In the case of sudden power failure, emergency supplies must be provided in a timely manner to minimize the loss of life and property caused by power failure and restore the living and working environment. Hou et al. [2] proposed a resource satisfaction transportation model for electrical emergency rescue based on node integral weights, and created an objective function with minimal material supply and time reduction. Gao et al. [14] created a multi-point fault repair optimization model considering the distribution of first-aid materials, the cooperation of maintenance teams and the repair sequence of damaged equipment.
However, traditional electrical material rescue exists mainly in urban areas, making it more practical in favorable traffic and road conditions. The disadvantages of traditional methods are more serious in areas with poor road conditions, making drone transportation a more effective means to improve transportation efficiency. The following are articles related to drone transportation.
Cheng et al. [15] further extended Dorling's research by modeling UAV power consumption as a nonlinear function of payload and trip time in the multi-trip UAV routing problem model, taking into account the time window. Their model provides logical and sub-gradient cuts to deal with nonlinear power functions, using branch and cut methods to solve the drone wiring problem. In the optimization problem of Gentili et al. [6], the value of emergency medical supplies is minimized based on their perishability. Given the battery life of drones when transporting medical supplies, they assumed that only one drone per platform could service one node at a time. The application of VRPD to heterogeneous UAVs has also been investigated in several studies. The application of VRPD to heterogeneous UAVs has also been investigated in several studies. For the routing problem of heterogeneous fixed fleet UAVs, Chowdhury et al. [16] proposed a mixed integer linear programming model to solve the cost of post-disaster inspection, because there are several aspects. Chen et al. [17] dealt with the UAV combat trajectory planning problem of UAVs with different capabilities in a multi-area system. They initially created an algorithm for grouping regions into clusters, taking inspiration from density-based clustering techniques. They then obtained an approximation of the ideal point-to-point path for the drone to perform the coverage task. In another work by Chen et al. [18], they focused on the problem of coverage combat trajectory planning for heterogeneous UAVs. They propose an ant colony system-based (ACS) approach to get enough drone paths and thoroughly and efficiently cover each area. According to the literature, the biggest obstacle in UAV distribution research is constraint, which is one of the reasons why the traditional vehicle routing problem model is not suitable for UAV logistics.
Most previous studies have examined this problem from the UAV-specific perspective, ignoring the coordination relationship between supply timeliness and UAV scheduling, as well as the target requirements and UAV safety considerations. Building drone scheduling models that focus on delivering electrical materials is critical, as models for emergency transport of powered drones are still uncommon.

Heuristic algorithms are more commonly used in UAV task assignment [19]. Based on this, Wu et al. [20] carried out a quadratic selection operation on the genetic algorithm (GA) to improve the population diversity, and adopted an improved simulated annealing algorithm (SA) to solve the cooperative multi-task assignment problem more effectively. Han et al. [21] proposed a fuzzy elite strategy genetic algorithm to deal with complex problems. In order to improve its ability to escape the local optimal trap and accelerate convergence, Wang et al. [22] propose to combine simulated annealing (SA) and large neighborhood search (LNS) algorithms. Liu et al. [23] also proposed a collaborative optimization method combining GA and clustering methods to meet the task assignment of UAVs on forest fires. Li et al. [24] used upgraded cellular automata (CA) and optimal generative tree technology to build a path network and find the best route between each endpoint. Zhang et al. [25] dynamically partitioned the particle swarm according to the particle mass and changed the topology of the algorithm. In addition, dynamic problems in task processing problems are also common. In addition, the PSO algorithm has been used in other combinations as a heuristic method: in their research, Geng et al. [26] propose a quantified particle swarm optimization algorithm for the task assignment problem of UAV clusters; Shao et al. [27] proposed a hybrid strategy based on discrete particle swarm optimization in the case of constructing quantitative particle swarm optimization technology for many-to-one task planning problem. Chen et al. [28] enhanced the simulated annealing algorithm using Levi's distribution strategy, which is suitable for both dynamic and static task assignment models. Yang et al. [29] adopted a distributed rational clustering algorithm for UAV clusters based on sensor networks and mobile information to improve the completion rate of UAV task assignment and provide a new distributed algorithm for task assignment.