On the performance comparison of multi-objective evolutionary UAV path planners

The big number of evolutionary planners for Unmanned Aerial Vehicles (UAVs) that have been developed demonstrates the good acceptance that the evolutionary techniques enjoy within the UAV community. However, the minor or nonexistent statistical characterization of the results obtained by the majority of the planners not only makes it difficult to assess their actual performance but also to justify the selection and/or parameterization of their supporting algorithms. To fill the gap, this paper proposes a method for comparing the planners performance by jointly employing several general and problem-specific quality indexes, which take into account the complexity and particularities of the problem. The generality of the performance metrics adopted, which are able to deal with any multi-objective dominance definition, makes them equally applicable to multi-objective planners with different relation operations (such as Pareto dominance, weighted objectives aggregation, and others). The specificity of the other indexes, which consider the types of solutions preferred by the problem experts, makes them especially attractive to characterize their planners’ behavior. The paper also shows how to analyze the results of the quality indexes graphically in order to identify, for a particular UAV planning problem, the best planners within a set of 36 variants (based on Genetic Algorithms, Particle Swarm Optimization and Differential Evolution).