Manipulators are widely used in manufacturing to replace manual work, reduce production risk, and improve production efficiency. However, the high-quality demand of industrial production puts forward specific requirements for the operation accuracy of manipulators. Considering the application scenario of ingot grinding in non-ferrous metal factories, the operation tracks of the task are highly repetitive. In addition, the practical application environment is complex and there are many uncertainties such as environmental interferences and system parameter changes, which result in that fixed control parameters are difficult to ensure the continuous and stable operation of the system. Therefore, an adaptive sliding mode iterative learning tracking control method is designed in this paper. The controller includes a PD term with parameter self-tuning, the symbolic function term based on sliding mode, and a control input of the last iteration. Especially, the PD control parameters are adjusted by the fuzzy logic system to improve the robustness of the system. Furthermore, the closed-loop stability is proven by using Lyapunov techniques. Finally, the effectiveness and robustness of the proposed control method are verified through several numerical simulations.