CN115147054A - Goods packing planning method and device - Google Patents

Goods packing planning method and device Download PDF

Info

Publication number
CN115147054A
CN115147054A CN202211075691.2A CN202211075691A CN115147054A CN 115147054 A CN115147054 A CN 115147054A CN 202211075691 A CN202211075691 A CN 202211075691A CN 115147054 A CN115147054 A CN 115147054A
Authority
CN
China
Prior art keywords
goods
cargo
loading space
model
point cloud
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
Application number
CN202211075691.2A
Other languages
Chinese (zh)
Other versions
CN115147054B (en
Inventor
林以明
盖晨阳
周鹏程
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Innovation Qizhi Qingdao Technology Co ltd
Original Assignee
Innovation Qizhi Qingdao Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Innovation Qizhi Qingdao Technology Co ltd filed Critical Innovation Qizhi Qingdao Technology Co ltd
Priority to CN202211075691.2A priority Critical patent/CN115147054B/en
Publication of CN115147054A publication Critical patent/CN115147054A/en
Application granted granted Critical
Publication of CN115147054B publication Critical patent/CN115147054B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/86—Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/75—Determining position or orientation of objects or cameras using feature-based methods involving models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10028—Range image; Depth image; 3D point clouds
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10032—Satellite or aerial image; Remote sensing
    • G06T2207/10044—Radar image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Remote Sensing (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Quality & Reliability (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Geometry (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Development Economics (AREA)
  • Electromagnetism (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application relates to the technical field of logistics transportation, and provides a cargo packing planning method and device. The method comprises the following steps: acquiring a point cloud image and a two-dimensional image of the container, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining a loading space of the container; according to the space for loading the material, determining the model of the target cargo; acquiring various goods to be boxed corresponding to the model of the target goods from various unpacked goods; and sequentially and adjacently placing the goods to be boxed into the loading space, and after placing the goods to be boxed into the current loading space each time, acquiring a current point cloud image and a current two-dimensional image of the container and inputting the current point cloud image and the current two-dimensional image into the neural network model to update the current loading space of the container until the placing of the goods to be boxed is completed or the updated current loading space cannot meet the placing of the goods to be boxed, and determining a boxing planning result of the goods to be boxed. The goods packing planning method provided by the embodiment of the application can improve the storage and taking efficiency of goods.

Description

Cargo packing planning method and device
Technical Field
The application relates to the technical field of logistics transportation, in particular to a cargo packing planning method and device.
Background
With the rapid development of the internet technology, the transaction amount of goods is more and more, which puts higher requirements on the transportation of goods, and the loading of the goods affects the transportation efficiency of the goods.
At present, goods vanning usually loads the goods according to the quantity of goods, and along with the increase of goods quantity, plans the loading of how to better goods to improve the access efficiency of goods, be the problem that needs to solve at present urgently.
Disclosure of Invention
The present application is directed to solving at least one of the technical problems occurring in the related art. Therefore, the application provides a cargo packing planning method which can improve the storage and taking efficiency of cargos.
The application also provides a goods vanning planning device.
The application also provides an electronic device.
The application also provides a computer readable storage medium.
According to the cargo packing planning method in the embodiment of the first aspect of the application, the method comprises the following steps:
acquiring a point cloud image and a two-dimensional image of a container, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining a loading space of the container;
determining the model of the target cargo according to the loading space;
acquiring each goods to be boxed corresponding to the target goods model from each unpacked goods;
sequentially and adjacently placing each goods to be boxed into the loading space, obtaining the current point cloud image and the current two-dimensional image of the container and inputting the current point cloud image and the current two-dimensional image into the neural network model after each time of placing the goods to be boxed into the current loading space, so as to update the current loading space of the container until the placing of each goods to be boxed is completed or the updated current loading space cannot meet the placing of the goods to be boxed, and determining a boxing planning result of each goods to be boxed;
the neural network model is obtained by training each training image set, and any training image set comprises a point cloud image sample and a two-dimensional image sample of any packing box at any moment;
the point cloud image and each point cloud image sample are obtained through a laser radar, and the two-dimensional image and each two-dimensional image sample are obtained through a camera device;
the laser radar and the camera device are located at the same position and orientation relative to the cargo box.
The goods packing planning method provided by the embodiment of the application, the loading space of the packing box is determined through the point cloud image and the two-dimensional image of the packing box, the target goods model is determined according to the obtained loading space, in order to obtain each goods to be packed corresponding to the target goods model from each goods, each goods to be packed are sequentially and adjacently placed into the loading space, so that the size of the loading space of the packing box can be accurately determined, the goods of a certain proper model can be selected for packing, the packing efficiency is improved, the goods of the same model can be adjacently and continuously placed, the goods of different models are prevented from being mixed, the follow-up unified management of the packed goods is ensured, and the goods storage and taking efficiency is improved. Meanwhile, the loading postures of the cargo planning of each time are possibly different, so that the loading space is re-planned after any cargo is subjected to the loading planning, and the feasibility of the finally formed cargo loading planning is ensured.
According to one embodiment of the application, the method for acquiring the point cloud image and the two-dimensional image of the container, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining the loading space of the container comprises the following steps:
acquiring a first characteristic of the point cloud image and a second characteristic of the two-dimensional image;
according to the three-dimensional coordinate feature in the first feature and the two-dimensional coordinate feature in the second feature, outliers of which the three-dimensional coordinate feature is not matched with the two-dimensional coordinate feature are removed from the point cloud image, and a target point cloud image is obtained;
inputting the first characteristic and the second characteristic of the target point cloud image into the neural network model, and determining the loading space of the container.
According to one embodiment of the application, the target cargo model is determined according to the loading space of the cargo box, and the method comprises the following steps:
according to the loading space, obtaining the model of each goods to be selected from a candidate goods model list;
determining the model of the target cargo from the models of the to-be-selected cargos;
the candidate cargo model list records the cargo models of the cargoes which are not finished to be boxed;
the goods corresponding to the to-be-selected goods model can be placed in the loading space.
According to an embodiment of the application, determining the target cargo model from each of the to-be-selected cargo models comprises:
and determining the type of the goods to be selected, which is in each type of the goods to be selected, wherein the length difference between the length of the corresponding goods and the length of the loading space meets the preset length, as the type of the target goods.
According to an embodiment of the application, determining, as the target cargo model, the cargo model to be selected, in which a length difference between a length of the corresponding cargo and a length of the loading space satisfies a preset length, among the cargo models to be selected, includes:
acquiring the length of the corresponding goods from the types of the goods to be selected, wherein the length difference between the length of the goods and the length of the loading space meets each preset goods type with preset length;
and determining the preset goods model with the maximum corresponding quantity of the goods in the preset goods models as the target goods model.
According to an embodiment of the application, determining the target cargo model from the candidate cargo models comprises:
and determining the type of the goods to be selected with the largest bottom area of the corresponding goods to be boxed in the types of the goods to be selected as the type of the target goods.
According to an embodiment of the present application, further comprising:
and mapping the loading space and the boxing planning result and then storing the mapping result.
According to the second aspect of the application, the cargo packing planning device comprises:
the space determining module is used for acquiring a point cloud image and a two-dimensional image of the container, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining the loading space of the container;
the model determining module is used for determining the model of the target cargo according to the loading space of the cargo box;
the goods acquisition module is used for acquiring goods to be boxed corresponding to the target goods model from goods which are not boxed;
the packing planning module is used for sequentially and adjacently placing each piece of goods to be packed into the loading space, acquiring the current point cloud image and the current two-dimensional image of the packing box and inputting the current point cloud image and the current two-dimensional image into the neural network model after each time of placing the goods to be packed into the current loading space, so as to update the current loading space of the packing box until the placement of each piece of goods to be packed is completed or the updated current loading space cannot meet the placement of the goods to be packed, and determining a packing planning result of each piece of goods to be packed;
the neural network model is obtained by training each training image set, and any training image set comprises a point cloud image sample and a two-dimensional image sample of any container at any moment;
the point cloud image and each point cloud image sample are obtained through a laser radar, and the two-dimensional image and each two-dimensional image sample are obtained through a camera device;
the laser radar and the camera device are located at the same position and orientation relative to the cargo box.
The electronic device according to the third aspect of the present application includes a processor and a memory storing a computer program, and the processor implements the cargo packing planning method according to any of the above embodiments when executing the computer program.
A computer-readable storage medium according to an embodiment of the fourth aspect of the present application, on which a computer program is stored, which computer program, when being executed by a processor, is adapted to carry out the method for planning the packing of goods according to any of the embodiments described above.
The computer program product according to an embodiment of the fifth aspect of the application comprises: the computer program, when executed by a processor, implements a method for cargo packing planning as described in any of the above embodiments.
One or more technical solutions in the embodiments of the present application have at least one of the following technical effects:
the loading space of packing box is confirmed through the point cloud image and the two-dimensional image of packing box, and according to the loading space that obtains the target goods model, in order from each goods, acquire each goods of waiting to vanning that correspond with the target goods model after, will respectively wait that the vanning goods is adjacent in proper order puts to the loading space in, thereby can confirm accurately that the size of the loading space of packing box carries out the vanning in order to select the goods of certain suitable model, improve vanning efficiency, and can carry out adjacent continuous putting with the goods of the same model, avoid the goods of different models to mix the pendulum, guarantee that follow-up can carry out unified management to the goods of vanning, improve the access efficiency of goods. Meanwhile, the loading postures of the cargo planning of each time are possibly different, so that the loading space is re-planned after any cargo is subjected to the loading planning, and the feasibility of the finally formed cargo loading planning is ensured.
Further, outliers are removed from the point cloud image through the three-dimensional coordinate features of the point cloud image and the two-dimensional coordinate features of the two-dimensional image, and then the first features of the target point cloud image and the second features of the two-dimensional image, which are obtained after the outliers are removed, are input into the space recognition submodel of the neural network model to determine the loading space of the container and improve the accuracy of the obtained loading space.
Furthermore, through the loading space of the container, the types of the goods to be selected are obtained from the candidate goods type list recorded with the types of the goods which are not packed, and then the type of the target goods is determined from the types of the goods to be selected, so that the situation that the selected goods of the type of the target goods do not exist is avoided, and the packing planning efficiency of the goods is improved.
Furthermore, the length of the corresponding goods in the goods models to be selected and the goods models to be selected, the length difference between which and the length of the loading space meets the preset length, are determined as the target goods models, so that a screening strategy can be formed by the lengths of the goods to determine the target goods models, the placing sequence with the possibility of better effect can be preferentially searched when the goods are subsequently planned for loading, and the planning result of the goods loading is controllable.
Furthermore, the length of the corresponding goods is obtained from the goods models to be selected, the length difference between the length of the goods to be selected and the length of the loading space meets the preset length of each preset goods model, and then the preset goods model with the largest quantity of the corresponding goods in each preset goods model is determined as the target goods model, so that the excessive quantity of the goods with certain goods model can be avoided when the goods packing planning is subsequently carried out.
Furthermore, the type of the goods to be selected with the largest bottom area of the corresponding goods to be boxed in the types of the goods to be selected is determined as the type of the target goods, so that the loading space can be fully utilized when the goods are subsequently planned for boxing, and the space utilization rate is improved.
Drawings
In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flowchart of a cargo packing planning method provided in an embodiment of the present application;
FIG. 2 is a schematic flow chart illustrating further details of the determination of the loading space of the cargo container planning method of FIG. 1 in an embodiment of the present application;
fig. 3 is a schematic flow chart illustrating further details of the determination of the target cargo model of the cargo encasement planning method of fig. 1 in an embodiment of the present application;
fig. 4 is a schematic structural diagram of a cargo boxing planning device provided in the embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
To make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Hereinafter, the cargo container planning method and apparatus provided in the embodiments of the present application will be described and explained in detail through several specific embodiments.
In one embodiment, a cargo packing planning method is provided, and the method is applied to a server and used for cargo packing planning of a user. The server can be an independent server or a server cluster formed by a plurality of servers, and can also be a cloud server providing basic cloud computing services such as cloud service, a cloud database, cloud computing, a cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN (content delivery network), big data and artificial intelligence sampling point equipment and the like.
As shown in fig. 1, the method for planning the packing of goods provided by this embodiment includes:
step S101, a point cloud image and a two-dimensional image of a container are obtained, the point cloud image and the two-dimensional image are input into a trained neural network model, and a loading space of the container is determined;
step S102, determining the type of the target cargo according to the loading space;
step S103, acquiring goods to be boxed corresponding to the target goods model from goods which are not boxed;
step S104, sequentially and adjacently placing each goods to be boxed into the loading space, and after each time of placing the goods to be boxed into the current loading space, obtaining the current point cloud image and the current two-dimensional image of the packing box and inputting the point cloud image and the current two-dimensional image into the neural network model to update the current loading space of the packing box until the placing of each goods to be boxed or the updated current loading space cannot meet the placing of the goods to be boxed, and determining a box-loading planning result of each goods to be boxed;
the neural network model is obtained by training each training image set, and any training image set comprises a point cloud image sample and a two-dimensional image sample of any container at any moment;
the point cloud image and each point cloud image sample are obtained through a laser radar, and the two-dimensional image and each two-dimensional image sample are obtained through a camera device;
the laser radar and the camera device are located at the same position and orientation relative to the cargo box.
The loading space of the container is determined through the point cloud image and the two-dimensional image of the container, the target cargo model is determined according to the obtained loading space, the cargo to be boxed corresponding to the target cargo model is obtained from each cargo, each cargo to be boxed is sequentially and adjacently placed into the loading space, the cargo to be boxed can be accurately determined to be boxed according to the size of the loading space of the container, the cargo of a certain proper model can be selected, the boxing efficiency is improved, the cargo of the same model can be adjacently and continuously placed, the cargo of different models is prevented from being mixed, the follow-up unified management of the boxed cargo is ensured, and the storage and taking efficiency of the cargo is improved. Meanwhile, the loading postures of the cargo planning of each time are possibly different, so that the loading space is re-planned after any cargo is subjected to the loading planning, and the feasibility of the finally formed cargo loading planning is ensured.
In one embodiment, the lidar is a product of a combination of laser technology and radar technology, for short, a laser detection and ranging system. The laser radar adopts a laser as a radar of a radiation source, and generally comprises a transmitter, an antenna, a receiver, a tracking frame, information processing and the like. The transmitter is a laser of various forms; the antenna is an optical telescope; the receiver adopts various forms of light spot detectors; the laser radar adopts two working modes of pulse or continuous wave, and the detection method is divided into direct detection and heterodyne detection. The laser detection and ranging system comprises a single-beam narrow-band laser and a receiving system. The laser generates and emits a beam of light pulses which impinge on the object and are reflected back and finally received by the receiver. The laser radar is used for accurately measuring the position (distance and angle), the motion state (speed, vibration and attitude) and the shape of a target, and detecting, identifying, distinguishing and tracking the target.
The point cloud image of the container is a mass point set which expresses the space distribution of the container and the surface characteristics of the container under the same space reference system, and after the space coordinates of each sampling point on the surface of the container are obtained, a point set is obtained and is called as point cloud.
The two-dimensional image of the cargo box is captured by an imaging device, such as a camera. In order to ensure that the training result of the neural network model is accurate enough and the loading space obtained according to the point cloud image and the two-dimensional image is accurate enough, the positions of the laser radar and the camera equipment relative to the container need to be the same, and the orientations of the laser radar and the camera equipment relative to the container are the same. Therefore, in the image acquired by the laser radar, most points can find a certain pixel which is the same as the abscissa and the ordinate of the point cloud image in the two-dimensional image, so that the point cloud image and the two-dimensional image can be conveniently matched and processed by the neural network model, and the point cloud image sample and the two-dimensional image sample can be conveniently matched and processed.
In an embodiment, the training of the neural network model may be to input the point cloud image sample and the two-dimensional image sample of any container at any time into the neural network model to obtain a storage space output by the neural network model, and then compare the storage space output by the neural network model with a storage space of the container actually measured, and adjust a loss function of the neural network model until a volume difference between the storage space output by the neural network model and the storage space of the container actually measured is within a preset range, thereby completing the training of the neural network model.
The point cloud image sample and the two-dimensional image sample are input into the data of the neural network model, may be a first feature of a point cloud image sample, and a second feature of a two-dimensional image sample. The first feature comprises a three-dimensional coordinate feature, and the second feature comprises a two-dimensional coordinate feature and a color feature. The three-dimensional coordinate characteristics comprise three-dimensional coordinates of each point in each point cloud, and the two-dimensional coordinate characteristics comprise coordinates of each pixel point in the image. The color characteristics are RGB information of the image.
After the training of the neural network model is completed, the first characteristic of the point cloud image of the container and the second characteristic of the two-dimensional image can be extracted, and then the first characteristic of the point cloud image and the second characteristic of the two-dimensional image are input into the trained neural network model to determine the loading space of the container.
In order to make the determined loading space more accurate, in an embodiment, as shown in fig. 2, the obtaining a point cloud image and a two-dimensional image of the container, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining the loading space of the container includes:
step S201, acquiring a first feature of the point cloud image and a second feature of the two-dimensional image;
step S202, according to the three-dimensional coordinate feature in the first feature and the two-dimensional coordinate feature in the second feature, outliers of which the three-dimensional coordinate feature is not matched with the two-dimensional coordinate feature are removed from the point cloud image, and a target point cloud image is obtained;
step S203, inputting the first characteristic and the second characteristic of the target point cloud image into the neural network model, and determining the loading space of the container.
In an embodiment, after the first feature of the point cloud image and the second feature of the two-dimensional image are obtained, the three-dimensional coordinate feature in the first feature of the point cloud image is matched with the two-dimensional coordinate feature of the second feature of the two-dimensional image. That is, the three-dimensional coordinates of each point in the point cloud image are matched with the two-dimensional coordinates of each pixel point of the two-dimensional image. If the abscissa and the ordinate of the three-dimensional coordinate of a certain point in the point cloud image are not matched with the abscissa and the ordinate of the two-dimensional coordinate of each pixel point, the point is represented as an outlier, and the outlier is removed from the point cloud image at the moment. And after all outliers are removed, a target point cloud image can be obtained, and the first characteristic and the second characteristic of the target point cloud image are input into the neural network model to obtain the loading space of the container. Therefore, when the loading space of the container is determined, the interference of outliers is eliminated, and the accuracy of the determined loading space is improved.
After the loading space is obtained, all goods which can be placed in the loading space can be determined according to the size of the loading space. Wherein, judge whether the goods can be put into the loading space, accessible obtains loading space's length L1, height H1 and width W1 to length L1, height H1 and width W1 with the loading space compare with the length L2 of goods, height H2 and width W2 respectively. If the length L1 is greater than or equal to the length L2, the height H1 is greater than or equal to the height H2, and the width W1 is greater than or equal to the width W2, it is determined that the cargo can be placed into the loading space.
After determining all the goods which can be placed in the loading space, the goods models of the goods are obtained, and all the goods models can be searched, so that one goods model is randomly selected from all the goods models to serve as a target goods model. The goods type of the goods is determined according to the size and the shape of the goods, and the goods with the same size and the same shape are divided into the same goods type.
Considering that the goods of a certain goods model may be completely loaded, if a goods model is randomly selected from all goods models as a target goods model, the goods of the selected target goods model do not exist, the search time is influenced, and the packing planning efficiency of the goods is influenced. To this end, in an embodiment, as shown in fig. 3, the determining the target cargo model according to the loading space of the cargo box includes:
step S301, acquiring the model of each to-be-selected cargo from a candidate cargo model list according to the loading space;
step S302, determining the target cargo model from the cargo models to be selected;
the candidate goods model list records the goods model of the goods which are not packaged completely;
the goods corresponding to the goods model to be selected can be placed in the loading space.
In an embodiment, after determining all the goods that can be placed in the loading space, matching the goods models of the goods with the goods models of the goods which are not completely boxed and recorded in the candidate goods model list to obtain the same goods model as the to-be-selected goods model. After the types of the goods to be selected are obtained, searching is carried out on the types of the goods to be selected, so that one type of the goods to be selected is selected as a target type of the goods from all the types of the goods to be selected, and therefore the goods of the selected target type of the goods do not exist.
Through the loading space of the container, the types of the goods to be selected are obtained from the candidate goods type list recorded with the types of the goods which are not packed, and then the type of the target goods is determined from the types of the goods to be selected, so that the situation that the selected goods of the type of the target goods do not exist is avoided, and the packing planning efficiency of the goods is improved.
After each model of the goods to be selected is obtained from the candidate goods model list, all the models of the goods to be selected can be selected as the target goods model, so that the uncertainty of the selected target goods model is high, the complexity of the search time for the goods of the target goods model is high, and finally the planning result of the goods packing is unpredictable. Therefore, in one embodiment, the target cargo model is determined by a preset selection strategy from various cargo models to be selected, so as to specify the cargo to be placed. Specifically, determining the target cargo model from each of the to-be-selected cargo models includes:
the length of the corresponding goods in each model of the goods to be selected, and determining the model of the goods to be selected, the length difference of which with the length of the loading space meets the preset length, as the model of the target goods.
In one embodiment, the preset selection policy may be a length selection policy. After the types of the goods to be selected are determined, the length of the goods corresponding to each type of the goods to be selected is obtained, then the length of the goods corresponding to any type of the goods to be selected is compared with the length of the loading space, and the length difference corresponding to the type of the goods to be selected is obtained. And then determining the model of the goods to be selected corresponding to the length difference meeting the preset length as the model of the target goods. The preset length may be the minimum length difference among the length differences corresponding to the types of the goods to be selected, that is, the length of the goods corresponding to the determined target type of the goods is the longest. Like this, can confirm the target goods model that corresponds length through the adjustment length of predetermineeing, make the length of the storing space that the goods of the target goods model of confirming occupy controllable, conveniently carry out remaining loading space's loading planning simultaneously.
The goods model to be selected, which meets the preset length through the length difference between the length of the corresponding goods and the length of the loading space in the goods models to be selected, is determined as the target goods model, so that a screening strategy can be formed by utilizing the length of the goods to determine the target goods model, the placing sequence with better effect can be preferentially searched during subsequent goods packing planning, and the planning result of goods packing is controllable.
In one embodiment, the preset selection policy may also be combined with a cargo quantity policy. Specifically, determining the type of the goods to be selected, in which the length difference between the length of the corresponding goods and the length of the loading space satisfies a preset length, as the type of the target goods includes:
acquiring the length of the corresponding goods from the types of the goods to be selected, wherein the length difference between the length of the goods and the length of the loading space meets each preset goods type with preset length;
and determining the preset goods model with the maximum corresponding quantity of the goods in the preset goods models as the target goods model.
In one embodiment, after the types of the goods to be selected are determined, the length of the goods corresponding to each type of the goods to be selected is obtained, and then the length of the goods corresponding to any type of the goods to be selected is compared with the length of the loading space, so that the length difference corresponding to the type of the goods to be selected is obtained. And then, marking the model of the goods to be selected corresponding to the length difference meeting the preset length as a preset goods model. And if only one preset cargo model exists, directly taking the preset cargo model as a target cargo model. And if a plurality of preset goods models exist, acquiring the quantity of goods corresponding to each preset goods model, and determining the preset goods model with the maximum quantity of the corresponding goods as the target goods model.
The length of the corresponding goods is obtained from the goods models to be selected, the length difference between the length of the goods to be selected and the length of the loading space meets each preset goods model with preset length, and then the preset goods model with the largest quantity of the corresponding goods in each preset goods model is determined as the target goods model, so that excessive goods leaving quantity of certain goods models can be avoided when the subsequent goods packing planning is carried out.
In an embodiment, the preset selection policy may also be a floor area maximum policy. Specifically, determining the target cargo model from each of the to-be-selected cargo models includes:
and determining the type of the goods to be selected with the largest bottom area of the corresponding goods to be boxed in the types of the goods to be selected as the type of the target goods.
In one embodiment, after the model of each to-be-selected goods is determined, the bottom area of the goods corresponding to each to-be-selected goods model is obtained, then the bottom areas corresponding to the to-be-selected goods models are compared, and after the largest bottom area is determined, the model of the to-be-selected goods corresponding to the largest bottom area is determined as the target goods model.
The type of the goods to be selected with the largest bottom area of the corresponding goods to be boxed in the types of the goods to be selected is determined as the type of the target goods, so that loading space can be fully utilized in subsequent goods boxing planning, and space utilization rate is improved.
In an embodiment, after the target cargo model is determined, all the cargos to be boxed belonging to the target cargo model can be obtained from the cargos not to be boxed according to the target cargo model, greedy boxing is performed according to the priority sequence of the height direction, the width direction and the length direction, the cargos to be boxed are sequentially and adjacently placed in a loading space, and the cargos to be boxed are placed to form a cubic shape as much as possible. The priority order of the height direction, the width direction and the length direction means that after the goods to be boxed are sequentially stacked in the height direction until the placement of the goods to be boxed is completed or the goods to be boxed cannot be placed more than much in the height direction, if the goods to be boxed still exist, the goods to be boxed are placed into a layer in the width direction until the placement of the goods to be boxed is completed or the goods to be boxed cannot be placed any more in the width direction, if the goods to be boxed still exist, and finally the goods to be boxed are placed in the length direction.
After the goods to be boxed are placed in the current loading space each time, the current point cloud image and the current two-dimensional image of the goods box are obtained again and input into the neural network model so as to update the current loading space of the goods box, the goods to be boxed are placed by using the updated current loading space until the placement of the goods to be boxed is completed or the updated current loading space cannot meet the placement of the goods to be boxed, and the boxing planning result of the goods to be boxed is determined until the goods to be boxed cannot be placed any more, so that the boxing planning result of the goods to be boxed is obtained.
In one embodiment, it is considered that not all the loaded goods to be boxed in all the postures can be put in the loading space, so that a plurality of preset placing postures compatible with the size of the loading space need to be determined. Wherein, because the goods of waiting to pack into the case are cube or cuboid shape, have 6 faces promptly, consequently corresponding predetermine and put the gesture and have 6 at most. However, the allowable placing postures are less than 6 according to the properties of the goods, for example, some goods can only be horizontally placed (the largest plane is below), cannot be laterally placed (the middle and large planes are below) or vertically placed (the smallest plane is below).
After a plurality of preset placing postures are obtained, aiming at any preset placing posture, placing the goods to be boxed into the loading space sequentially and adjacently according to the preset placing posture until the placing of the goods to be boxed is completed or the current loading space of the goods box cannot meet the placing of the goods to be boxed, and determining a planning result corresponding to the preset placing posture.
And after the planning results which are in one-to-one correspondence with the preset placing postures are obtained, the planning results meeting the preset conditions are obtained from the planning results to serve as boxing planning results. The preset condition can be that the number of the placed goods to be boxed is maximum, so that the goods loading capacity of the obtained boxing planning result is optimal; or the rest space of the loading space after placement is minimum, so that the obtained packing planning result is the planning result with the optimal space utilization rate.
By acquiring a plurality of preset placing postures and aiming at any preset placing posture, sequentially and adjacently placing each to-be-boxed cargo into a loading space according to the preset placing posture, after each planning result corresponding to each preset placing posture one to one is obtained, the boxing planning result meeting the preset conditions is acquired from each planning result, and the finally determined boxing planning result can meet the actual requirement.
After the packing planning result is determined, if the container has a remaining space, the remaining space may be marked as a loading space, and the steps 101 to 104 of the above embodiment are executed again until the container has no remaining space or the remaining space is no longer available for placing any goods, thereby completing the packing planning of the entire container.
In an embodiment, after the result of the packing plan is determined, the result of the packing plan may be mapped with the loading space and then stored in a mapping table describing a mapping relationship between the loading space and the result of the packing plan. Therefore, when a new loading space is acquired subsequently and is matched with any loading space recorded in the mapping table, the packing planning result corresponding to the new loading space can be determined directly from the mapping table without re-packing planning of goods, and the packing planning efficiency of the goods is improved.
The following describes the cargo packing planning apparatus provided in the present application, and the cargo packing planning apparatus described below and the cargo packing planning method described above may be referred to in correspondence with each other.
In one embodiment, as shown in fig. 4, there is provided a cargo packing planning apparatus, including:
the space determining module 210 is configured to obtain a point cloud image and a two-dimensional image of the container, input the point cloud image and the two-dimensional image into a trained neural network model, and determine a loading space of the container;
the model determining module 220 is used for determining the model of the target cargo according to the loading space of the cargo box;
a cargo obtaining module 230, configured to obtain each cargo to be boxed corresponding to the target cargo model from each unboxed cargo;
a packing planning module 240, configured to sequentially and adjacently place each of the goods to be packed into the loading space, and after each time the goods to be packed is placed into the current loading space, obtain the current point cloud image and the current two-dimensional image of the container and input the point cloud image and the current two-dimensional image into the neural network model, so as to update the current loading space of the container, until the placement of each of the goods to be packed is completed or the updated current loading space cannot meet the placement of the goods to be packed, and determine a packing planning result of each of the goods to be packed;
the neural network model is obtained by training each training image set, and any training image set comprises a point cloud image sample and a two-dimensional image sample of any container at any moment;
the point cloud image and each point cloud image sample are obtained through a laser radar, and the two-dimensional image and each two-dimensional image sample are obtained through a camera device;
the laser radar and the camera device are located at the same position and orientation relative to the cargo box.
In an embodiment, the space determining module is specifically configured to:
acquiring a first characteristic of the point cloud image and a second characteristic of the two-dimensional image;
according to the three-dimensional coordinate feature in the first feature and the two-dimensional coordinate feature in the second feature, outliers of which the three-dimensional coordinate feature is not matched with the two-dimensional coordinate feature are removed from the point cloud image, and a target point cloud image is obtained;
inputting the first characteristic and the second characteristic of the target point cloud image into the neural network model, and determining the loading space of the container.
In an embodiment, the model determining module 220 is specifically configured to:
according to the loading space, obtaining the model of each goods to be selected from a candidate goods model list;
determining the model of the target cargo from the models of the to-be-selected cargos;
the candidate cargo model list records the cargo models of the cargoes which are not finished to be boxed;
the goods corresponding to the to-be-selected goods model can be placed in the loading space.
In an embodiment, the model determining module 220 is specifically configured to:
and determining the type of the goods to be selected, which is the target type of the goods, in which the length difference between the length of the corresponding goods and the length of the loading space meets the preset length, in each type of the goods to be selected.
In an embodiment, the model determining module 220 is specifically configured to:
acquiring the length of the corresponding goods from the types of the goods to be selected, wherein the length difference between the length of the goods and the length of the loading space meets the preset length of each preset goods type;
and determining the preset goods model with the maximum corresponding quantity of the goods in the preset goods models as the target goods model.
In an embodiment, the model determination module 220 is specifically configured to:
and determining the type of the goods to be selected with the largest bottom area of the corresponding goods to be boxed in the types of the goods to be selected as the type of the target goods.
In an embodiment, the bin planning module 230 is further configured to:
and mapping the loading space and the boxing planning result and then storing the mapping result.
Fig. 5 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 5: a processor (processor) 810, a Communication Interface 820, a memory 830 and a Communication bus 840, wherein the processor 810, the Communication Interface 820 and the memory 830 communicate with each other via the Communication bus 840. The processor 810 may invoke computer programs in the memory 830 to perform a cargo bin planning method, including, for example:
acquiring a point cloud image and a two-dimensional image of a container, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining a loading space of the container;
determining the type of the target cargo according to the loading space;
acquiring each goods to be boxed corresponding to the target goods model from each unpacked goods;
sequentially and adjacently placing each goods to be boxed into the loading space, obtaining the current point cloud image and the current two-dimensional image of the container and inputting the current point cloud image and the current two-dimensional image into the neural network model after each time of placing the goods to be boxed into the current loading space, so as to update the current loading space of the container until the placing of each goods to be boxed is completed or the updated current loading space cannot meet the placing of the goods to be boxed, and determining a boxing planning result of each goods to be boxed;
the neural network model is obtained by training each training image set, and any training image set comprises a point cloud image sample and a two-dimensional image sample of any container at any moment;
the point cloud image and each point cloud image sample are obtained through a laser radar, and the two-dimensional image and each two-dimensional image sample are obtained through a camera device;
the laser radar and the camera device are located at the same position and orientation relative to the cargo box.
In addition, the logic instructions in the memory 830 can be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, an embodiment of the present application further provides a storage medium, where the storage medium includes a computer program, where the computer program is stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is capable of executing the cargo packing planning method provided in the foregoing embodiments, for example, including:
acquiring a point cloud image and a two-dimensional image of a container, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining a loading space of the container;
determining the model of the target cargo according to the loading space;
acquiring each goods to be boxed corresponding to the target goods model from each unpacked goods;
sequentially and adjacently placing each goods to be boxed into the loading space, obtaining the current point cloud image and the current two-dimensional image of the container and inputting the current point cloud image and the current two-dimensional image into the neural network model after each time of placing the goods to be boxed into the current loading space, so as to update the current loading space of the container until the placing of each goods to be boxed is completed or the updated current loading space cannot meet the placing of the goods to be boxed, and determining a boxing planning result of each goods to be boxed;
the neural network model is obtained by training each training image set, and any training image set comprises a point cloud image sample and a two-dimensional image sample of any packing box at any moment;
the point cloud image and each point cloud image sample are obtained through a laser radar, and the two-dimensional image and each two-dimensional image sample are obtained through a camera device;
the laser radar and the camera device are located at the same position and orientation relative to the cargo box.
On the other hand, embodiments of the present application further provide a processor-readable storage medium, where a computer program is stored in the processor-readable storage medium, where the computer program is configured to cause a processor to execute the method provided in each of the above embodiments, for example, the method includes:
acquiring a point cloud image and a two-dimensional image of a container, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining a loading space of the container;
determining the model of the target cargo according to the loading space;
acquiring each goods to be boxed corresponding to the target goods model from each unpacked goods;
sequentially and adjacently placing each goods to be boxed into the loading space, obtaining the current point cloud image and the current two-dimensional image of the container and inputting the current point cloud image and the current two-dimensional image into the neural network model after placing the goods to be boxed into the current loading space each time so as to update the current loading space of the container until the placing of each goods to be boxed or the updated current loading space cannot meet the placing of the goods to be boxed, and determining a boxing planning result of each goods to be boxed;
the neural network model is obtained by training each training image set, and any training image set comprises a point cloud image sample and a two-dimensional image sample of any container at any moment;
the point cloud image and each point cloud image sample are obtained through a laser radar, and the two-dimensional image and each two-dimensional image sample are obtained through a camera device;
the laser radar and the camera device are located at the same position and orientation relative to the cargo box.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims (10)

1. A method for planning the packing of goods, comprising:
acquiring a point cloud image and a two-dimensional image of a container, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining a loading space of the container;
determining the model of the target cargo according to the loading space;
acquiring each goods to be boxed corresponding to the target goods model from each unpacked goods;
sequentially and adjacently placing each goods to be boxed into the loading space, obtaining the current point cloud image and the current two-dimensional image of the container and inputting the current point cloud image and the current two-dimensional image into the neural network model after placing the goods to be boxed into the current loading space each time so as to update the current loading space of the container until the placing of each goods to be boxed or the updated current loading space cannot meet the placing of the goods to be boxed, and determining a boxing planning result of each goods to be boxed;
the neural network model is obtained by training each training image set, and any training image set comprises a point cloud image sample and a two-dimensional image sample of any container at any moment;
the point cloud image and each point cloud image sample are obtained through a laser radar, and the two-dimensional image and each two-dimensional image sample are obtained through a camera device;
the laser radar and the camera device are located at the same position and orientation relative to the container.
2. The cargo packing planning method according to claim 1, wherein the steps of obtaining a point cloud image and a two-dimensional image of a cargo box, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining the loading space of the cargo box comprise:
acquiring a first characteristic of the point cloud image and a second characteristic of the two-dimensional image;
according to the three-dimensional coordinate feature in the first feature and the two-dimensional coordinate feature in the second feature, outliers of which the three-dimensional coordinate feature is not matched with the two-dimensional coordinate feature are removed from the point cloud image, and a target point cloud image is obtained;
inputting the first characteristic and the second characteristic of the target point cloud image into the neural network model, and determining the loading space of the container.
3. The method for planning the packing of the cargo according to claim 1, wherein the determining the target cargo model according to the loading space of the cargo box comprises:
according to the loading space, obtaining the model of each goods to be selected from the candidate goods model list;
determining the model of the target cargo from the models of the to-be-selected cargos;
the candidate cargo model list records the cargo models of the cargoes which are not finished to be boxed;
the goods corresponding to the to-be-selected goods model can be placed in the loading space.
4. The method for planning the packing of the cargo according to claim 3, wherein the determining the target cargo model from the selected cargo models comprises:
and determining the type of the goods to be selected, which is the target type of the goods, in which the length difference between the length of the corresponding goods and the length of the loading space meets the preset length, in each type of the goods to be selected.
5. The cargo boxing planning method according to claim 4, wherein the target cargo model is determined as the cargo model to be selected, which is one of the cargo models to be selected, and which has a length difference between a length of the corresponding cargo and a length of the loading space, and which satisfies a preset length, and the method comprises the following steps:
acquiring the length of the corresponding goods from the types of the goods to be selected, wherein the length difference between the length of the goods and the length of the loading space meets each preset goods type with preset length;
and determining the preset goods model with the maximum corresponding quantity of the goods in the preset goods models as the target goods model.
6. The method for planning the packing of the cargo according to claim 3, wherein the determining the target cargo model from the selected cargo models comprises:
and determining the type of the goods to be selected with the largest bottom area of the corresponding goods to be boxed in the types of the goods to be selected as the type of the target goods.
7. The cargo encasement planning method of claim 1, further comprising:
and mapping the loading space and the boxing planning result and then storing the mapping result.
8. A cargo encasement planning apparatus, comprising:
the space determining module is used for acquiring a point cloud image and a two-dimensional image of the container, inputting the point cloud image and the two-dimensional image into a trained neural network model, and determining the loading space of the container;
the model determining module is used for determining the model of the target cargo according to the loading space of the cargo box;
the goods obtaining module is used for obtaining goods to be boxed corresponding to the target goods model from the goods which are not boxed;
the packing planning module is used for sequentially and adjacently placing each piece of goods to be packed into the loading space, acquiring the current point cloud image and the current two-dimensional image of the packing box and inputting the point cloud image and the two-dimensional image into the neural network model after each time of placing the goods to be packed into the current loading space, so as to update the current loading space of the packing box until the placing of each piece of goods to be packed or the updated current loading space cannot meet the placing of the goods to be packed is completed, and determining a packing planning result of each piece of goods to be packed;
the neural network model is obtained by training each training image set, and any training image set comprises a point cloud image sample and a two-dimensional image sample of any container at any moment;
the point cloud image and each point cloud image sample are obtained through a laser radar, and the two-dimensional image and each two-dimensional image sample are obtained through a camera device;
the laser radar and the camera device are located at the same position and orientation relative to the cargo box.
9. An electronic device comprising a processor and a memory storing a computer program, wherein the processor, when executing the computer program, implements the method of cargo encasement planning of any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method for cargo packing planning according to any one of claims 1 to 7.
CN202211075691.2A 2022-09-05 2022-09-05 Goods packing planning method and device Active CN115147054B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211075691.2A CN115147054B (en) 2022-09-05 2022-09-05 Goods packing planning method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211075691.2A CN115147054B (en) 2022-09-05 2022-09-05 Goods packing planning method and device

Publications (2)

Publication Number Publication Date
CN115147054A true CN115147054A (en) 2022-10-04
CN115147054B CN115147054B (en) 2022-12-02

Family

ID=83415378

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211075691.2A Active CN115147054B (en) 2022-09-05 2022-09-05 Goods packing planning method and device

Country Status (1)

Country Link
CN (1) CN115147054B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116452656A (en) * 2023-06-16 2023-07-18 星猿哲科技(深圳)有限公司 Method, device, equipment and storage medium for neatly placing objects in box
CN117557193A (en) * 2023-11-14 2024-02-13 广州方舟信息科技有限公司 Method, device, equipment and storage medium for formulating boxing strategy

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103761635A (en) * 2014-01-14 2014-04-30 大连理工大学 Three-dimensional multi-box specially-structured cargo loading optimizing method
CN110853004A (en) * 2019-11-04 2020-02-28 华东交通大学 Intelligent shape recognition and optimized packing method and device for part load goods
CN112815941A (en) * 2019-11-18 2021-05-18 北京京邦达贸易有限公司 Goods picking path planning method, device, equipment and storage medium
CN113610172A (en) * 2021-08-13 2021-11-05 北京地平线信息技术有限公司 Neural network model training method and device, and sensing data fusion method and device
CN113963044A (en) * 2021-09-30 2022-01-21 北京工业大学 RGBD camera-based intelligent loading method and system for cargo box
CN114612030A (en) * 2020-12-08 2022-06-10 顺丰科技有限公司 Cabin allocation optimization method and device, computer equipment and storage medium

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103761635A (en) * 2014-01-14 2014-04-30 大连理工大学 Three-dimensional multi-box specially-structured cargo loading optimizing method
CN110853004A (en) * 2019-11-04 2020-02-28 华东交通大学 Intelligent shape recognition and optimized packing method and device for part load goods
CN112815941A (en) * 2019-11-18 2021-05-18 北京京邦达贸易有限公司 Goods picking path planning method, device, equipment and storage medium
CN114612030A (en) * 2020-12-08 2022-06-10 顺丰科技有限公司 Cabin allocation optimization method and device, computer equipment and storage medium
CN113610172A (en) * 2021-08-13 2021-11-05 北京地平线信息技术有限公司 Neural network model training method and device, and sensing data fusion method and device
CN113963044A (en) * 2021-09-30 2022-01-21 北京工业大学 RGBD camera-based intelligent loading method and system for cargo box

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘宁君等: "运输机群货物装载方案生成方法", 《北京航空航天大学学报》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116452656A (en) * 2023-06-16 2023-07-18 星猿哲科技(深圳)有限公司 Method, device, equipment and storage medium for neatly placing objects in box
CN116452656B (en) * 2023-06-16 2023-11-28 星猿哲科技(深圳)有限公司 Method, device, equipment and storage medium for neatly placing objects in box
CN117557193A (en) * 2023-11-14 2024-02-13 广州方舟信息科技有限公司 Method, device, equipment and storage medium for formulating boxing strategy

Also Published As

Publication number Publication date
CN115147054B (en) 2022-12-02

Similar Documents

Publication Publication Date Title
CN115147054B (en) Goods packing planning method and device
US20220286512A1 (en) Server device, learned model providing program, learned model providing method, and learned model providing system
WO2019136369A1 (en) Network computer system to evaluate freight loads
US20130188841A1 (en) Densifying and colorizing point cloud representation of physical surface using image data
US20180094966A1 (en) Generating load characteristic information based on sensor data
US20120114181A1 (en) Vehicle pose estimation and load profiling
US20170227674A1 (en) Method of imaging an object for tracking and documentation in transportation and storage
US11087273B1 (en) Item recognition system using reference images
CN111895989A (en) Robot positioning method and device and electronic equipment
CN112700552A (en) Three-dimensional object detection method, three-dimensional object detection device, electronic apparatus, and medium
CN116502785B (en) Warehouse logistics intelligent management method, device, equipment and storage medium
JP2017019586A (en) Physical distribution system and physical distribution management method
US20230290056A1 (en) Systems and Methods for Fine Adjustment of Roof Models
CN109871765A (en) A kind of article stacking detection method, system and electronic equipment based on image
CN109272050B (en) Image processing method and device
CN115205610A (en) Training method and training device for perception model and electronic equipment
CN115271149A (en) Container space optimization method and device, electronic equipment and storage medium
CN113378750A (en) Charging pile docking method and device, computer equipment and storage medium
KR102310612B1 (en) Method for predicting object of 2D image using object information of point group of a lidar, and computer program recorded on record-medium for executing method therefor
CN111739164A (en) Box body modeling method and device, robot picking system, electronic device and medium
US11498776B1 (en) Automated guided vehicle control and organizing inventory items using dissimilarity models
CN116957459A (en) Intelligent matching method, system, equipment and storage medium based on freight scene
CN113345023B (en) Box positioning method and device, medium and electronic equipment
CN110880089A (en) Knowledge graph-based warehouse space distribution method and system
CN111090110A (en) Method for acquiring or providing VRS service data, user equipment and storage medium

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