Currently, multi-robot cooperative algorithms are widely used in the field of agriculture, which greatly improves the efficiency of agricultural production. However, the multi-robot cooperative operation of agricultural machinery is mostly limited to the efficiency and accuracy of scheduling. To address the mentioned shortcomings, a novel multi-task scheduling method based on improved particle swarm optimization algorithm is proposed, which is applied to the efficient collaborative scheduling problem of tomato picking robots and transfer vehicles in greenhouse cultivation of different scales. Firstly, the scene of collaborative scheduling between tomato automatic picking and transshipment is described, and the mathematical model of multi-machine collaborative scheduling is established with the shortest waiting time of picking robots and the minimum number of transshipment vehicles as the optimization objectives. Secondly, an improved particle swarm optimization algorithm is expounded in detail, which customizes the fitness function and enhances the particle update strategy. Finally, the experimental results show that the improved particle swarm optimization algorithm can not only determine the optimal number and execution order of cooperative robots, but also reduce the task execution time by 47% compared with the unimproved method.