Objective To enhance the operational efficiency and stability of Delta robots in food automation sorting systems.Methods Based on an analysis of food automation sorting systems, a Delta robot sorting strategy combining machine vision, a multi-objective model, and an improved dung beetle optimizer algorithm was proposed. Machine vision was used to obtain real-time food position data, and a sorting optimization model was established with the goal of achieving the optimal balance between running path efficiency and stability. The model was solved using an improved mantis optimization algorithm, and the performance of the proposed method was experimentally verified.Results Compared with conventional methods, the proposed method achieved a faster average sorting speed and lower end vibration impact, with an average sorting speed of <0.60 pieces/s and an average end acceleration of <16 m/s².Conclusion The integration of machine vision, a multi-objective model, and intelligent algorithms can effectively improve the sorting efficiency and stability of Delta robots.