The invention discloses a warehouse logistics AGV path planning algorithm based on an ant colony algorithm and an improved genetic algorithm, relates to the ant colony algorithm and the genetic algorithm, and overcomes the defects that a traditional method is time-consuming in calculation, easy to premature and converge and easy to fall into local optimum. The invention comprises the following steps: 1, establishing gridding division and coding for a site; 2, generating an initial AGV path for genetic evolution by using an improved ant colony algorithm based on obstacle information; 3, iteratively selecting an AGV optimal path based on a three-stage genetic algorithm; 4, carrying out tail end intersection on different AGV paths with overlapping points; 5, performing favorable variation on the AGV path; and 6, recalculating the AGV path fitness, judging whether iteration is terminated or not, and carrying out trajectory smoothing processing on the terminated AGV path. The basic idea of the invention is that the improved ant colony algorithm and the improved genetic algorithm are combined, the iterative convergence speed is increased, the AGV path with higher operation efficiency is obtained, and the engineering applicability is high.