handled, and calculates load-bearing
And geometry splicing (geometry splicing is exactly crash tests, and chest is gone to simulate by collision model, for example, chest A and chest B it
Before have a gap, it is desirable to a chest C is filled in, but the length of chest C or width etc. be greater than gap when, lead to
Crossing crash tests and can be obtained by chest C cannot be placed between chest A and chest B) when the property be also more convenient for using binary tree
It is calculated to carry out traversal load-bearing.To effect simple, stable and efficient with calculating process.
Embodiment three
Fig. 4 is please referred to, a kind of packing method is present embodiments provided.The present embodiment based on the above embodiment, and to wherein
The step of further illustrated.It is specific as follows:
The layered relationship that cargo is recorded by AABB tree, and judge whether each cargo meets load-bearing limitation, packet
It includes:
Step S301 obtains the supporter that the bottom of newly-increased cargo directly contacts in newly-increased cargo；
Step S302 updates the load-bearing letter of each supporter according to the contact relation of the weight of newly-increased cargo and supporter
Breath.
Other steps of the present embodiment are same as the previously described embodiments, specifically may refer to above-described embodiment, no longer superfluous herein
It states.
In the present embodiment, it when node is unfolded in emulation each time in an iterative process, i.e., in newly-increased cargo, obtains new
The bottom for increasing cargo directly contacts supporter.Wherein, emulation is directed to newly-increased Item Number each time, this newly-increased cargo body
Product, weight and load-bearing information, the location information etc. that cargo will be put.The bottom for obtaining newly-increased cargo by analysis is direct
The weight of the supporter of contact, the newly-increased cargo will directly affect the load-bearing information of above support.
In the present embodiment, in the information of acquisition supporter and then according to the weight of newly-increased cargo and connecing for supporter
Touching relationship updates the load-bearing information of each supporter.Wherein, contact relation, can be whether the chest directly contacted is newly-increased
Main support, the Auxiliary support of chest? directly contact the accounting etc. of the relatively total bearing area of bearing area of chest.Therefore,
When increasing newly-increased cargo, corresponding weight can be obtained by the contact relation and shares relationship.Then according to weight point
The weight of newly-increased cargo is updated on each supporter by booth relationship.Due to, in record layered relationship, being recorded by AABB tree
When each cargo, the load-bearing limitation and current load-bearing information of the cargo also will record.To need to only obtain each chest
Weight share after relationship, newly-increased weight of sharing is increased on each supporter.
The load-bearing information of the supporter directly contacted is updated according to the weight for increasing cargo newly.It is straight that this place is merely illustrative update
The load-bearing information of the supporter of contact.It is understood that if being also present in the supporter under above support and directly connecing
Other supporters of touching, then the load-bearing information change of the supporter, will also conduct to other supporters.So with continuous
The weight of two-stage conduction, the newly-increased cargo will be updated to always tree root, to complete the update of all load-bearing information.
Packing method disclosed in the present embodiment executes load-bearing when executing and emulating by AABB tree and updates, thus
Can judge in time emulation whether meet load-bearing limitation, so as to interrupt in time do not meet load-bearing limitation emulation continue to hold
Row, and then whole operand is reduced, system resource is saved, the waiting time is shortened.
Further, which is optimized using the representation method of non-interconnected digraph.Specifically, each cargo is in figure
A node, the side of node A to node B indicates that A supports B.By the non-interconnected digraph representation method, algorithm can be by wide
Degree first traversal quickly updates cabinet load-bearing.
Example IV
Fig. 5 is please referred to, a kind of packing method is present embodiments provided.The present embodiment based on the above embodiment, and to wherein
The step of further illustrated.It is specific as follows:
The basis increases the weight of cargo newly and directly contacts the contact relation of cargo, updates the load-bearing letter of each supporter
Breath, comprising:
The newly-increased cargo is connected to upper one layer of above support, and obtains the newly-increased cargo by step S401
With the contact area of each above support, and the gross contact area with all above supports；
Step S402, in the weight or load-bearing information change of upper one layer of cargo, the supporter of this layer is according to the change
Change and updates load-bearing information；Also,
This layer of supporter according to the contact area with upper one layer of cargo, the gross contact area for the cargo that mono- layer of Zhan Shang
Ratio shares the variation；Wherein,
The lowest level supporter is the supporting surface of the container.
Other steps of the present embodiment are same as the previously described embodiments, specifically may refer to above-described embodiment, no longer superfluous herein
It states.
In the present embodiment, when executing emulation, the newly-increased cargo is connected to upper one layer of above support, and
Obtain the contact area of the newly-increased cargo and each above support, and the gross contact area with all above supports.
Wherein, upper one layer that cargo is connected to the supporter directly contacted in AABB tree is increased newly.Then increases newly and connect again on cargo at this
When connecing other cargos, the load-bearing of the newly-increased cargo generates variation for being influenced by other cargos, should so that being located at
The load-bearing information of the supporter of newly-increased cargo lower layer also accordingly generates variation.Above-mentioned contact area is all the ground of newly-increased cargo
On the area that is supported.For example, the supporter directly contacted there are two the lower sections of newly-increased cargo, and respective contact area
For 10cm2, then gross contact area is 20cm2；Upper one layer of two supporters of the newly-increased cargo as below.
In the present embodiment, contact area and gross contact area and then execution are being obtained: in the weight of upper one layer of cargo
When amount or load-bearing information change, the supporter of this layer updates load-bearing information according to the variation；Also, this layer of supporter according to
With the contact area of upper one layer of cargo, the ratio of the gross contact area for the cargo that mono- layer of Zhan Shang shares the variation, wherein
The lowest level supporter is the supporting surface of the container.Wherein, the case where weight of upper one layer of the cargo changes
It is suitable for, upper one layer of supporter when having increased the newly-increased cargo newly.The load-bearing information of upper one layer of the cargo changes
The case where be suitable for, when the load-bearing information change of upper one layer of cargo of supporter.The weight of i.e. newly-increased cargo will be directly changed tree
On next layer of cargo load-bearing information, and indirectly change tree on more lower cargo load-bearing information；The mistake of the influence
Journey carries out in a manner of successively transmitting, and passes up to tree location of root.
Specifically, the ratio that weight is shared is as follows: if upper one layer of a supporter has increased cargo or upper one layer of goods newly
The load-bearing information change of object, then the weight or the load-bearing information of variation will be conducted to the supporter.Firstly, obtain supporter with
The contact area of upper one layer of the cargo and upper one layer cargo and area in the contact of all supporting goods；Then with
The contact area accounts for the ratio of gross contact area, which shares the weight or load-bearing information change of one layer of the cargo
Amount.
Such as:
As shown in figure 8, upper one layer of supporter H has newly-increased cargo A, the weight of the newly-increased cargo A is 10kg；Wherein, should
The contact area that supporter H contacts the newly-increased cargo A is 10 square centimeters, and newly-increased cargo A and all supporter T's of lower layer connects
Touching the gross area is 50 square centimeters；Therefore, supporter H will share the weight of 10kg*10/50=2kg.The 2kg weight will increase
It adds in the load-bearing information of the supporter.And remaining 8kg will share to other supporters.
The load-bearing information of upper one layer of cargo of supporter H produces variation, so that the weight of supporter H increases 2kg；Its
In, the direct contact area of supporter H and next one cargo B of layer are 40 square centimeters, supporter H and all of lower layer
The gross contact area for supportting object is 50 square centimeters；Therefore, one cargo B of next layer will share the weight of 2kg*40/50=1.6kg
Amount.The 1.6kg weight will be increased in the load-bearing information of one cargo B of next layer.And remaining 0.4kg will share to lower layer its
On his supporter.
Packing method provided by the present embodiment establishes newly-increased cargo on AABB tree first in newly-increased cargo, and
And it establishes on the upper layer of supporter.Then, then when there are weight and load-bearing information change, according to sequence from top to bottom, by
Layer transmits and shares weight, shares always to tree location of root.Finally, using the supporting surface of the container as lowest level supporter,
That is tree root position.Then whole weight will all be recorded to the supporting surface of container, to judge whether the weight of entire cargo surpasses
Cross the load-bearing limitation of container.To packing method provided by the present embodiment, by using the mode of bearing area accounting come by
Layer shares weight, then has reasonable, the fireballing effect of Computation for apportionment of sharing weight.
Embodiment five
Fig. 6 is please referred to, a kind of packing method is present embodiments provided.The present embodiment based on the above embodiment, and additionally increases
Step is added.It is specific as follows:
The packing method further include:
Step S501, in carrying out the simulation process in the search of Monte Carlo tree, the expansion of a tree node of first layer is accorded with
In the case where closing space limitation, the expansion of the relevant second layer tree node of the tree node is carried out；Otherwise the another of first layer is executed
The expansion of tree node；
Step S502 by placing cargo in the space of discretization, and judges that newly-increased cargo is in simulation process
It is no to meet space limitation.
Other steps of the present embodiment are same as the previously described embodiments, specifically may refer to above-described embodiment, no longer superfluous herein
It states.
In the present embodiment, in carrying out the simulation process in the search of Monte Carlo tree, the exhibition of a tree node of first layer
It opens in the case where meeting space limitation, carries out the expansion generation of the relevant second layer tree node of the tree node；Otherwise first layer is executed
Another tree node expansion.Wherein, the limitation in space can be by judging whether the cumulative length, width and height of cargo are higher than volume
Length, width and height judge.During one tree node of emulation expansion each time, that is, carry out the judgement of cargo length, width and height；If meeting empty
Between limit, then can emulate the next tree node of expansion.If being unsatisfactory for space limitation, the tree node of the expansion is abandoned, and is held
Row others tree node expansion.
In the present embodiment, in simulation process, when judging whether to meet space limitation, by by cargo in discretization
Space is placed, and judges whether newly-increased cargo meets space limitation.Wherein, by discretization expression of space, so as to drop
The low occupancy to resource can simplify calculating, improve computational efficiency.
Embodiment six
Present embodiments provide a kind of packing method.
Fig. 7 is please referred to, a kind of packing method is sorted for logistics, and the packing method includes:
Step S601 obtains the receiving information of container, and waits the bin information of the cargo loaded；
Existing container is divided into multiple regions when the volume of the container is greater than preset value by step S602；
Step S603 is obtained one by one according to the receiving information and the bin information, and according to the search of Monte Carlo tree
, each vanning scheme in each space；
Step S604, when searching for the vanning scheme in a space, more according to the vanning scheme in other spaces obtained before
The receiving information of new container.
In the present embodiment, the receiving information of container is obtained first, and waits the bin information of the cargo loaded.Its
In, as described above, the receiving information of container includes the length, width and height of container, and internal already existing Item Information.For example, leaning on
Angle information, or have been placed in internal chest information etc..The bin information of cargo to be installed may include cargo to be installed
ID, weight, volume, whether there is the requirement placed upward, if having load-bearing limitation etc..
In the present embodiment, the receiving information and bin information are being obtained and then is being greater than in the volume of the container
When preset value, existing container is divided into multiple regions.Wherein, preset value can be preset, such as preset value is 0.2m3
When 1m3 or 2m3 etc..When volume is more than preset value, either four, two regions, three regions can be divided the container into
Region etc..
In the present embodiment, multiple regions are being divided the container into and then are being believed according to the receiving information and the vanning
Breath, and obtained one by one according to the search of Monte Carlo tree, each vanning scheme in each space.Wherein, according to Monte Carlo tree
It, can be according to the space utilization rate of container, and/or according to the load-bearing utilization rate of container, to obtain reward function when search.And
During according to the search of Monte Carlo tree, to the judgement that space limitation and load-bearing limit, so that the search search of Monte Carlo tree
Process meet space limitation and/or load-bearing limitation.The judgement for specifically how carrying out space limitation judgement and load-bearing limitation can be with
Referring to the above embodiments, details are not described herein.After being searched for by Monte Carlo tree, it can export any one region
Vanning scheme.As shown in figure 8, container is divided into two regions, it is when freighting first to region 1, and region 1 being filled
Only.
In the present embodiment, in the vanning scheme for obtaining a region, then when searching for the vanning scheme in another space, according to
The receiving information of the vanning scheme more new container in other spaces obtained before.Wherein, by the dress in other regions obtained before
The set of case scheme, starting State when being calculated as new space.As shown in figure 8, when region 1 is filled, then zoning 2
Vanning, calculate when filled using region 1 when as region 2 starting State.
Packing method provided by the present embodiment forms multiple zonules, then right by the way that bulk container to be split
Each zonule is cased one by one, so that stochastic arithmetic amount when Monte Carlo tree is searched for, avoids optional space excessive, and
Lead to the calculating overlong time of each step.
Embodiment seven
Present embodiments provide a kind of boxing apparatus.
The boxing apparatus includes reservoir and processor；The reservoir stores vanning program；
The processor is for executing the vanning program, to execute packing method described in any of the above-described embodiment
The step of.
Since the present embodiment has all technical characteristics of above-mentioned packing method, the present embodiment also has above-mentioned vanning
Beneficial effect possessed by method.Above-described embodiment specifically is please referred to, details are not described herein.
Embodiment eight
Present embodiments provide a kind of computer readable storage medium.
Vanning program is stored on the computer readable storage medium, the vanning program is executed by processor above-mentioned
The step of packing method described in one.
Since the present embodiment has all technical characteristics of above-mentioned packing method, the present embodiment also has above-mentioned vanning
Beneficial effect possessed by method.Above-described embodiment specifically is please referred to, details are not described herein.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the device that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or device institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or device.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in a storage medium
In (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that a terminal (can be mobile phone, computer, service
Device, air conditioner or network equipment etc.) execute method described in each embodiment of the present invention.
The embodiment of the present invention is described with above attached drawing, but the invention is not limited to above-mentioned specific
Embodiment, the above mentioned embodiment is only schematical, rather than restrictive, those skilled in the art
Under the inspiration of the present invention, without breaking away from the scope protected by the purposes and claims of the present invention, it can also make very much
Form, all of these belong to the protection of the present invention.
Claims (10)
1. a kind of packing method, sorted for logistics, which is characterized in that the packing method includes:
Obtain the receiving information of container and the bin information of cargo to be installed；
According to the receiving information and the bin information, and is searched for according to Monte Carlo tree and obtain vanning scheme；
During searching for acquisition vanning scheme according to Monte Carlo tree, reward function is obtained according to space utilization rate.
2. packing method as described in claim 1, which is characterized in that
It is described to accommodate the shape and location information that information includes the shape of the container and the object accommodated；
The bin information includes the shape, weight and load-bearing information of the cargo to be installed；
The vanning scheme includes each cabinet position finally in a reservoir and orientation information.
3. packing method as claimed in claim 2, which is characterized in that the packing method further include:
In carrying out the simulation process in the search of Monte Carlo tree, the expansion of a tree node of first layer meets the feelings of load-bearing limitation
Under condition, the expansion of the relevant second layer tree node of the tree node is carried out；Otherwise the expansion of another tree node of first layer is executed.
4. packing method as claimed in claim 3, which is characterized in that the packing method further include:
During searching for acquisition vanning scheme according to Monte Carlo tree, the layered relationship of cargo is recorded by AABB tree, and
And judge whether each cargo meets load-bearing limitation.
5. packing method as claimed in claim 4, which is characterized in that the layered relationship that cargo is recorded by AABB tree,
And judge whether each cargo meets load-bearing limitation, comprising:
In newly-increased cargo, the supporter that the bottom of newly-increased cargo directly contacts is obtained；
According to the contact relation of the weight of newly-increased cargo and supporter, the load-bearing information of each supporter is updated.
6. packing method as claimed in claim 5, which is characterized in that the basis increases the weight of cargo and connecing for supporter newly
Touching relationship updates the load-bearing information of each supporter, comprising:
The newly-increased cargo is connected to upper one layer of above support, obtains the newly-increased cargo and each above support
Contact area, and the gross contact area with all above supports；
In the weight or load-bearing information change of upper one layer of cargo, the supporter of this layer updates load-bearing letter according to the variation
Breath；Also,
This layer of supporter is according to the contact area with upper one layer of cargo, the ratio of the gross contact area for the cargo that mono- layer of Zhan Shang
Value, shares the variation；Wherein,
The lowest level supporter is the supporting surface of the container.
7. packing method as claimed in claim 2, which is characterized in that the packing method further include:
In carrying out the simulation process in the search of Monte Carlo tree, the expansion of a tree node of first layer meets the feelings of space limitation
Under condition, the expansion of the relevant second layer tree node of the tree node is carried out；Otherwise the expansion of another tree node of first layer is executed；
In simulation process, by placing cargo in the space of discretization, and judge whether newly-increased cargo meets space limit
System.
8. a kind of packing method, sorted for logistics, which is characterized in that the packing method includes:
The receiving information of container is obtained, and waits the bin information of the cargo loaded；
When the volume of the container is greater than preset value, existing container is divided into multiple regions；
It is obtained one by one according to the receiving information and the bin information, and according to the search of Monte Carlo tree, each space
Each vanning scheme；
When searching for the vanning scheme in a space, believed according to the receiving of the vanning scheme more new container in other spaces obtained before
Breath.
9. a kind of boxing apparatus, which is characterized in that the boxing apparatus includes reservoir and processor；The reservoir stores
Vanning program；
The processor is for executing the vanning program, to execute the described in any item vanning sides of the claims 1 to 8
The step of method.
10. a kind of computer readable storage medium, which is characterized in that be stored with vanning journey on the computer readable storage medium
The step of sequence, the vanning program realizes packing method as claimed in any one of claims 1 to 8 when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811626116.0A CN109711624B (en) | 2018-12-28 | 2018-12-28 | Boxing method, equipment and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811626116.0A CN109711624B (en) | 2018-12-28 | 2018-12-28 | Boxing method, equipment and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109711624A true CN109711624A (en) | 2019-05-03 |
| CN109711624B CN109711624B (en) | 2021-06-11 |
Family
ID=66259168
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811626116.0A Active CN109711624B (en) | 2018-12-28 | 2018-12-28 | Boxing method, equipment and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109711624B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111125878A (en) * | 2019-11-20 | 2020-05-08 | 深圳蓝胖子机器人有限公司 | Boxing method, system and computer readable storage medium |
| CN117114524A (en) * | 2023-10-23 | 2023-11-24 | 香港中文大学（深圳） | Logistics sorting method based on reinforcement learning and digital twin |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101381033A (en) * | 2008-10-15 | 2009-03-11 | 江苏科技大学 | Container loading method based on ant colony algorithm |
| CN101957945A (en) * | 2010-08-20 | 2011-01-26 | 上海电机学院 | Method and device for optimizing goods loading of container |
| CN102841961A (en) * | 2012-07-20 | 2012-12-26 | 中南大学 | Method for detecting three-dimensional lifting dynamic collision based on cache information |
| CN104680237A (en) * | 2015-03-10 | 2015-06-03 | 西南科技大学 | Three-dimensional encasement novel genetic algorithm model under multi-constrain condition |
| CN105279629A (en) * | 2015-09-30 | 2016-01-27 | 上海东源计算机自动化工程有限公司 | Optimized intelligent packaging system |
| CN108015767A (en) * | 2017-11-30 | 2018-05-11 | 北京邮电大学 | A kind of space manipulator emergency operating device |
| CN108985458A (en) * | 2018-07-23 | 2018-12-11 | 东北大学 | A kind of double tree monte carlo search algorithms of sequential synchronous game |
-
2018
- 2018-12-28 CN CN201811626116.0A patent/CN109711624B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101381033A (en) * | 2008-10-15 | 2009-03-11 | 江苏科技大学 | Container loading method based on ant colony algorithm |
| CN101957945A (en) * | 2010-08-20 | 2011-01-26 | 上海电机学院 | Method and device for optimizing goods loading of container |
| CN102841961A (en) * | 2012-07-20 | 2012-12-26 | 中南大学 | Method for detecting three-dimensional lifting dynamic collision based on cache information |
| CN104680237A (en) * | 2015-03-10 | 2015-06-03 | 西南科技大学 | Three-dimensional encasement novel genetic algorithm model under multi-constrain condition |
| CN105279629A (en) * | 2015-09-30 | 2016-01-27 | 上海东源计算机自动化工程有限公司 | Optimized intelligent packaging system |
| CN108015767A (en) * | 2017-11-30 | 2018-05-11 | 北京邮电大学 | A kind of space manipulator emergency operating device |
| CN108985458A (en) * | 2018-07-23 | 2018-12-11 | 东北大学 | A kind of double tree monte carlo search algorithms of sequential synchronous game |
Non-Patent Citations (2)
| Title |
|---|
| ALEXANDRE LATERRE等: "Ranked Reward: Enabling Self-Play Reinforcement", 《ARXIV：1807.01672V3》 * |
| 王丽君: "基于Web模式的3D装箱系统可视化关键技术及应用研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111125878A (en) * | 2019-11-20 | 2020-05-08 | 深圳蓝胖子机器人有限公司 | Boxing method, system and computer readable storage medium |
| CN117114524A (en) * | 2023-10-23 | 2023-11-24 | 香港中文大学（深圳） | Logistics sorting method based on reinforcement learning and digital twin |
| CN117114524B (en) * | 2023-10-23 | 2024-01-26 | 香港中文大学（深圳） | Logistics sorting method based on reinforcement learning and digital twin |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109711624B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109685278A (en) | Packing method, equipment and computer readable storage medium | |
| CN110443549B (en) | Method and system for generating boxing scheme of articles in logistics warehouse | |
| CN103455841B (en) | Based on improving ant group algorithm and the container loading method of heuritic approach | |
| CN110097315A (en) | Container determines method, container determining device, medium and calculates equipment | |
| CN103870893A (en) | Optimization method for solving encasement problem under multiple weight restrictions based on three-dimensional space | |
| Chen et al. | The self-learning particle swarm optimization approach for routing pickup and delivery of multiple products with material handling in multiple cross-docks | |
| JP7453722B2 (en) | Optimization method and system for multi-box three-dimensional loading based on multi-tree search | |
| Kavousi-Fard et al. | Optimal distribution feeder reconfiguration for increasing the penetration of plug-in electric vehicles and minimizing network costs | |
| CN104915817A (en) | Loading method and apparatus for container | |
| CN109665328A (en) | A kind of dynamic optimization chest palletizing method, information data processing terminal | |
| CN109711624A (en) | Packing method, equipment and computer readable storage medium | |
| CN105976108A (en) | Distributed energy storage planning method of power distribution network | |
| CN108229867A (en) | Material arranges task generation, material method for sorting and device | |
| CN105279629B (en) | A kind of intelligent packaging system of optimization | |
| CN109670263A (en) | Packing method, equipment and computer readable storage medium | |
| CN104680237A (en) | Three-dimensional encasement novel genetic algorithm model under multi-constrain condition | |
| CN110046845A (en) | A kind of place vehicle loading method and Load System | |
| CN108448731A (en) | A kind of collaborative wireless sense network energy compensation process and its wireless sense network | |
| CN108062615A (en) | Production plan apparatus and method and non-transitory computer-readable storage media | |
| CN107958314A (en) | A kind of electric automobile fast charge station charging service takes pricing method and device | |
| CN108959783A (en) | A kind of layout simulation optimization method and device in intelligence workshop | |
| CN107067136A (en) | Charging electric vehicle distribution method and device | |
| Iliopoulou et al. | Robust electric transit route network design problem (RE-TRNDP) with delay considerations: Model and application | |
| CN113135290B (en) | Article placing method and device suitable for air transportation | |
| Grignon et al. | Bi-objective optimization of components packing using a genetic algorithm |
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 | ||
| CB02 | Change of applicant information |
Address after: B701-702, industrialization building, Shenzhen Virtual University Park, No.2, Yuexing Third Road, Nanshan District, Shenzhen, Guangdong Province Applicant after: Shenzhen Lan pangzi machine intelligence Co., Ltd Address before: B701-702, industrialization building, Shenzhen Virtual University Park, No.2, Yuexing Third Road, Nanshan District, Shenzhen, Guangdong Province Applicant before: SHENZHEN DORABOT ROBOTICS Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |