机械

针对现有智能物流搬运小车在干扰条件下视觉识别精度和实时性不足以及路径规划偏差大等问题，设计了一款基于STM32和OpenMV的智能物流搬运小车。小车以STM32F103RCT6为核心控制器，结合OpenMV视觉识别模块、二维码扫描模块、升降式机械臂模块和麦克纳姆轮底盘，实现了从物料识别、抓取到精准放置的全自动化流程。系统通过在LAB颜色空间中进行阈值分割和Blob连通域分析，并运用质心定位算法和位移计算模型，显著提高了对物料、色环的颜色识别及定位的精度与速度。此外，小车采用分层式供电架构和UART232通讯协议，有效解决了多模块间的时序同步与数据交互问题，确保了系统的高可靠性和实时响应能力。简化实验结果表明，该智能物流搬运小车在视觉识别精度和路径规划能力等方面具有显著提升，为复杂环境下的智能物流搬运提供了技术支持。

In this paper, an intelligent logistics handling trolley based on STM32 and OpenMV is designed, aiming at solving the problems of the existing intelligent logistics handling trolley, including insufficient visual recognition accuracy and real-time performance under disturbing conditions, as well as significant deviation in path planning. The trolley takes STM32F103RCT6 as the core controller, and combines OpenMV visual recognition module, QR code scanning module, lifting robotic arm module and Mc Namee wheel chassis to realize a fully automated process from material recognition, gripping to accurate placement. The system significantly improves the accuracy and speed of colour recognition and positioning of materials and colour rings by performing threshold segmentation and Blob connectivity domain analysis in LAB colour space and applying the centre-of-mass positioning algorithm and displacement calculation model. In addition, the trolley adopts layered power supply architecture and UART232 communication protocol, which effectively solves the problem of timing synchronization and data interaction among multiple modules and ensures the high reliability and real-time response capability of the system. Simplified experimental results show that the intelligent logistics handling trolley has significant improvement in visual recognition accuracy and path planning ability, which provides technical support for intelligent logistics handling in complex environments.