China has become the largest consumer of industrial robots in the world, but the gap of related technical talents is expanding year by year. The important reason is the high cost and dangerous operation of industrial robots, which leads to difficulties in teaching and programming of robots. To solve this problem, based on HTC Vive (a VR device jointly developed by HTC and VALVE), a virtual industrial robot teaching programming system is developed by combining virtual reality (VR) technology with robot teaching programming. First, the VR environment characteristics of the system is analyzed, and a three-dimensional UI (User Interface) designed for the main interface interaction of the system. Then the kinematic modeling of the robot is carried out to realize the teaching method of dragging the robot directly. Using lexical analysis, syntax analysis and sentence execution mechanism, the virtual interpreter is developed, and the program editing mode of inserting instructions quickly is realized. Finally, the system realizes the robot instruction programming function which can carry out complex trajectory motion. Experimental tests show that the system has good interactivity and can be used for virtual robot teaching and programming. The system is safe to operate, and can reduce the cost of robot teaching and programming, contribute to the training of robot related talents, provide solutions for robot simulation in industrial production, and help the development of China’s manufacturing industry.

Coronary heart disease (CHD) , as a complex multifactorial condition, presents significant challenges for patient education due to the abstract and specialized nature of its pathological mechanisms. Conventional health education models relying primarily on static media such as pamphlets and instructional videos exhibit notable limitations in knowledge dissemination efficacy and patient engagement. To address these issues, this study innovatively develops an augmented reality (AR) -based interactive CHD health education system. The system integrates multimodal hardware including HoloLens2, Kinect DK, and Pico4 to establish a HoloLens2-Kinect DK-Pico4 multi-terminal collaborative framework, enabling shared physician-patient AR perspectives. Clinicians overlay real-time cardiovascular models onto patients’ anatomies to simulate disease progression, while patients synchronously observe personalized pathological evolution through headsets, thereby enhancing situational cognition. At the technical implementation level: Personalized educational resources undergo parameterization to optimize multimedia visual presentation; A novel depth-camera registration maintenance algorithm utilizing spatial anchors is proposed, coupled with pose update frequency optimization to significantly enhance cardiovascular model registration stability; Real-time skeletal data captured by Kinect DK is mapped to Unity’s Avatar module, ensuring stable synchronization of cardiovascular models with patient movement; Cross-platform applications (PC server, clinician AR terminal, patient VR terminal) are developed via Unity3D, with seamless HoloLens2-Kinect DK-Pico4 communication achieved through custom data packet structures, message types, and process-pool mechanisms.

To address the problem of severe lack of defect data in workpieces to train the deep-learning-based defect visual detection systems, this paper introduces a generative sample synthesis method that integrates generative adversarial networks (GAN) with a physical-based rendering (PBR) pipeline for data augmentation. The method employs ConSinGAN as the defect feature generation model and enhances the discriminator by incorporating a coordinate attention (CA) mechanism, enabling more precise identification of defect features in images. Additionally, the loss function is adjusted by introducing a weighted combination of reconstruction loss and multi-scale structural similarity loss to alleviate the gradient vanishing in small sample training and improve the quality of generated samples. The PBR pipeline is used to output the augmented samples, which first constructs a 3D model for the workpiece to be augmented, and then use poisson blending to merge the generated defect features with the original model texture. Finally, defect samples of the workpiece are rendered in a simulated production environment using a virtual camera. Experimental results on public datasets demonstrate the effectiveness of the proposed method in augmenting small samples of workpiece defects.