面向柔性精密装配任务的工业机器人作业机理与技能优化研究

随着制造产业智能化、柔性化的发展，如何提高装配机器人工作站的定位与作业精度、能力和成功率，使其既能达到装配工艺的精度要求，又能兼顾产品多型共线与快速迭代等柔性生产需求，是当前迫切需要解决的问题。本项目面向电子与电气产品制造业复杂零部件高精度装配作业需求，重点研究基于视觉和激光传感的多维视觉测量和补偿方法，实现对精密装配过程中的实时测量补偿、微小目标在线三维测量；研究机器人柔性装配状态获取与装配过程描述方法，揭示多模信息描述的装配过程相互作用机理；探索具有自主决策与学习能力的模型与算法，并用于零部件精密装配过程中，实现工业机器人柔性装配技能获得与优化；构建通用性仿真与真实平台，对提出的算法与模型进行验证，为产业化应用提供基础理论与技术支撑。

With the development of intelligent and flexible manufacturing industry，how to improve the positioning accuracy, ability and success rate of assembly robot is an urgent problem to be solved. It can not only meet the accuracy requirements of the assembly process, but also take into account the flexible production requirements such as multi product collinear and fast iteration. This project is aimed at the high precise assembly of complex parts in the electronic and electrical products manufacturing industry. The multi-dimensional vision measurement and compensation methods with vision and laser sensor are studied to realize real-time measurement compensation and online three-dimensional measurement of micro targets in the process of precision assembly. The method of robot flexible assembly state acquisition and assembly process description is studied. The interaction mechanism of multi-mode information description is revealed. The model and algorithm with autonomous decision-making and learning ability are explored to achieve the acquisition and optimization of flexible assembly skills of industrial robots. Finally, a general simulation and real platform is built to verify the proposed algorithm and model. This project provides basic theory and technical support for industrial application.