依据我国煤矿智能化发展战略，深入分析了国内外智能掘进研究现状，结合我国煤炭赋存条件复杂，巷道掘进问题突出，智能掘进挑战严峻等实际，提出了直接影响和制约我国煤矿巷道智能掘进加快发展的智能截割、智能导航、智能协同控制和远程智能测控四大关键共性技术并给出了解决思路和方法。针对掘进系统智能截割问题，提出了基于视觉伺服的掘进系统智能定形截割控制方法和基于遗传算法优化的BP(GA-BP)神经网络的自适应截割控制方法，旨在提高巷道截割成形质量和效率;针对掘进系统智能导航问题，提出了基于惯导与视觉信息融合的履带式掘进系统智能导航控制方法和基于惯导、数字全站仪与油缸行程信息融合的液压推移式掘进系统智能导航控制方法，旨在提高掘进定位定向精度，实现智能导航;针对掘进系统中掘进、支护、钻锚、运输等多系统协同控制和多任务并行控制问题，提出了基于强化学习的并行作业控制方法和基于Agent的并行控制方法，以及leader-follower法和基于行为法的智能协同控制方法，旨在实现多机器人系统或智能设备的智能协同控制和并行作业，提高掘进效率;针对掘进系统智能测控问题，创建了本地控制层、近程集控层和远程监控层的智能测控系统架构，提出了数字孪生驱动的虚拟远程智能控制方法，旨在保证掘进系统安全、可靠、高效运行，实现身临其境的虚拟远程智能测控。破解煤矿巷道智能掘进的四大关键共性技术难题。

According to the national strategy of “intelligent development of coal mines”,the current research status of roadway intelligent heading in the world is analyzed.Combining with the complex conditions of coal mine prominent problems of roadway heading and severe challenges with intelligent heading,four key common technologies,such as intelligent cutting,intelligent navigation,intelligent collaborative control and remote intelligent measurement and control that directly affect and restrict the accelerated development of intelligent heading of coal mine are proposed,and the solutions are also given.Aiming at the problem of intelligent cutting in the heading system,the intelligent shaping cutting control method based on visual servo and the adaptive cutting control method of BP neural network optimized by genetic algorithm are proposed to improve the quality and efficiency of cutting and forming.Aiming at the problem of intelligent navigation of the heading system,the intelligent navigation control method of the crawler heading system based on the fusion of inertial navigation and visual information and the intelligent navigation control method of the hydraulic traveling heading system based on the fusion of inertial navigation,digital total station and cylinder stroke information are proposed to improve the accuracy of positioning and orientation testing and realize intelligent navigation.Aiming at the problems of multi-system coordinated control and multi-task parallel control in the heading system,such as heading,support,drilling anchor and transportation,the parallel operation control methods based on reinforcement learning and agent and the intelligent collaborative control methods based on leader-follower and behavior are put forward,in order to realize the intelligent collaborative control and parallel operation of multirobot systems or smart devices and improve heading efficiency.Aiming at the problem of intelligent measurement and control of the heading system,the intelligent measurement and control system architecture of the local control layer,the short-range centralized control layer and the remote monitoring layer are constructed,and a virtual remote intelligent control method driven by a digital twin are proposed,in order to ensure the safe,reliable and efficient operation of the heading system and realize immersive virtual remote intelligent measurement and control.Cracking the four key common technical problems of intelligent heading in coal mines.