To address the limitations of traditional constant-power operation mode in electric sweepers and overcome the excessive subjectivity and lack of theoretical support in fuzzy control, an energy-saving operation strategy integrating road cleaning perception and fuzzy optimization was proposed. First, the YOLOv8-seg model was used to segment and quantify road surface garbage, calculating the road cleanliness index based on garbage type and coverage area. Then, a fuzzy controller was designed with the road cleanliness index and vehicle speed as inputs and the operating motor speed and torque as outputs.Eighteen membership function parameters were selected, and the grey wolf optimizer(GWO) was employed to reduce operational energy consumption. Finally, the effectiveness of the proposed strategy was verified through MATLAB/Simulink under the established simulation conditions. Experimental results demonstrate that the proposed strategy effectively reduces power consumption, with energy usage decreasing by 9.85% compared to the unoptimized method. The optimized battery state of charge exhibits a smoother variation trend.The findings provide a new technical pathway for the intelligent and energy-efficient development of electric sweepers.