Research on multi-objective path planning for UAVs in complex environments based on improved adaptive genetic algorithm
With the popularization of unmanned aerial vehicles (UAVs) in scenarios such as military reconnaissance, logistics transportation, and post-disaster rescue, Generating optimal flight paths that guarantee both safety and timeliness amidst high-density barriers and unknown environmental factors presents a formidable challenge in autonomous navigation. Standard genetic algorithms are often hindered by premature convergence, suboptimal local solutions, and diminished genetic diversity in later iterations. In response to these challenges, we propose an Improved Adaptive Genetic Algorithm (IAGA) that synergizes an adaptive mutation strategy with an elite retention framework.This method first constructs a comprehensive cost model combining path length and obstacle threat potential field based on the grid concept, thereby converting the nonlinear path planning problem into a computable mathematical form; secondly, it designs a linearly decreasing adaptive mutation operator that can dynamically adjust the mutation probability according to the population evolution stage, thus maintaining a strong global search ability in the early stages of the algorithm and strengthening local fine search in the later stages; finally, it introduces an elite retention strategy to ensure that excellent individuals are retained and participate in subsequent evolution, avoiding population degradation. Simulation experiments conducted in a Python environment show that in a complex obstacle environment of 100m × 100m× 100m, the IAGA algorithm can effectively plan the optimal path, and the convergence speed is significantly improved compared to traditional genetic algorithms, verifying the effectiveness and robustness of this method in handling multi-constraint path planning problems.