In response to the current situation of low delivery efficiency, high costs and numerous overdue orders in the real-time logistics system, a rider-drone collaborative delivery model is proposed. Taking into account factors such as the coordinates of the store, the coordinates of the drone station, the coordinates of the customers and their demand, the maximum single order capacity of the rider, and the maximum flight distance of the drone, a routing model for the collaborative delivery by riders and drones is constructed with the goal of minimizing the total delivery cost. A hybrid genetic algorithm is developed to solve the model. A series of multi-type cases were used for the experiments. The results show that the method proposed in this paper can solve the distribution path planning scheme for riders and drones that meets the decision-making goals within a relatively short time. The proposed approaches can effectively reduce the total distribution cost, decrease the number of overdue orders, reduce the overdue penalty cost, and expand the distribution scope of real-time logistics.