Intelligent monitoring and alarming system suitable for running action of clothing manufacturing robot

Abstract

The invention belongs to the technical field of clothing supervision, and particularly relates to an intelligent monitoring and alarming system suitable for running actions of a clothing robot, which comprises a data acquisition preprocessing unit, a neural radiation field model unit, a robot control unit with a body, an intelligent monitoring and alarming unit, an evaluation optimization analysis unit and a background terminal, the physical model and the object recognition technology are integrated into the nerve radiation field, training is performed based on the worker work video data to drive the robot to complete the clothing making task, the automatic clothing making process is achieved, the abnormal behaviors of the clothing making robot are found and corrected in time through real-time monitoring and alarming, and the clothing making efficiency is improved. The system is simple in structure and convenient to operate, guarantees smooth implementation of the garment manufacturing process, improves production efficiency and product quality, can effectively prevent safety accidents caused by robot operation faults, guarantees personal safety of operators, and is beneficial to promoting the garment manufacturing industry to develop towards the intelligentization and automation directions.

Description

Technical Field

[0001] The present invention relates to the technical field of garment manufacturing supervision, and specifically to an intelligent monitoring and alarm system for the operation actions of garment-making robots. Background Art

[0002] With the development of intelligent manufacturing technology, embodied garment-making robots have gradually become an important tool in the garment-making industry. However, the garment-making process involves complex physical interactions and fine operations, which pose extremely high requirements for the action accuracy and response speed of the robots.

[0003] Most of the existing garment-making robot systems lack an effective operation action monitoring mechanism, making it difficult to issue early warnings in a timely manner when the robot's actions are abnormal or the garment-making quality does not meet the standards, which affects the garment-making efficiency and product quality and is not conducive to promoting the development of the garment-making industry towards the direction of intelligence and automation.

[0004] In view of the above technical deficiencies, a solution is proposed herein. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent monitoring and alarm system for the operation actions of garment-making robots, which solves the problems in the prior art that it is difficult to issue early warnings in a timely manner when the robot's actions are abnormal or the garment-making quality does not meet the standards, affecting the garment-making efficiency and product quality and being not conducive to promoting the development of the garment-making industry towards the direction of intelligence and automation.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] An intelligent monitoring and alarm system for the operation actions of garment-making robots, comprising a data acquisition and preprocessing unit, a neural radiance field model unit, an embodied robot control unit, an intelligent monitoring and alarm unit, an evaluation and optimization analysis unit, and a background terminal;

[0008] The data acquisition and preprocessing unit acquires high-definition video data of the work of garment-making workers, including the garment-making process under multi-angle and multi-illumination conditions, and provides high-quality data input for model training through preprocessing steps such as video frame division, denoising, and enhancement.

[0009] The neural radiance field model unit constructs a basic NeRF model for 3D scene reconstruction. The embodied robot control unit configures a multi-joint robotic arm and a high-precision sensor, and responds to the instructions of the monitoring system through the interaction protocol with the NeRF model, and controls the garment-making robot based on the control algorithm of reinforcement learning.

[0010] During the operation of the garment-making robot, the intelligent monitoring and alarm unit monitors the operation actions of the garment-making robot in real time. When abnormal behaviors are detected, it automatically triggers the alarm mechanism and sends alarm information to the background terminal; the evaluation and optimization analysis unit judges the system operation performance based on various evaluation indicators and sends the judgment information to the background terminal to continuously optimize the system performance.

[0011] Furthermore, the neural radiance field model unit incorporates and integrates an object recognition module, a physical property prediction module, and a temporal information processing module.

[0012] Furthermore, the object recognition module uses deep learning object detection and segmentation algorithms to identify target objects in the video and outputs object bounding boxes and semantic segmentation masks; among them, the target objects include clothes and tools.

[0013] Furthermore, the physical property prediction module combines computer vision and physical simulation to predict the physical properties of clothes and outputs the physical parameters of each part of the clothes; among them, the predicted physical properties include the softness, elasticity, and thickness of the clothes.

[0014] Furthermore, the temporal information processing module uses a recurrent neural network or a temporal convolutional network to capture the dynamic information in the garment-making process and outputs a feature representation related to time series.

[0015] Furthermore, when the evaluation and optimization analysis unit conducts system operation performance analysis, the evaluation indicators it refers to include 3D reconstruction accuracy, object recognition accuracy, physical property prediction error, robot action imitation accuracy, and garment-making task completion quality.

[0016] Compared with the prior art, the beneficial effects of the present invention are:

[0017] By integrating the physical model and object recognition technology into the neural radiance field and training based on the worker's work video data to drive the embodied robot to complete the garment-making task, the automated garment-making process is realized, and through real-time monitoring and alarm to ensure the smooth progress of the garment-making process and effectively prevent safety accidents, the production efficiency and product quality are improved, and the garment-making industry is promoted to develop in the direction of intelligence and automation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the accompanying drawings;

[0019] Figure 1 is the overall system block diagram of the present invention;

[0020] Figure 2 is the system block diagram of the neural radiance field model unit in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0021] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0022] Embodiment 1: As Figure 1-2 shown, the intelligent operation action monitoring and alarm system applicable to garment-making robots proposed by the present invention includes a data acquisition and preprocessing unit, a neural radiance field model unit, an embodied robot control unit, an intelligent monitoring and alarm unit, an evaluation and optimization analysis unit, and a background terminal;

[0023] The data acquisition and preprocessing unit collects high-definition video data of garment-making workers' work, including the garment-making process under multiple angles and multiple lighting conditions, and provides high-quality data input for subsequent model training through preprocessing steps such as video frame division, denoising, and enhancement.

[0024] The neural radiance field model unit constructs a basic NeRF (neural radiance field) model for 3D scene reconstruction, integrates an object recognition module, and integrates a physical property prediction module and a temporal information processing module; among them, the object recognition module uses deep learning object detection and segmentation algorithms (such as Mask R-CNN) to identify target objects (including clothes and tools, etc.) in the video, and outputs object bounding boxes and semantic segmentation masks;

[0025] The physical property prediction module combines computer vision and physical simulation to predict the physical properties of clothes, including the softness, elasticity, and thickness of clothes, etc., and outputs the physical parameters of each part of the clothes; the temporal information processing module uses a recurrent neural network (RNN) or a temporal convolutional network (TCN) to capture the dynamic information in the garment-making process and outputs a temporal-related feature representation.

[0026] The embodied robot control unit (Embodied Robots refer to robots that have a physical form and can perform tasks and interact with the environment in the real world, and can perform complex tasks and operations, such as identifying objects, understanding instructions, and planning actions) is configured with a multi-joint robotic arm and high-precision sensors, and precisely controls the garment-making robot based on a reinforcement learning-based control algorithm; and designs an interaction protocol with the NeRF model to ensure that the robot can accurately respond to the instructions of the monitoring system.

[0027] During the operation of the garment-making robot, the intelligent monitoring and alarm unit monitors the operation actions of the garment-making robot in real time. When abnormal behaviors are detected, it automatically triggers an alarm mechanism and sends an alarm message to the background terminal, reminding the operator to take corresponding measures through sound, light, and other means;

[0028] It should be noted that the monitored abnormal behaviors include slow or stopped actions, program crashes or unresponsiveness, sensor failures, runaway phenomena, mechanical failures, and abnormal connection states. Among them, when the clothing-making robot is performing clothing-making tasks, if it shows slow or completely stopped actions, it may be due to internal failures of the robot, such as motor failures, damage to transmission components, etc. In addition, insufficient power supply or controller failures may also cause the robot to move slowly or stop.

[0029] The control system of the clothing-making robot is the key to its normal operation. If the control system experiences a program crash or unresponsiveness, the robot will be unable to continue performing clothing-making tasks. This situation may be caused by software vulnerabilities, program errors, or external electromagnetic interference, etc.

[0030] The clothing-making robot is equipped with various sensors to sense various parameters and states during the clothing-making process. If a sensor fails, such as a dirty camera lens, a malfunction of the lidar transmitter and receiver, etc., the robot will be unable to accurately obtain information during the clothing-making process.

[0031] The runaway phenomenon refers to the abnormal phenomenon of the robot accelerating out of control during operation. This may be caused by controller failures, motor driver failures, sensor signal interference, or brake system failures, etc. The runaway phenomenon will not only have a serious impact on the clothing-making process, but may also pose a safety threat to surrounding equipment and personnel.

[0032] Mechanical structure failures are also one of the common abnormal behaviors of the clothing-making robot. For example, the robot arm and joints may move inflexibly, which may be due to poor lubrication or wear of mechanical components. In addition, inaccurate robot positioning may also be caused by malfunctions or improper adjustment of the positioning system.

[0033] An abnormal connection state between the clothing-making robot and the controller or network may also cause abnormal behaviors of the robot. For example, unstable network connections or faults in the communication line between the controller and the robot may cause the robot to be unable to receive correct instructions or unable to feedback status information to the controller.

[0034] The evaluation and optimization analysis unit judges the system operation performance based on evaluation indicators such as 3D reconstruction accuracy, object recognition accuracy, physical property prediction error, robot motion imitation accuracy, and clothing-making task completion quality, and sends the judgment information to the background terminal to continuously optimize the system performance.

[0035] Working principle of the present invention: During use, by integrating physical models and object recognition technologies into a Neural Radiance Field (NeRF), training is carried out through the video data of workers' operations. Finally, it drives an embodied robot to complete the clothing manufacturing task, enabling it to recognize clothes and understand their physical properties, thereby realizing an automated clothing manufacturing process. Moreover, through real-time monitoring and alarming, abnormal behaviors of the clothing manufacturing robot can be promptly detected and corrected, ensuring the smooth progress of the clothing manufacturing process, improving production efficiency and product quality, effectively preventing safety accidents caused by robot operation failures, guaranteeing the personal safety of operators, and contributing to the development of the clothing industry towards the direction of intelligence and automation.

[0036] All the above formulas are dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation. The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not elaborate on all the details, nor do they limit the invention to only the specific implementation manners. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principle and practical application of the present invention, so that those skilled in the relevant technical field can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. An intelligent monitoring and alarm system for the operation of a garment-making robot, characterized in that: It includes data acquisition preprocessing unit, neural radiation field model unit, embodied robot control unit, intelligent monitoring and alarm unit, evaluation optimization and analysis unit and background terminal; The data acquisition and preprocessing unit collects high-definition video data of garment workers at work, including the garment-making process under multiple angles and lighting conditions, and provides high-quality data input for model training through video framing, denoising and enhancement preprocessing steps; The neural radiation field model unit constructs a basic NeRF model for 3D scene reconstruction. The embodied robot control unit is equipped with a multi-joint robotic arm and high-precision sensors. It responds to the instructions of the monitoring system through an interactive protocol with the NeRF model and controls the clothing robot based on a control algorithm based on reinforcement learning. During the operation of the garment-making robot, the intelligent monitoring and alarm unit monitors the operation of the garment-making robot in real time. When abnormal behavior is detected, the alarm mechanism is automatically triggered and the alarm information is sent to the background terminal. The evaluation and optimization analysis unit judges the system performance based on various evaluation indicators, and sends the judgment information to the background terminal to continuously optimize the system performance.

2. The intelligent monitoring and alarm system for the operation of a clothing-making robot according to claim 1 is characterized in that: The neural radiation field model unit incorporates and integrates an object recognition module, a physical property prediction module and a time series information processing module.

3. The intelligent monitoring and alarm system for the operation of a clothing-making robot according to claim 2 is characterized in that: The object recognition module uses deep learning target detection and segmentation algorithms to identify target objects in the video and output object bounding boxes and semantic segmentation masks; among them, target objects include clothes and tools.

4. The intelligent monitoring and alarm system for the operation of a clothing-making robot according to claim 2 is characterized in that: The physical property prediction module combines computer vision and physical simulation to predict the physical properties of clothing and outputs the physical parameters of each part of the clothing; among them, the predicted physical properties include the softness, elasticity and thickness of the clothing.

5. The intelligent monitoring and alarm system for the operation of a clothing-making robot according to claim 2 is characterized in that: The temporal information processing module uses a recurrent neural network or a temporal convolutional network to capture the dynamic information in the clothing making process and outputs temporal-related feature representations.

6. The intelligent monitoring and alarm system for the operation of a clothing-making robot according to claim 1 is characterized in that: When the evaluation optimization analysis unit conducts system operation performance analysis, the reference evaluation indicators include 3D reconstruction accuracy, object recognition accuracy, physical property prediction error, robot motion imitation accuracy and clothing task completion quality.

Images