Search

CN-121073148-B - Intelligent scheduling system and method for intelligent palletizing robot

CN121073148BCN 121073148 BCN121073148 BCN 121073148BCN-121073148-B

Abstract

The invention provides an intelligent scheduling system and method of an intelligent palletizing robot, the method comprises the steps of S1, collecting material point cloud data through a depth camera, identifying the size of a material, obtaining the weight of the material, S2, calculating the volume of the material according to the size of the material, combining the weight of the material to obtain a density value, calculating the stroke parameter of a pneumatic clamping jaw, generating a palletizing scheme through an intelligent planning algorithm, S3, controlling the pneumatic clamping jaw to open and close based on the stroke parameter to grab the material, executing the palletizing scheme to finish material stacking, and S4, detecting the shape of the stacked material, and enabling a WMS storage scheduling module to distribute the material to a proper storage according to a detection result. According to the invention, aiming at the morphological characteristics of the flexible material, the size identification method combining convex hull analysis and outlier filtering is adopted, so that the measurement precision of irregular objects is improved, and the problems of slow mold changing, high damage and low scheduling efficiency of multi-specification flexible bulk materials such as tea in a stacking link are solved.

Inventors

  • SUN JUNWEN
  • HU CHENXIN
  • ZHAO ZIYI

Assignees

  • 上海交大智邦科技有限公司

Dates

Publication Date
20260505
Application Date
20251103

Claims (7)

  1. 1. An intelligent scheduling system of intelligent palletizing robot, which is characterized by comprising: the three-dimensional intelligent identification module is used for extracting size data of the materials from the collected material point cloud data; the material density estimation module is used for calculating the material density according to the size data and the weight information of the materials; The pneumatic clamping jaw stroke adapting module is used for extracting the maximum dimension of the material in the length direction as a clamping jaw opening and closing reference, dynamically increasing the bidirectional safety margin according to the density on the basis of the maximum dimension and compensating the morphological change of the flexible material during grabbing; The self-adaptive stacking scheme generation module is used for generating an optimal stacking scheme through an intelligent planning algorithm based on size data and safety margin of materials; The WMS storage scheduling module screens out candidate storage bits with adaptive size from idle storage bits according to size data and safety margin of materials, matches optimal storage bits through a double-layer optimization algorithm and issues storage bit allocation instructions; The WMS storage scheduling module acquires idle storage data through a WMS interface, establishes a standardized model containing storage codes, areas, layer heights, size limits and access priorities, takes the length and width heights of cargoes obtained through appearance detection as input, introduces safety margin parameters to correct actual requirements, and screens out candidate storage with adaptive size from the idle storage; The WMS storage scheduling module adopts a double-layer optimization algorithm to select optimal storage, calculates the height utilization rate according to constraint sequencing of access storage groups, and selects storage which is most matched with the cargo height requirement; And the IOT communication platform is used for completing the interactive control of the WMS software system and the equipment automation control system.
  2. 2. An intelligent scheduling system for an intelligent palletizing robot as recited in claim 1, wherein the process of extracting dimensional data of the material by the three-dimensional intelligent recognition module comprises: filtering abnormal values of the original point cloud data, and removing noise points which deviate from the mean value by more than 3 times of standard deviation by adopting a Z-score algorithm; Projecting the preprocessed point cloud on an XY plane, constructing a convex hull model to fit the outer contour of the material, determining the length of the material by calculating the maximum distance between the convex hull vertexes, and calculating the width by the projection difference value perpendicular to the longest axis direction; and directly extracting the maximum coordinate difference value of the point cloud in the Z-axis direction to obtain the height of the material.
  3. 3. Intelligent scheduling system for an intelligent palletizing robot according to claim 1, wherein the process of generating an optimal palletizing scheme by an intelligent planning algorithm comprises: introducing a safety allowance parameter to correct the actual size of the material; Two placing directions are automatically simulated, the quantity of single-layer receivable materials is calculated respectively, and the direction with larger receiving capacity is selected as the optimal placing mode; And determining the maximum stacking layer number according to the ratio of the limited height of the tray to the corrected height of the material and combining the maximum allowable layer number threshold.
  4. 4. An intelligent scheduling system for an intelligent palletizing robot according to claim 3, wherein the two placing directions are automatically simulated to include an original direction and a 90 ° rotation direction, and the single-layer compatible material quantity is calculated by integer division of the tray size and the corrected material size.
  5. 5. An intelligent scheduling method of an intelligent palletizing robot, based on the intelligent scheduling system of the intelligent palletizing robot as claimed in any one of claims 1-4, characterized by comprising: step S1, acquiring material point cloud data through a depth camera, identifying the size of a material, and acquiring the weight of the material; Step S2, calculating the material volume according to the material size and combining the material weight to obtain a density value, calculating the travel parameter of the pneumatic clamping jaw, dynamically increasing a bidirectional safety margin according to the density on the basis of the maximum size, compensating the form change when the flexible material is grabbed, and generating a stacking scheme through an intelligent planning algorithm; Step S3, controlling the opening and closing of the pneumatic clamping jaw based on the travel parameter to grab materials, and executing the stacking scheme to finish material stacking; and S4, detecting the appearance of the stacked materials, and enabling the WMS storage scheduling module to distribute the materials to a proper storage position according to a detection result to finish warehousing.
  6. 6. The intelligent scheduling method of an intelligent palletizing robot of claim 5, wherein the calculating process of the material size comprises: removing noise points which deviate from the mean value by more than 3 times of standard deviation by adopting a Z-score algorithm; constructing a convex hull model on an XY plane to calculate the length and width of the material; And obtaining the height of the material according to the maximum coordinate difference value in the Z-axis direction.
  7. 7. The intelligent scheduling method of an intelligent palletizing robot of claim 5, wherein the palletizing scheme includes a material placement direction, a layout per layer, a number per layer, a total number of layers, a total accommodation amount, and a remaining spatial distribution.

Description

Intelligent scheduling system and method for intelligent palletizing robot Technical Field The invention relates to the technical field of industrial robot control and intelligent logistics scheduling systems, in particular to an intelligent scheduling system and method of an intelligent palletizing robot, and especially relates to an intelligent scheduling system and method of an intelligent palletizing robot for flexible bulk materials with multiple specifications. Background The existing tea product stacking technology mostly adopts a fixed program robot, and has the following core defects: 1. The specification adaptability is poor, only can adapt to single packaging form, the transformation time is long, and the requirements of multiple specifications of materials such as tea powder, tea extract and the like (such as Hybrid packaging). The tea product industry has multiple varieties of small-batch production fields, and the traditional equipment is difficult to be compatible with material specification fluctuation (such as package size deviation due to fixed mechanical structure)。 2. The visual recognition accuracy is insufficient, namely, the 2D visual positioning error is large, the deformation characteristics of flexible bulk materials (such as the shape fluctuation of tea powder bags caused by the difference of filling quantity) cannot be accurately recognized, and the stacking stability is poor (the collapse rate is high). And the material package is damaged when serious. 3. The scheduling strategy is stiff, stacking and storage links are split, an AGV logistics system is not linked, the equipment utilization rate is low, the storage space utilization rate is low due to the lack of a dynamic scheduling model in the traditional scheme, and the material distribution delay rate is high. Patent document CN220130470U discloses a dried tea pile up neatly device for dried tea production and processing, its workflow is carried the dried tea through the conveyer belt, in-process dried tea is through holding the board smooth slip into weighing platform top bin with the help of the deflector, after the dry quantity of dried tea reaches weighing platform and sets up weight, push away the bin to pile up neatly platform department by first push pedal, later push away the inside bin of placing to weighing platform top by the second push pedal, continue the vanning processing, when pile up neatly platform department bin quantity reaches three, push it to pile up neatly platform other end. However, the above technical problems cannot be solved in this patent document. The root cause is that the existing system lacksThe mechanical structure does not realize self-adaptive adjustment and does not construct codesIs difficult to adapt to the multi-specification and nonstandard characteristics of flexible bulk materials such as tea. Disclosure of Invention Aiming at the defects in the prior art, the invention aims to provide an intelligent scheduling system and method of an intelligent palletizing robot. The invention provides an intelligent scheduling system of an intelligent palletizing robot, which comprises the following components: the three-dimensional intelligent identification module is used for extracting size data of the materials from the collected material point cloud data; the material density estimation module is used for calculating the material density according to the size data and the weight information of the materials; the pneumatic clamping jaw travel adaptation module calculates travel parameters of the pneumatic clamping jaw according to size data of materials and material density and superimposes a safety margin; The self-adaptive stacking scheme generation module is used for generating an optimal stacking scheme through an intelligent planning algorithm based on size data and safety margin of materials; The WMS storage scheduling module screens out candidate storage bits with adaptive size from idle storage bits according to size data and safety margin of materials, matches optimal storage bits through a double-layer optimization algorithm and issues storage bit allocation instructions; And the IOT communication platform is used for completing the interactive control of the WMS software system and the equipment automation control system. Preferably, the process of extracting the size data of the material by the three-dimensional intelligent identification module comprises the following steps: filtering abnormal values of the original point cloud data, and removing noise points which deviate from the mean value by more than 3 times of standard deviation by adopting a Z-score algorithm; Projecting the preprocessed point cloud on an XY plane, constructing a convex hull model to fit the outer contour of the material, determining the length of the material by calculating the maximum distance between the convex hull vertexes, and calculating the width by the projection difference value perpendicular to