Global Path Planning in Robotics Based on Chaos-Enhanced Multi-Strategy Wild Goose Algorithm

In response to the limitations of the emerging goose optimization algorithm (GOOSE) in global search capability and its tendency to fall into local optima, we propose an improved GOOSE algorithm (IGOOSE). IGOOSE enhances population diversity by integrating chaos mapping and elite reverse learning strategies, thereby improving its global search ability. We employ a nonlinear sine Alpha control function to prevent premature convergence, and introduce a chaotic random Levy flight strategy to enhance the search efficiency in high-dimensional spaces. To balance local and global search capabilities, IGOOSE combines an improved Cauchy inverse cumulative distribution function with a golden sine strategy, which accelerates the convergence speed and improves local exploitation. Additionally, we adopt the optimal explosion particle strategy to effectively avoid local optima, and a redundancy elimination strategy further optimizes path planning. Simulation results demonstrate that the IGOOSE algorithm exhibits significant advantages in both 2D plane and 3D flight path planning: compared to GOOSE, the shortest path is shortened by 6.31%, the average path length is reduced by 7.53%, and stability is improved by 34.72%, validating its practical applicability in complex environments.