Intelligent storage machine object recognition and sorting system and method

Abstract

This invention relates to the field of intelligent warehousing technology, specifically to an intelligent warehousing machine sorting and identification system and method, comprising: a collection layer, an analysis layer, and a transfer layer; when goods are loaded onto a truck, the collection layer collects the outline information of the goods transported on the conveyor belt, and constructs a three-dimensional model of the goods based on the outline information; the analysis layer uploads the specifications of the cargo container, and the specifications of the cargo container are synchronously fed back to the collection layer; the collection layer constructs a three-dimensional model of the cargo container based on the specifications of the cargo container; by constructing a three-dimensional model of the goods, this invention further uses the three-dimensional model of the cargo container as reference data to bring about a planning effect on the stacking order of goods when stacking them inside the container, thereby stacking goods more effectively, ensuring that the goods are stacked reasonably and compactly inside the container, reducing or eliminating the waste of loading space inside the container, and thus achieving the effect of reducing the cost of goods transportation.

Description

Technical Field

[0001] The present invention relates to the field of intelligent warehousing technology, and in particular to an intelligent warehousing machine object recognition and sorting system and method. Background Art

[0002] Intelligent warehousing and logistics uses advanced technologies and equipment to achieve automation, intelligence and efficiency in warehousing and logistics processes.

[0003] The invention patent with application number 201910704536.4 discloses an intelligent sorting system, characterized in that the system comprises: a storage robot, the storage robot is used to receive a first transport instruction, the first transport instruction includes first material box information and / or cargo information to be transported, and according to the first transport instruction, transport the first material box and / or cargo to a preset position on a preset shelf, the preset position is used to place the first material box and / or cargo, and the preset position corresponds to the shipping destination of the first material box and / or cargo;

[0004] The storage robot is further used to: lock the sorting grid where the target material box to be replaced is located, and according to the material box replacement instruction, move to the sorting grid where the target material box is located; after carrying the empty material box to the sorting grid, send a sorting grid release instruction and / or a material box replacement completion instruction to release the sorting grid, wherein the sorting grid is in a locked state during the material box replacement process;

[0005] The material box replacement instruction includes the storage location information of the empty material box, and the storage robot is also used to: move to the empty material box storage location according to the material box replacement instruction, and instruct the handling component to store the empty material box into the storage unit; wherein, there are multiple storage units, and the storage robot grabs one empty material box at a time at the storage location of the empty material box and places it in the corresponding storage unit, and grabs the empty material box according to the number of target material boxes that need to be replaced.

[0006] This application aims to solve the problem that "manual movement of material boxes to the pallets or temporary storage locations corresponding to the shipping destination affects the efficiency of cargo sorting."

[0007] However, in the process of intelligent warehousing and shipping, most of the cargo loading is still done manually. The goods to be shipped are stacked and packed, and the goods are transported after the goods fill the cargo container. In this process, the stacking of goods into the cargo container is often done by the staff based on previous cargo stacking experience. This method cannot maximize the rationality of cargo stacking, resulting in a certain waste of internal space in the cargo container, which in turn increases the cost of cargo transportation. Summary of the invention

[0008] In view of the above-mentioned shortcomings of the prior art, the present invention provides an intelligent storage machine object recognition and sorting system and method, which solves the technical problems raised in the above-mentioned background technology.

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0010] In the first aspect, an intelligent storage machine object recognition and sorting system includes: a collection layer, an analysis layer, and a circulation layer;

[0011] When the warehouse goods are loaded, the goods transmitted on the conveyor belt are collected through the collection layer to collect the contour information of the goods, and the three-dimensional model of the goods is constructed based on the contour information of the goods. The analysis layer is loaded with the specification parameters of the cargo container, and the specification parameters of the cargo container are synchronously fed back to the collection layer. The collection layer is applied to construct the three-dimensional model of the cargo container based on the specification parameters of the cargo container. The analysis layer synchronously receives the three-dimensional model of the cargo, analyzes the loading adaptability of the cargo corresponding to the three-dimensional model of the cargo, and feeds back the loading adaptability of the cargo corresponding to the three-dimensional model of the cargo to the circulation layer. The circulation layer decides the cargo transmission target based on the loading adaptability of the cargo.

[0012] The analysis layer includes a receiving module, an analysis module and a refresh module. The receiving module is used to receive the three-dimensional model of the cargo constructed in the acquisition layer and send the three-dimensional model of the cargo to the analysis module. The analysis module is used to obtain the three-dimensional model of the cargo fed back in the receiving module and perform loading adaptability analysis on the three-dimensional model of the cargo. The refresh module is used to refresh the operation of the analysis layer.

[0013] The loading adaptability analysis logic of the cargo three-dimensional model in the analysis module is expressed as follows:

[0014]

[0015] Where: f is the load adaptability judgment value of the cargo three-dimensional model; S box is a set of inner surfaces of the container 3D model; S goods is a set of surfaces of the three-dimensional model of the cargo; f′ is the correction result of the loading adaptability judgment value f of the three-dimensional model of the cargo; χ is the correction factor;

[0016] Among them, SIM(S box ,S goods ) down The SIM(S) represents the similarity between the bottom surface of the inner surface of the container 3D model and the bottom surface of the cargo 3D model. box ,S goods ) facadeIndicates the similarity between the facade of the inner surface of the container 3D model that is attached to the cargo and the facade of the cargo 3D model. The correction factor value is 0.1>χ≥0, and the more cargo currently loaded in the cargo container, the smaller the value of the correction factor χ. On the contrary, the larger the value of the correction factor χ, f′≥1, indicating that the cargo 3D model corresponding to the cargo is suitable for loading. On the contrary, it indicates that the cargo 3D model corresponding to the cargo is not suitable for the current loading.

[0017] The receiving module is interactively connected to the building module via a wireless network, the building module is interactively connected to the identification module and the sensor module via a wireless network, the receiving module is interactively connected to the analysis module and the refresh module via a wireless network, the refresh module is interactively connected to the determination module via a wireless network, and the determination module is interactively connected to the transmission module via a wireless network.

[0018] Furthermore, the acquisition layer includes a sensing module, an identification module and a construction module, the sensing module is used to sense the distance from its own ranging end to the surface of the cargo, the identification module is used to receive and traverse the distance from its own ranging end to the surface of the cargo sensed by the sensing module, identify available distance parameters, and filter available distance parameters based on the identification results, and the construction module is used to obtain the available distance parameters identified in the identification module, and use the available distance parameters to construct a three-dimensional model of the cargo;

[0019] Among them, the sensor module is a ranging sensor, and the sensor module is provided with several groups. The sensor modules of the several groups are deployed in an equidistant array on the left, right and upper sides of the cargo conveyor belt. The sensor modules deployed on the left and right sides and the upper side of the cargo conveyor belt in the identification module are respectively provided with two groups of comparison distance parameters, and the two groups of comparison distance parameters are respectively the spacing between the sensor modules deployed on the left and right sides of the cargo conveyor belt and the distance from the sensor module deployed on the upper side to the surface of the cargo conveyor belt. When the identification module selects the available distance parameters based on the identification result, the source ranging module deployment position of the distance from the ranging end to the cargo surface is first obtained, and then the corresponding comparison distance parameter is selected according to the ranging module deployment position. The selected comparison distance parameter is compared with the distance from the ranging end to the cargo surface sensed by the ranging module. When the two groups of distances are equal, the distance from the ranging end to the cargo surface sensed by the ranging module is discarded. Otherwise, the distance from the ranging end to the cargo surface sensed by the ranging module is the available distance parameter screened by the identification module.

[0020] Furthermore, any two sets of sensor modules closest to the input end of the cargo conveyor belt run in real time to perform the distance measurement task. When the distance measured by its own distance measurement end is sensed for the first time and both are the two sets of comparison distance parameters set in the non-identification module, the operating frequency of the sensor module is controlled to double. When the operating frequency is doubled, when the sensor module senses that the distance measured by its own distance measurement end is both the two sets of comparison distance parameters set in the identification module, the sensor module currently performing the distance measurement task stops running, and synchronously triggers all sensor modules to run synchronously once to perform the distance measurement task.

[0021] Among them, when the sensor module is synchronously triggered to perform the ranging task, the ranging result is received and processed by the recognition module. When the recognition module and the construction module receive or transmit the distance parameter, they are transmitted in sequence according to the deployment order of the ranging module from which the distance parameter comes.

[0022] Furthermore, when constructing the three-dimensional model of the cargo, the construction module determines the construction space of the three-dimensional model of the cargo based on the two sets of compared distance parameters, determines the contour of the cargo based on the available distance parameters screened by the recognition module, and constructs the three-dimensional model of the cargo using the contour of the cargo and the available distance parameters.

[0023] Furthermore, the receiving module is further used to upload the specification parameters of the cargo container. The specification parameters of the cargo container uploaded in the receiving module are synchronously sent to the construction module in the collection layer. The construction module constructs a three-dimensional model of the cargo container based on the specification parameters of the cargo container, and further feeds the constructed three-dimensional model of the cargo container back to the analysis module for storage.

[0024] The contact surface between the goods and the conveyor belt surface, the contact surface between the goods and the cargo container placement surface, and the cargo surface corresponding to the bottom surface of the cargo three-dimensional model are all the same surface of the goods. The refresh module is manually controlled and run by the system end user. After each row of goods is stacked in the cargo container, the refresh analysis layer runs.

[0025] Furthermore, when the refresh module refreshes the analysis layer, the refresh module synchronously obtains the three-dimensional model of the cargo container and the three-dimensional model of the stacked cargo, and based on the stacking position in the cargo container, segments the three-dimensional model of the cargo corresponding to the stacked cargo at the corresponding position in the three-dimensional model of the cargo container, and segments the remaining three-dimensional model of the cargo container and sends it to the analysis module, and iterates the three-dimensional model of the original cargo container stored in the analysis module;

[0026] Among them, when stacking goods in cargo containers, they follow the stacking logic from inside to outside and from bottom to top.

[0027] Furthermore,

[0028] Where: SIM(S box ,S goods) is the similarity between a set of inner surfaces of the container 3D model and a set of surfaces of the cargo 3D model; Lbox, L goods W is the length of a set of inner surfaces of the container 3D model and a set of surfaces of the cargo 3D model; box , W goods is the width of a set of inner surfaces of the container 3D model and a set of surfaces of the cargo 3D model; n is the set of angles in the surface figure; (r i ) box is the angle value of the i-th group of angles in a set of inner surfaces of the container 3D model; (r i ) goods is the angle value of the i-th group of angles in a set of surfaces of the three-dimensional model of the cargo;

[0029] SIM(S box ,S goods ) is closer to 1, indicating that the similarity of the two groups of surfaces is higher. When the refresh module runs and the refresh analysis layer runs, the analysis module runs again. When analyzing the loading adaptability of the cargo three-dimensional model, SIM(S box ,S goods ) down Where Sbox can be the upper surface of the three-dimensional model of the cargo corresponding to the uppermost cargo in any group of cargo containers. SIM(S box ,S goods ) facade The source of Sbox is the inner facade of the three-dimensional model of the cargo container corresponding to the inner facade of the cargo container attached to the cargo among the inner facades of the latest three-dimensional model of the cargo container stored in the analysis module;

[0030] Among them, SIM(S box ,S goods ) facade After determining the inner facade of the three-dimensional model of the cargo container, Sbox obtains all areas in the three-dimensional model of the cargo container that represent the upper surface of the three-dimensional model of the cargo corresponding to the uppermost layer of the cargo. The upper space corresponds to the local model, and the facade of the local model is Sbox.

[0031] Furthermore, the circulation layer includes a determination module and a transmission module, the determination module is used to receive the loading adaptability analysis result of the three-dimensional model of the goods in the analysis module, and determine whether the goods corresponding to the three-dimensional model of the goods are suitable for loading based on the analysis result, and the transmission module is used to obtain the determination result of whether the three-dimensional model of the goods is suitable for loading in the determination module, and further transmit the goods corresponding to the three-dimensional model of the goods based on the determination result;

[0032] Among them, the transmission module is integrated by a conveyor belt, and the conveyor belt has the same function as the cargo conveyor belt. When the transmission module transmits the cargo, the transmission target is the initial transmission position of the cargo conveyor belt or the cargo container. When the determination result of the determination module is yes, the transmission target of the cargo by the transmission module is the cargo container. When the determination result of the determination module is no, the transmission target of the cargo by the transmission module is the initial transmission position of the cargo conveyor belt.

[0033] In a second aspect, a method for intelligent warehousing machine object recognition and sorting includes the following steps:

[0034] Step 1: Obtain the contour information of the goods being transported on the goods conveyor belt, and construct a three-dimensional model of the goods based on the contour information of the goods;

[0035] Step 11: Setting the logic of building the cargo 3D model;

[0036] Step 2: Upload the specification parameters of the cargo container, and build a three-dimensional model of the cargo container based on the specification parameters of the cargo container;

[0037] Step 3: Analyze the loading adaptability of the cargo 3D model corresponding to the cargo by combining the cargo container 3D model and the cargo 3D model;

[0038] Step 31: Cargo loading adaptability setting stage;

[0039] Step 4: Determine whether the cargo is suitable for the current loading of the cargo container based on the cargo three-dimensional model of the cargo loading adaptability analysis result;

[0040] Step 5: Obtaining a determination result of whether the cargo is suitable for current loading in the cargo container, and making a decision on further transport of the cargo based on the determination result;

[0041] Step 51: Decision logic setting phase.

[0042] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0043] The present invention provides an intelligent storage machine object recognition and sorting system. During operation, the system constructs a three-dimensional model of the goods by collecting the contour information of the goods, and further uses the three-dimensional model of the cargo container as reference data to bring a planning effect to the stacking sequence of the goods when stacking inside the container, so as to more effectively stack the goods, ensure that the goods stacked in the container are reasonable and compact, and waste of the loading space inside the container is not easy or less likely to occur, thereby achieving the effect of reducing the cost of cargo transportation, and thus, the cargo stacking inside the container can be made more compact, thereby improving the stability of the container during the loading and transportation process;

[0044] Furthermore, an intelligent storage machine object recognition and sorting method is used to assist the operation of the above system, making the system operation logic more complete, and making the technical solution composed of the system and method more stable and effective in the actual application stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0046] Figure 1 This is a structural diagram of an intelligent warehousing machine object recognition and sorting system;

[0047] Figure 2 A schematic diagram of a process for an intelligent warehousing machine to identify and sort objects;

[0048] Figure 3 It is a schematic diagram of the deployment status of the hardware modules of the system collection layer and the circulation layer and the cargo conveyor belt in the present invention;

[0049] Figure 4 This is a diagram showing the construction principle of the three-dimensional model of the cargo in the present invention;

[0050] Figure 5 A schematic diagram of a three-dimensional model of a cargo container stored in an iterative analysis module of the present invention;

[0051] The numbers in the figure represent: 1. conveyor belt; 2. distance measuring sensor; 3. cargo container; 4. conveyor belt one; 5. conveyor belt two. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] The present invention will be further described below in conjunction with the embodiments.

[0054] Embodiment 1:

[0055] An intelligent storage machine object recognition and sorting system of this embodiment, such as Figure 1 As shown, it includes: a collection layer, an analysis layer and a circulation layer;

[0056] When the warehouse goods are loaded, the goods transmitted on the conveyor belt are collected through the collection layer to collect the contour information of the goods, and the three-dimensional model of the goods is constructed based on the contour information of the goods. The analysis layer is loaded with the specification parameters of the cargo container, and the specification parameters of the cargo container are synchronously fed back to the collection layer. The collection layer is applied to construct the three-dimensional model of the cargo container based on the specification parameters of the cargo container. The analysis layer synchronously receives the three-dimensional model of the cargo, analyzes the loading adaptability of the cargo corresponding to the three-dimensional model of the cargo, and feeds back the loading adaptability of the cargo corresponding to the three-dimensional model of the cargo to the circulation layer. The circulation layer decides the cargo transmission target based on the loading adaptability of the cargo.

[0057] The acquisition layer includes a sensing module, an identification module and a construction module. The sensing module is used to sense the distance from its own ranging end to the surface of the cargo. The identification module is used to receive and traverse the distance from its own ranging end to the surface of the cargo sensed by the sensing module, identify the available distance parameters, and filter the available distance parameters based on the identification results. The construction module is used to obtain the available distance parameters identified in the identification module and use the available distance parameters to construct a three-dimensional model of the cargo.

[0058] Among them, the sensor module is a distance measuring sensor, and the sensor module is provided with several groups, and the several groups of sensor modules are deployed in an equidistant array on the left, right and upper sides of the cargo conveyor belt. The sensor modules deployed on the left and right sides and the upper side of the cargo conveyor belt in the identification module are respectively provided with two groups of comparison distance parameters, and the two groups of comparison distance parameters are respectively the spacing between the sensor modules deployed on the left and right sides of the cargo conveyor belt and the distance from the sensor module deployed on the upper side to the surface of the cargo conveyor belt. When the identification module selects the available distance parameters based on the identification results, the source distance measuring module deployment position of the distance from the ranging end to the cargo surface is first obtained, and then the corresponding comparison distance parameter is selected according to the ranging module deployment position. The selected comparison distance parameter is compared with the distance from the ranging end to the cargo surface sensed by the ranging module. When the two groups of distances are equal, the distance from the ranging end to the cargo surface sensed by the ranging module is discarded. Otherwise, the distance from the ranging end to the cargo surface sensed by the ranging module is the available distance parameter selected by the identification module.

[0059] The analysis layer includes a receiving module, an analysis module and a refresh module. The receiving module is used to receive the three-dimensional model of the cargo constructed in the acquisition layer and send the three-dimensional model of the cargo to the analysis module. The analysis module is used to obtain the three-dimensional model of the cargo fed back in the receiving module and perform loading adaptability analysis on the three-dimensional model of the cargo. The refresh module is used to refresh the operation of the analysis layer.

[0060] The loading adaptability analysis logic of the cargo three-dimensional model in the analysis module is expressed as follows:

[0061]

[0062] Where: f is the load adaptability judgment value of the cargo three-dimensional model; S boxis a set of inner surfaces of the container 3D model; S goods is a set of surfaces of the three-dimensional model of the cargo; f′ is the correction result of the loading adaptability judgment value f of the three-dimensional model of the cargo; χ is the correction factor;

[0063] Among them, SIM(S box ,S goods ) down The SIM(S) represents the similarity between the bottom surface of the inner surface of the container 3D model and the bottom surface of the cargo 3D model. box ,S goods ) facade Indicates the similarity between the facade of the inner surface of the container 3D model that is attached to the cargo and the facade of the cargo 3D model. The correction factor value is 0.1>χ≥0, and the more cargo currently loaded in the cargo container, the smaller the value of the correction factor χ. On the contrary, the larger the value of the correction factor χ, f′≥1, indicating that the cargo 3D model corresponding to the cargo is suitable for loading. On the contrary, it indicates that the cargo 3D model corresponding to the cargo is not suitable for the current loading.

[0064] The circulation layer includes a determination module and a transmission module. The determination module is used to receive the loading adaptability analysis result of the three-dimensional cargo model in the analysis module, and determine whether the cargo corresponding to the three-dimensional cargo model is suitable for loading based on the analysis result. The transmission module is used to obtain the determination result of whether the cargo corresponding to the three-dimensional cargo model is suitable for loading in the determination module, and further transmit the cargo corresponding to the three-dimensional cargo model based on the determination result.

[0065] Wherein, the transmission module is integrated by a conveyor belt, and the conveyor belt has the same function as the cargo conveyor belt. When the transmission module transmits cargo, the transmission target is the initial transmission position of the cargo conveyor belt or the cargo container. When the determination result of the determination module is yes, the transmission target of the cargo by the transmission module is the cargo container. When the determination result of the determination module is no, the transmission target of the cargo by the transmission module is the initial transmission position of the cargo conveyor belt.

[0066] The receiving module is interactively connected to the building module through a wireless network, the building module is interactively connected to the identification module and the sensor module through a wireless network, the receiving module is interactively connected to the analysis module and the refresh module through a wireless network, the refresh module is interactively connected to the determination module through a wireless network, and the determination module is interactively connected to the transmission module through a wireless network.

[0067] In this embodiment, the sensing module operates to sense the distance from its own ranging end to the surface of the cargo, the identification module synchronously receives and traverses the distance from its own ranging end to the surface of the cargo sensed by the sensing module, identifies the available distance parameters, and selects the available distance parameters based on the identification results. The construction module obtains the available distance parameters identified in the identification module in real time, and applies the available distance parameters to construct a three-dimensional model of the cargo. The receiving module then receives the three-dimensional model of the cargo constructed in the acquisition layer, and sends the three-dimensional model of the cargo to the analysis module. The analysis module further obtains the three-dimensional model of the cargo fed back in the receiving module, and performs loading adaptability analysis on the three-dimensional model of the cargo. The refresh module refreshes the analysis layer in real time. Finally, the determination module receives the loading adaptability analysis result of the three-dimensional model of the cargo in the analysis module, and determines whether the cargo corresponding to the three-dimensional model of the cargo is suitable for loading based on the analysis result. The transmission module synchronously obtains the determination result of whether the cargo corresponding to the three-dimensional model of the cargo is suitable for loading in the determination module, and further transmits the cargo corresponding to the three-dimensional model of the cargo based on the determination result.

[0068] Based on the operation of the system in the above embodiment, an intelligent control effect is brought to the stacking container of logistics goods in intelligent warehousing, ensuring that the goods to be transported can be stacked more reasonably inside the container, thereby achieving the purpose of saving goods transportation costs and ensuring stable goods transportation;

[0069] See also Figure 3 As shown in the figure, the deployment of the sensor module and the transmission module in the system is further demonstrated. It should be noted that based on Figure 3 The transmission belt 1 4 and the transmission belt 2 5 are shown to represent the transmission module;

[0070] See also Figure 4 As shown, the external three-dimensional figure in the figure represents the construction space of the three-dimensional model of the goods, the internal three-dimensional figure in the figure represents the three-dimensional model of the goods, and the lines in the figure represent the distance measured by the sensor module;

[0071] See also Figure 5 As shown in FIG. 1 , the three-dimensional model of the cargo container stored in the iterative analysis module is further displayed, and based on the arrows in the figure, it is further helpful to understand the "when analyzing the loading adaptability of the three-dimensional model of the cargo, SIM (S box ,S goods ) down Where Sbox can be the upper surface of the three-dimensional model of the cargo corresponding to the uppermost cargo in any group of cargo containers. SIM(S box ,S goods ) facade The source of Sbox is the inner facade of the three-dimensional model of the cargo container corresponding to the inner facade of the cargo container attached to the cargo among the inner facades of the latest three-dimensional model of the cargo container stored in the analysis module;

[0072] SIM(S box ,S goods ) facade After determining the inner facade of the three-dimensional model of the cargo container, Sbox obtains all areas in the three-dimensional model of the cargo container that represent the upper surface of the three-dimensional model of the cargo corresponding to the cargo loaded on the top layer. The upper space corresponds to the local model, and the facade of the local model is Sbox.

[0073] Embodiment 2:

[0074] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The intelligent storage machine object recognition and sorting system in Example 1 is further described in detail:

[0075] Any two sets of sensor modules closest to the input end of the cargo conveyor belt run in real time to perform the distance measurement task. When the first time they sense that the distance measured by their own distance measurement end is the two sets of comparison distance parameters set in the non-identification module, they control the operating frequency of the sensor module to double. When the operating frequency is doubled, when the sensor module senses that the distance measured by its own distance measurement end is the two sets of comparison distance parameters set in the identification module, the sensor module currently performing the distance measurement task stops running, and synchronously triggers all sensor modules to run synchronously once to perform the distance measurement task.

[0076] Among them, when the sensor module is synchronously triggered to perform the ranging task, the ranging result is received and processed by the recognition module. When the recognition module and the construction module receive or transmit the distance parameter, they are transmitted in sequence according to the deployment order of the ranging module from which the distance parameter comes;

[0077] When constructing the three-dimensional model of the cargo, the construction module determines the construction space of the three-dimensional model of the cargo based on two sets of compared distance parameters, determines the contour of the cargo based on the available distance parameters screened by the recognition module, and constructs the three-dimensional model of the cargo using the contour of the cargo and the available distance parameters.

[0078] Through the above settings, the deployment structure of the sensor module in the system is further limited to ensure that the acquisition layer in the system can stably complete the construction of the three-dimensional model of the goods.

[0079] like Figure 1 As shown, the receiving module is also used to upload the specification parameters of the cargo container. The specification parameters of the cargo container uploaded in the receiving module are synchronously sent to the construction module in the acquisition layer. The construction module constructs a three-dimensional model of the cargo container based on the specification parameters of the cargo container, and further feeds the constructed three-dimensional model of the cargo container back to the analysis module for storage;

[0080] The contact surface between the goods and the conveyor belt surface, the contact surface between the goods and the cargo container placement surface, and the bottom surface of the cargo three-dimensional model corresponding to the cargo surface are all the same surface of the goods. The refresh module is manually controlled by the system end user to run. After each row of goods is stacked in the cargo container, the refresh analysis layer runs;

[0081] When the refresh module refreshes the analysis layer, the refresh module synchronously obtains the three-dimensional model of the cargo container and the three-dimensional model of the stacked cargo, and based on the stacking position in the cargo container, divides the three-dimensional model of the cargo corresponding to the stacked cargo at the corresponding position in the three-dimensional model of the cargo container, divides the remaining three-dimensional model of the cargo container and sends it to the analysis module, and iterates the three-dimensional model of the original storage cargo container in the analysis module;

[0082] Among them, when stacking goods in cargo containers, they follow the stacking logic from inside to outside and from bottom to top;

[0083]

[0084] Where: SIM(S box ,S goods ) is the similarity between a set of inner surfaces of the container 3D model and a set of surfaces of the cargo 3D model; Lbox, L goods W is the length of a set of inner surfaces of the container 3D model and a set of surfaces of the cargo 3D model; box , W goods is the width of a set of inner surfaces of the container 3D model and a set of surfaces of the cargo 3D model; n is the set of angles in the surface figure; (r i ) box is the angle value of the i-th group of angles in a set of inner surfaces of the container 3D model; (r i ) goods is the angle value of the i-th group of angles in a set of surfaces of the three-dimensional model of the cargo;

[0085] SIM(S box ,S goods ) is closer to 1, indicating that the similarity of the two groups of surfaces is higher. When the refresh module runs and the refresh analysis layer runs, the analysis module runs again. When analyzing the loading adaptability of the cargo three-dimensional model, SIM(S box ,S goods ) down Where Sbox can be the upper surface of the three-dimensional model of the cargo corresponding to the uppermost cargo in any group of cargo containers. SIM(S box ,S goods ) facade The source of Sbox is the inner facade of the three-dimensional model of the cargo container corresponding to the inner facade of the cargo container attached to the cargo among the inner facades of the latest three-dimensional model of the cargo container stored in the analysis module;

[0086] Among them, SIM(S box ,S goods ) facade After determining the inner facade of the three-dimensional model of the cargo container, Sbox obtains all areas in the three-dimensional model of the cargo container that represent the upper surface of the three-dimensional model of the cargo corresponding to the uppermost layer of the cargo. The upper space corresponds to the local model, and the facade of the local model is Sbox.

[0087] In this embodiment, based on the above settings, the system in Embodiment 1 is further limited in operation logic, and The setting provides further operation logic parameter support for the loading adaptability analysis logic of the cargo three-dimensional model in the analysis module.

[0088] Embodiment 3:

[0089] In terms of specific implementation, based on Example 1, this example refers to Figure 2 The intelligent storage machine object recognition and sorting system in Example 1 is further described in detail:

[0090] An intelligent storage machine object recognition and sorting method comprises the following steps:

[0091] Step 1: Obtain the contour information of the goods being transported on the goods conveyor belt, and construct a three-dimensional model of the goods based on the contour information of the goods;

[0092] Step 11: Setting the logic of building the cargo 3D model;

[0093] Step 2: Upload the specification parameters of the cargo container, and build a three-dimensional model of the cargo container based on the specification parameters of the cargo container;

[0094] Step 3: Analyze the loading adaptability of the cargo 3D model corresponding to the cargo by combining the cargo container 3D model and the cargo 3D model;

[0095] Step 31: Cargo loading adaptability setting stage;

[0096] Step 4: Determine whether the cargo is suitable for the current loading of the cargo container based on the cargo three-dimensional model of the cargo loading adaptability analysis result;

[0097] Step 5: Obtaining a determination result of whether the cargo is suitable for current loading in the cargo container, and making a decision on further transport of the cargo based on the determination result;

[0098] Step 51: Decision logic setting phase.

[0099] In summary, during the application of the system and method in the above embodiments, a three-dimensional model of the cargo is constructed by collecting cargo contour information, and the three-dimensional model of the cargo container is further used as reference data to bring a planning effect to the stacking order of the cargo when stacking inside the container, so as to stack the cargo more effectively, ensure that the cargo in the container is stacked reasonably and tightly, and waste of the loading space inside the container is less or less likely to occur, thereby achieving the effect of reducing the cost of cargo transportation, and thereby, the cargo stacking inside the container can be made more compact, thereby improving the stability of the container during the loading and transportation process.

[0100] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent storage machine object recognition and sorting system, characterized in that: include: Collection layer, analysis layer and circulation layer; When the warehouse goods are loaded, the goods transmitted on the conveyor belt are collected through the collection layer to collect the contour information of the goods, and the three-dimensional model of the goods is constructed based on the contour information of the goods. The analysis layer is loaded with the specification parameters of the cargo container, and the specification parameters of the cargo container are synchronously fed back to the collection layer. The collection layer is applied to construct the three-dimensional model of the cargo container based on the specification parameters of the cargo container. The analysis layer synchronously receives the three-dimensional model of the cargo, analyzes the loading adaptability of the cargo corresponding to the three-dimensional model of the cargo, and feeds back the loading adaptability of the cargo corresponding to the three-dimensional model of the cargo to the circulation layer. The circulation layer decides the cargo transmission target based on the loading adaptability of the cargo. The analysis layer includes a receiving module, an analysis module and a refresh module. The receiving module is used to receive the three-dimensional model of the cargo constructed in the acquisition layer and send the three-dimensional model of the cargo to the analysis module. The analysis module is used to obtain the three-dimensional model of the cargo fed back in the receiving module and perform loading adaptability analysis on the three-dimensional model of the cargo. The refresh module is used to refresh the operation of the analysis layer. The loading adaptability analysis logic of the cargo three-dimensional model in the analysis module is expressed as follows: ； Where: Loading adaptability judgment value for the cargo three-dimensional model; is a set of inner surfaces of the three-dimensional model of the container; A set of surfaces of a three-dimensional model of the cargo; Loading adaptability judgment value for cargo 3D model The correction result of is the correction factor; in, Indicates the similarity between the bottom surface of the inner surface of the container 3D model and the bottom surface of the cargo 3D model. Indicates the similarity between the facade of the inner surface of the container 3D model that is attached to the cargo and the facade of the cargo 3D model. The correction factor is set at 0.1> ≥0, and subject to the cargo container currently loaded with more cargo, the correction factor The smaller the value of The larger the value, ≥1, indicating that the cargo 3D model corresponds to the cargo suitable for loading; conversely, indicating that the cargo 3D model corresponds to the cargo not suitable for current loading; The receiving module is also used to upload the specification parameters of the cargo container. The specification parameters of the cargo container uploaded in the receiving module are synchronously sent to the construction module in the collection layer. The construction module constructs a three-dimensional model of the cargo container based on the specification parameters of the cargo container, and further feeds the constructed three-dimensional model of the cargo container to the analysis module for storage; The contact surface between the goods and the conveyor belt surface, the contact surface between the goods and the cargo container placement surface, and the bottom surface of the cargo three-dimensional model corresponding to the cargo surface are all the same surface of the goods. The refresh module is manually controlled by the system end user to run. After each row of goods is stacked in the cargo container, the refresh analysis layer runs; When the refresh module refreshes the analysis layer, the refresh module synchronously obtains the three-dimensional model of the cargo container and the three-dimensional model of the stacked cargo, and based on the stacking position in the cargo container, divides the three-dimensional model of the cargo corresponding to the stacked cargo at the corresponding position in the three-dimensional model of the cargo container, divides the remaining three-dimensional model of the cargo container and sends it to the analysis module, and iterates the three-dimensional model of the original cargo container stored in the analysis module; Among them, when stacking goods in cargo containers, they follow the stacking logic from inside to outside and from bottom to top.

2. According to claim 1, the intelligent storage machine object recognition and sorting system is characterized in that: The acquisition layer includes a sensing module, an identification module and a construction module. The sensing module is used to sense the distance from its own ranging end to the surface of the goods. The identification module is used to receive and traverse the distance from its own ranging end to the surface of the goods sensed by the sensing module, identify the available distance parameters, and filter the available distance parameters based on the identification results. The construction module is used to obtain the available distance parameters identified in the identification module and use the available distance parameters to construct a three-dimensional model of the goods. Among them, the sensor module is a ranging sensor, and the sensor module is provided with several groups. The sensor modules of the several groups are deployed in an equidistant array on the left, right and upper sides of the cargo conveyor belt. The sensor modules deployed on the left and right sides and the upper side of the cargo conveyor belt in the identification module are respectively provided with two groups of comparison distance parameters, and the two groups of comparison distance parameters are respectively the spacing between the sensor modules deployed on the left and right sides of the cargo conveyor belt and the distance from the sensor module deployed on the upper side to the surface of the cargo conveyor belt. When the identification module selects the available distance parameters based on the identification result, the source ranging module deployment position of the distance from the ranging end to the cargo surface is first obtained, and then the corresponding comparison distance parameter is selected according to the ranging module deployment position. The selected comparison distance parameter is compared with the distance from the ranging end to the cargo surface sensed by the ranging module. When the two groups of distances are equal, the distance from the ranging end to the cargo surface sensed by the ranging module is discarded. Otherwise, the distance from the ranging end to the cargo surface sensed by the ranging module is the available distance parameter screened by the identification module.

3. The intelligent storage machine object recognition and sorting system according to claim 2 is characterized in that: Any two sets of sensor modules closest to the input end of the cargo conveyor belt run in real time to perform the distance measurement task. When the first time they sense that the distance measured by their own distance measurement end is the two sets of comparison distance parameters set in the non-identification module, they control the operating frequency of the sensor module to double. When the operating frequency is doubled, when the sensor module senses that the distance measured by its own distance measurement end is the two sets of comparison distance parameters set in the identification module, the sensor module currently performing the distance measurement task stops running, and synchronously triggers all sensor modules to run synchronously once to perform the distance measurement task. Among them, when the sensor module is synchronously triggered to perform the ranging task, the ranging result is received and processed by the recognition module. When the recognition module and the construction module receive or transmit the distance parameter, they are transmitted in sequence according to the deployment order of the ranging module from which the distance parameter comes.

4. The intelligent storage machine object recognition and sorting system according to claim 3 is characterized in that: When constructing the three-dimensional model of the cargo, the construction module determines the construction space of the three-dimensional model of the cargo based on the two sets of compared distance parameters, determines the contour of the cargo based on the available distance parameters screened by the recognition module, and constructs the three-dimensional model of the cargo using the contour of the cargo and the available distance parameters.

5. The intelligent storage machine object recognition and sorting system according to claim 4 is characterized in that: ； Where: is the similarity between a set of inner surfaces of the container 3D model and a set of surfaces of the cargo 3D model; , is the length of a set of inner surfaces of the container 3D model and a set of surfaces of the cargo 3D model; , is the width of a set of inner surfaces of the container 3D model and a set of surfaces of the cargo 3D model; is the set of angles in the surface figure; is the angle value of the i-th group of angles in a set of inner surfaces of the three-dimensional model of the container; is the angle value of the i-th group of angles in a set of surfaces of the three-dimensional model of the cargo; The closer the value is to 1, the higher the similarity of the two sets of surfaces. When the refresh module runs and the refresh analysis layer runs, the analysis module runs again and analyzes the loading adaptability of the three-dimensional model of the cargo. middle It can be the upper surface of the three-dimensional model of the cargo corresponding to the uppermost cargo loaded in any group of cargo containers. middle The source is the three-dimensional model inner facade of the cargo container corresponding to the inner facade of the cargo container attached to the cargo among the latest three-dimensional model inner facades of the cargo container stored in the analysis module; in, middle After determining the interior elevation of the cargo container 3D model, obtain all areas in the cargo container 3D model that represent the uppermost layer of cargo loading corresponding to the upper surface of the cargo 3D model. The upper space corresponds to the local model. The elevation of the local model is .

6. The intelligent storage machine object recognition and sorting system according to claim 5 is characterized in that: The circulation layer includes a determination module and a transmission module. The determination module is used to receive the loading adaptability analysis result of the three-dimensional cargo model in the analysis module, and determine whether the cargo corresponding to the three-dimensional cargo model is suitable for loading based on the analysis result. The transmission module is used to obtain the determination result of whether the cargo corresponding to the three-dimensional cargo model is suitable for loading in the determination module, and further transmit the cargo corresponding to the three-dimensional cargo model based on the determination result. Among them, the transmission module is integrated by a conveyor belt, and the conveyor belt has the same function as the cargo conveyor belt. When the transmission module transmits the cargo, the transmission target is the initial transmission position of the cargo conveyor belt or the cargo container. When the determination result of the determination module is yes, the transmission target of the cargo by the transmission module is the cargo container. When the determination result of the determination module is no, the transmission target of the cargo by the transmission module is the initial transmission position of the cargo conveyor belt.

7. The intelligent storage machine object recognition and sorting system according to claim 6 is characterized in that: The receiving module is interactively connected to the building module via a wireless network, the building module is interactively connected to the identification module and the sensor module via a wireless network, the receiving module is interactively connected to the analysis module and the refresh module via a wireless network, the refresh module is interactively connected to the determination module via a wireless network, and the determination module is interactively connected to the transmission module via a wireless network.

8. A method for intelligent warehousing machine object recognition and sorting, the method is an implementation method of the intelligent warehousing machine object recognition and sorting system as claimed in claim 7, characterized in that: The following steps are involved: Step 1: Obtain the contour information of the goods being transported on the goods conveyor belt, and construct a three-dimensional model of the goods based on the contour information of the goods; Step 11: Setting the logic of building the cargo 3D model; Step 2: Upload the specification parameters of the cargo container, and build a three-dimensional model of the cargo container based on the specification parameters of the cargo container; Step 3: Analyze the loading adaptability of the cargo 3D model corresponding to the cargo by combining the cargo container 3D model and the cargo 3D model; Step 31: Cargo loading adaptability setting stage; Step 4: Determine whether the cargo is suitable for the current loading of the cargo container based on the cargo three-dimensional model of the cargo loading adaptability analysis result; Step 5: Obtaining a determination result of whether the cargo is suitable for current loading of the cargo container, and making a decision on further transport of the cargo based on the determination result; Step 51: Decision logic setting phase.

Images