In the complex storage environment, the global path planning in the process of automated guided vehicle(AGV) handling has attracted much attention, and the minimization of energy consumption is a key goal. The path planning of the traditional ant colony algorithm does not fully consider the energy consumption. To this end, the improved ant colony algorithm was proposed to realize the minimum energy consumption planning of AGV transportation path. Firstly, the principle and basis of ant colony algorithm were introduced, and then a mathematical model of the whole process of motion energy consumption of single AGV comprehensive acceleration, deceleration and turning was constructed to provide a basis for algorithm improvement. Based on this model, three improvement strategies were proposed: first, the trend function improves the pheromone update mechanism to improve the convergence speed and accuracy. The second was to introduce energy consumption factors to optimize path selection and use code models to enhance the attention of energy efficiency. The third was to reverse eliminate redundant inflection points, simplify the path, and reduce energy consumption. Finally, the simulation was compared with the original ACO experiment to verify the effectiveness of the improved algorithm. The results show that compared with the traditional algorithm, the energy consumption of the transportation path planned in the complex storage environment is lower, more concise and efficient, which is of great significance for improving the handling efficiency and reducing the energy consumption of AGV, and has significant application value.