A method and system for constructing a fiber laying process parameter model based on imitation learning

Abstract

The invention discloses a method and system for constructing a fiber laying process parameter model based on imitation learning. The system includes modules such as a manual laying teaching module, and the teaching operator holds a small fiber laying device to carry out fiber laying work on the surface of the mold. The motion capture module records the real-time position of the laying point in the mold, the video recording module records the operation process in real time, and the process parameter sensing module records the fiber laying process parameters in real time. The above data is transmitted to the industrial computer for data storage and processing. Through imitation learning The algorithm module completes the learning process of manual laying and teaching operations, and outputs the probability mapping model of "process parameters - surface curvature". The automatic fiber laying system based on the collaborative robotic arm uses the learned model results and combines intelligent control algorithms to realize fiber placement. Auto-layout work on curved surfaces. The invention improves the control precision of the laying path and process parameters in the laying process of the composite material, thereby reducing laying defects.

Description

technical field [0001] The invention provides a method and system for constructing a fiber laying process parameter model based on imitation learning, and relates to the fields of advanced composite material manufacturing, artificial intelligence and robot technology. Background technique [0002] Fiber-reinforced composite materials refer to composite materials composed of high-strength fibers and matrix materials. Because of their advantages such as high specific strength, fatigue resistance, corrosion resistance, and strong designability, they are widely used in aerospace, marine, automotive, sports, medical and other fields. There are a wide range of application requirements. [0003] Automatic placement of composite materials is one of the widely used high-performance and low-cost forming methods. According to different prepregs and processing methods, it is mainly divided into automatic fiber tape placement and automatic fiber filament placement. In the process of aut...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problems that the acquisition of the current fiber placement process parameters depends on a large number of experiments, and the obtained parameters have poor generalization ability and the control accuracy of the process parameters of the robotic fiber automatic placement system is not high, the present invention proposes a fiber placement method based on imitation learning. The construction method and system of the laying process parameter model, through the manual teaching and imitation learning method, constructs the probability mapping model of the fiber laying process parameters; combined with the designed intelligent control algorithm, the robust force / power of the multi-axis robotic arm in the complex laying environment is realized. Bit mixing control and multivariable decoupling control of fiber placement head to improve precise regulation of laying path and process parameters

Images