Abstract: In view of the problems of unstable image acquisition, large mechanical impact and insufficient flexible control of the robotic arm in the process of flotation tailings ash detection in the current intelligent coal preparation plant. In this paper, a six-axis manipulator compliance control optimization strategy is proposed by combining the TJS-DACC interpolation optimization algorithm and the reward function mechanism in reinforcement learning. Firstly, the DH parameter model of the six-degree-of-freedom manipulator was established, the forward and reverse kinematics analysis was completed, and the trajectory simulation and visualization of the manipulator were realized by combining the Robotics Toolbox toolkit in MATLAB. Secondly, by constructing a TJS-DACC interpolation control framework that fuses the task space and joint space of the six-axis manipulator, the interpolation factor α dynamic weight interpolation is introduced, and the optimal interpolation factor training is realized by using the reward function mechanism in reinforcement learning by using the reward function mechanism in reinforcement learning to realize the optimal interpolation factor training by using the response speed and contact flexibility of the end effector under different interpolation factor conditions, and finally the optimal value is determined to be α=0.74, which is verified by establishing a physical experiment platform. The results show that the optimized robotic arm controls the ash collection of flotation tailings more smoothly, the trajectory response of the end effector of the robotic arm is faster, the compliance is significantly enhanced, and the sampling efficiency is increased by 26.13% on average compared with the traditional algorithm. The compliant control optimization strategy has good trajectory stability and environmental adaptability, which can improve the sampling efficiency of flotation tailings while better reducing the mechanical impact of the manipulator, and provides effective technical support for the control system design of flotation tailings intelligent detection equipment.