In order to solve the problems that the traditional flotation tailings ash detection mostly depends on the color of the surface slurry, the correlation is not high and it is susceptible to environmental interference, a new method for tailings ash detection based on robotic arm-assisted stratification is proposed and implemented. Firstly, the difference in settlement velocity between coal and gangue in the process of static and flow dumping was analyzed by using the principles of physical mechanics, so as to provide theoretical support for the stratification of coal gangue in tailings. Subsequently, a six-degree-of-freedom manipulator model was established based on the DH parameter method, and its kinematics and working space were simulated with the help of MATLAB Robotics Toolbox, which verified that the selected manipulator could reach and run smoothly in the space environment of the coal preparation plant. In order to achieve fast and efficient online monitoring, this paper further uses the TOPP-RA algorithm to plan the time-optimal trajectory of the robotic arm, which significantly shortens the operation cycle while satisfying the speed and acceleration constraints. During the experiment, the manipulator arm made the coal and gangue in the tailings obviously stratified through reasonable tilt angle and vibration action. Then, the image processing technology was used to extract the area proportion of coal and gangue after stratification, and compared it with the ash value obtained by the actual laboratory analysis. The results show that the Pearson correlation coefficient between coal gangue proportion and tailings ash can reach 0.42 under the conditions of multiple coal types and working conditions, which is significantly better than the traditional detection method based on the color of surface slurry. This study provides a feasible automation solution for the prediction of flotation tailings ash, and lays an important foundation for the intelligent construction and refined production control of subsequent coal preparation plants.