To plan a more energy-efficient picking path, an artificial bee colony-adaptive genetic algorithm for warehouse robots is proposed. The performance of the basic genetic algorithm is considered, which depends on the quality of the initial population, the selection, crossover, and mutation operations. Firstly, the artificial bee colony algorithm is used to initialize the initial population, which can enhance the diversity of the population. The path length, turn times, and robot running energy consumption are taken as the evaluation indexes of the fitness function. Then, on the basis of crossover and mutation operators designed by adaptive strategic adjustable trigonometric functions, it can improve the convergence speed of the algorithm. Simulation results show that the path energy consumption of the artificial bee colony-adaptive genetic algorithm proposed in this article is 5. 22% lower than that of the basic genetic algorithm in the 20×20 grid map. On the 40×40 grid map, the energy consumption of the path planned by the artificial bee colonyadaptive genetic algorithm proposed in this article is 9. 08% less than that of the basic genetic algorithm. Finally, experimental results show that the energy consumption of the artificial bee colony-adaptive genetic algorithm is reduced by 7. 64% , and the planned path is smoother, which is more suitable for the path planning of warehouse robots.