CN-122006985-A - Intelligent spraying method, device, equipment and storage medium

CN122006985ACN 122006985 ACN122006985 ACN 122006985ACN-122006985-A

Abstract

The application discloses an intelligent spraying method, device, equipment and storage medium, which relate to the field of robot control and comprise the steps of obtaining a three-dimensional CAD model of a workpiece to be sprayed, determining target geometric features of the workpiece to be sprayed based on the three-dimensional CAD model, obtaining a target parameter sequence, obtaining a predicted coating thickness distribution map corresponding to the workpiece to be sprayed based on the target parameter sequence and the target geometric features by utilizing a target prediction model, determining target thickness deviation based on the predicted coating thickness distribution map and the target coating thickness, and generating a target spraying strategy based on the target thickness deviation, so as to control a target robot to spray the workpiece to be sprayed based on the target spraying strategy. The application realizes the durable, stable and high-precision control of the spraying quality of the complex curved surface.

Inventors

DENG GUOTAO
DENG SHUANGWEI
LI BOXUAN
QIU SHIGUANG
Qu pei
Wan Die

Assignees

中航(成都)无人机系统股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260203

Claims (10)

1. An intelligent spraying method is characterized by comprising the following steps: Acquiring a three-dimensional CAD model of a workpiece to be sprayed, and determining target geometric features of the workpiece to be sprayed based on the three-dimensional CAD model, wherein the target geometric features comprise surface curvature distribution features, edge contour features and potential shadow areas; Acquiring a target parameter sequence, and acquiring a predicted coating thickness distribution map corresponding to the workpiece to be sprayed based on the target parameter sequence and the target geometric characteristic by utilizing a target prediction model, wherein the target parameter sequence is a sequence formed by real-time spraying process parameters, and the spraying process parameters comprise a robot tool center point speed, a spray gun coating flow, an atomization pressure, a forming air pressure and a spray gun-to-workpiece surface distance; And determining a target thickness deviation based on the predicted coating thickness distribution diagram and the target coating thickness, and generating a target spraying strategy based on the target thickness deviation so as to control a target robot to spray the workpiece to be sprayed based on the target spraying strategy.
2. The intelligent spray method according to claim 1, wherein the determining the target geometric feature of the workpiece to be sprayed based on the three-dimensional CAD model comprises: determining curvature information of each point on the surface of the workpiece to be sprayed based on the three-dimensional CAD model, and determining surface curvature distribution characteristics corresponding to the workpiece to be sprayed based on the curvature information; determining edge profile characteristics of the workpiece to be sprayed based on the three-dimensional CAD model by utilizing a target edge detection algorithm, wherein the edge profile characteristics comprise physical edges, edge lines and corresponding edge types; And determining a potential shadow area of the workpiece to be sprayed based on a spray cone angle model of the target spray gun, the relative pose of the spray gun and the surface of the workpiece and the three-dimensional CAD model, wherein the potential shadow area is an area which is shielded by the workpiece to be sprayed under the preset spray gun pose and cannot directly receive the coating.
3. The intelligent spraying method according to claim 1, wherein the obtaining, by using a target prediction model, a predicted coating thickness profile corresponding to the workpiece to be sprayed based on the target parameter sequence and the target geometric feature includes: inputting the target parameter sequence and the target geometric feature into a target prediction model; The first model branch of the target prediction model is utilized to fuse all the target geometric features to obtain a target feature map, and the first model branch is a model branch constructed based on a convolutional neural network; acquiring a target dependency relationship based on the target parameter sequence by utilizing a second model branch of the target prediction model, wherein the second model branch is a model branch constructed based on a long-short-time memory network, and the target dependency relationship is a time sequence dependency relationship of the spraying process parameter on a scanning path; and fusing the target feature map and the target dependency relationship to obtain a fused feature quantity, and acquiring a predicted coating thickness distribution map corresponding to the workpiece to be sprayed based on the fused feature quantity by utilizing a full-connection layer of the target prediction model.
4. The smart spray method of claim 1, wherein the generating a target spray strategy based on the target thickness deviation comprises: Judging whether the target thickness deviation is smaller than a target deviation threshold value or not; if the target thickness deviation is smaller than the target deviation threshold, determining a preset basic spraying strategy as a target spraying strategy; and if the target thickness deviation is larger than the target deviation threshold, carrying out track normal correction and spray process parameter adjustment on the basic spray strategy based on a preset safety constraint, and determining the adjusted basic spray strategy as a target spray strategy.
5. The intelligent spray method of claim 1, wherein the generating a target spray strategy based on the target thickness deviation to control a target robot to spray the workpiece to be sprayed based on the target spray strategy comprises: determining the corresponding geometric complexity of the workpiece to be sprayed based on the target geometric features; Determining a target spraying responsibility area based on the geometric complexity, the target thickness deviation and the corresponding working space range of each target robot; Generating a target spraying strategy corresponding to the target spraying responsibility area based on the geometric complexity and the target thickness deviation; And distributing corresponding target robots for the target spraying responsibility areas based on the working space range corresponding to each target robot so as to control the target robots to spray the workpiece to be sprayed based on the target spraying strategy.
6. The intelligent spraying method according to claim 5, wherein the controlling the target robot to spray the workpiece to be sprayed based on the target spraying strategy comprises: overlapping spraying areas are arranged at the junctions of the adjacent target spraying responsibility areas; When at least one target robot corresponding to the overlapped spraying area sprays the overlapped spraying area based on the target spraying strategy, synchronizing data transmission speeds of all the target robots corresponding to the overlapped spraying area, and performing position-based weighted fusion on spraying flow instructions corresponding to the target robots so as to spray the workpiece to be sprayed based on the fused spraying flow instructions.
7. The intelligent spraying method according to claim 1, wherein after the target robot is controlled to spray the workpiece to be sprayed based on the target spraying strategy, further comprising: acquiring actual coating thickness data corresponding to the target spraying strategy; constructing a target training data set based on the actual coating thickness data and the target parameter sequence when the workpiece to be sprayed is sprayed; Updating the target prediction model based on the target training data set by using a preset increment learning mechanism to obtain an updated target prediction model, and obtaining a new predicted coating thickness distribution diagram corresponding to the workpiece to be sprayed by using the updated target prediction model when a spraying task is executed on the new workpiece to be sprayed.
8. An intelligent spraying device, characterized by comprising: The geometrical feature determining module is used for acquiring a three-dimensional CAD model of the workpiece to be sprayed and determining target geometrical features of the workpiece to be sprayed based on the three-dimensional CAD model, wherein the target geometrical features comprise surface curvature distribution features, edge contour features and potential shadow areas; The predicted thickness determining module is used for obtaining a target parameter sequence and obtaining a predicted coating thickness distribution map corresponding to the workpiece to be sprayed based on the target parameter sequence and the target geometric characteristic by utilizing a target prediction model, wherein the target parameter sequence is a sequence formed by real-time spraying process parameters, and the spraying process parameters comprise the center point speed of a robot tool, the flow rate of spray gun coating, the atomizing pressure, the forming air pressure and the distance from the spray gun to the surface of the workpiece; and the workpiece spraying module is used for determining a target thickness deviation based on the predicted coating thickness distribution diagram and the target coating thickness, and generating a target spraying strategy based on the target thickness deviation so as to control a target robot to spray the workpiece to be sprayed based on the target spraying strategy.
9. An electronic device, comprising: A memory for storing a computer program; A processor for executing the computer program to implement the smart spray method as claimed in any one of claims 1 to 7.
10. A computer readable storage medium for storing a computer program, wherein the computer program when executed by a processor implements the intelligent spraying method according to any one of claims 1 to 7.

Description

Intelligent spraying method, device, equipment and storage medium Technical Field The invention relates to the field of robot control, in particular to an intelligent spraying method, device, equipment and storage medium. Background In the automatic spraying of complex curved surface workpieces in the fields of aerospace, automobiles, high-end equipment and the like, how to ensure the uniformity of the thickness of a coating is a great challenge. Workpiece geometry (e.g., sharp edges, deep concave regions, high curvature curved surfaces) can cause "edge effects" and "shadow effects" resulting in systematic, predictable thickness deviations. In the prior art, there are some methods for parameter compensation based on workpiece geometry. These methods improve uniformity to some extent by identifying high and low risk regions and applying predefined compensation strategies. However, the static compensation system has inherent defects that firstly, the core compensation rule (such as the adjustment amount of speed and flow) is determined by manual experience or off-line experiments, and is difficult to accurately match the dynamic change actual working condition, secondly, once the system is deployed, the performance of the system is fixed, slow time-varying interference such as spray gun abrasion, paint characteristic fluctuation and the like cannot be dealt with, so that the compensation effect gradually declines, and frequent manual recalibration is needed. Therefore, an intelligent spraying system capable of self-evolution, more use and more precision is urgently needed to overcome the limitation of a static compensation system and realize durable, stable and high-precision control of the spraying quality of a complex curved surface. Disclosure of Invention In view of the above, the present invention aims to provide an intelligent spraying method, device, equipment and storage medium, which can realize persistent, stable and high-precision control of the spraying quality of complex curved surfaces. The specific scheme is as follows: In a first aspect, the application discloses an intelligent spraying method, comprising the following steps: Acquiring a three-dimensional CAD model of a workpiece to be sprayed, and determining target geometric features of the workpiece to be sprayed based on the three-dimensional CAD model, wherein the target geometric features comprise surface curvature distribution features, edge contour features and potential shadow areas; Acquiring a target parameter sequence, and acquiring a predicted coating thickness distribution map corresponding to the workpiece to be sprayed based on the target parameter sequence and the target geometric characteristic by utilizing a target prediction model, wherein the target parameter sequence is a sequence formed by real-time spraying process parameters, and the spraying process parameters comprise a robot tool center point speed, a spray gun coating flow, an atomization pressure, a forming air pressure and a spray gun-to-workpiece surface distance; And determining a target thickness deviation based on the predicted coating thickness distribution diagram and the target coating thickness, and generating a target spraying strategy based on the target thickness deviation so as to control a target robot to spray the workpiece to be sprayed based on the target spraying strategy. Optionally, the determining the target geometric feature of the workpiece to be sprayed based on the three-dimensional CAD model includes: determining curvature information of each point on the surface of the workpiece to be sprayed based on the three-dimensional CAD model, and determining surface curvature distribution characteristics corresponding to the workpiece to be sprayed based on the curvature information; determining edge profile characteristics of the workpiece to be sprayed based on the three-dimensional CAD model by utilizing a target edge detection algorithm, wherein the edge profile characteristics comprise physical edges, edge lines and corresponding edge types; And determining a potential shadow area of the workpiece to be sprayed based on a spray cone angle model of the target spray gun, the relative pose of the spray gun and the surface of the workpiece and the three-dimensional CAD model, wherein the potential shadow area is an area which is shielded by the workpiece to be sprayed under the preset spray gun pose and cannot directly receive the coating. Optionally, the obtaining, by using a target prediction model, a predicted coating thickness distribution map corresponding to the workpiece to be sprayed based on the target parameter sequence and the target geometric feature includes: inputting the target parameter sequence and the target geometric feature into a target prediction model; The first model branch of the target prediction model is utilized to fuse all the target geometric features to obtain a target feature map, and the first model branch is a m