采煤机是综采工作面的核心装备，研发智能采煤机器人是实现综采工作面智能化的关键。综合分析当前采煤机机器人化研究进程中的传感检测、位姿控制、速度控制、截割轨迹规划与跟踪控制等技术的研究现状，提出研发智能采煤机器人必须破解的“智能感知、位姿控制、速度控制、截割轨迹规划与跟踪控制、位−姿−速协同控制”五大关键技术，并给出解决方案。针对智能感知问题，提出了构建智能感知系统思路，给出了智能采煤机器人智能感知系统的架构，实现对运行状态、位姿、环境等全面感知，为智能采煤机器人安全、可靠运行提供保障；针对位姿控制问题，提出了智能PID位姿控制思路，给出了改进遗传算法的PID位姿控制方法，实现了智能采煤机器人位姿精准控制；针对速度控制问题，提出了融合“力−电”异构数据的截割载荷测量思路，给出了基于神经网络算法的截割载荷测量方法，实现了截割载荷的精准测量；提出牵引与截割速度自适应控制思路，给出了人工智能算法牵引与截割速度决策方法和滑模自抗扰控制的牵引与截割速度控制方法，实现了智能采煤机器人速度精准自适应控制；针对截割轨迹规划与跟踪控制问题，提出了截割轨迹精准规划思路，给出了融合地质数据和历史截割数据的截割轨迹规划模型，实现了截割轨迹的精准规划；提出了截割轨迹精准跟踪控制思路，给出了智能插补算法的截割轨迹跟踪控制方法，实现了智能采煤机器人截割轨迹高精度规划与精准跟踪控制；针对“位−姿−速”协同控制问题，提出了“位−姿−速”协同控制参数智能优化思路，给出了基于多系统互约束的改进粒子群“位−姿−速”协同控制参数优化方法，实现了智能采煤机器人智能高效作业。深入研究五大关键技术破解思路，有利于加快推动研发高性能、高效率、高可靠的智能采煤机器人。

Coal mining machine is the core equipment of completely automated working face, and the research and development of intelligent coal mining robot is crucial for achieving the intellectualization of fully mechanized working face. This paper comprehensively analyzes the current research status of sensing detection, position and attitude control, speed control, cutting trajectory planning, and tracking control in the coal mining machine roboticization process, and proposes five key technologies that must be solved in the development of intelligent shearer robots, including sensing and detection, pose control, velocity control, cutting trajectory planning and tracking control. Aiming at the problem of intelligent perception, this paper proposes the construction thought of a coal mining robot intelligent perception system, as well as the architecture of a coal mining robot intelligent per-ception system. The architecture of the intelligent perception system for coal mining robots is outlined, enabling a comprehensive sensing of running state, posture, environment, and so on, thereby ensuring the safe and reliable operation of intelligent coal mining robots. In terms of the position and attitude control problem of intelligent coal mining robots, the intelligent PID position and attitude control thought is proposed, along with an improved genetic algorithm-based PID pose control method, enabling precise pose control for the coal mining robot. As to the problem of velocity control, the thought of cutting load measurement based on the fusion of “force-electricity” heterogeneous data is proposed. Additionally, a neural network-based algorithm for cutting load measurement is presented, achieving an accurate load measurement. Furthermore, a traction and cutting speed adaptive control approach is proposed, including an artificial intelligence-based decision-making method for traction and cutting speed and a sliding mode control method for traction and cutting speed with disturbance rejection. This approach enables a precise and adaptive speed control for the coal mining robot. Regarding the problem of cutting trajectory planning and tracking control, the precise cutting trajectory planning thought is proposed, incorporating geological data and historical cutting data into a cutting trajectory planning model. The precise cutting trajectory tracking control thought is proposed, and an intelligent interpolation algorithm-based cutting trajectory tracking control method is given, achieving a high-precision trajectory planning and accurate tracking control for the coal mining robot. Considering the “position-attitude-velocity” collaborative control problem, the intelligent optimization idea of "position-attitude-velocity" collaborative control parameters is proposed, which utilizes an improved particle swarm optimization method based on multi-system constraints to optimize the coordinated control parameters, resulting in intelligent and efficient operation of the coal mining robot. The in-depth investigation of these five key technologies for intelligent coal mining robot provides some valuable insights for accelerating the development of high-performance, efficient, and reliable intelligent coal mining robot.