Mole-Inspired Deterministic Optimization with Guided Sensing

Metaheuristic algorithms are widely used for solving complex nonlinear optimization problems; however, many rely on stochastic mechanisms that may lead to unstable search behavior and limited reproducibility. This paper introduces a novel deterministic optimization framework, termed Mole-Inspired Deterministic Optimization with Guided Sensing (MIDO-GS), which is inspired by sensory-driven navigation mechanisms observed in subterranean animals.The proposed method employs a structured sensing strategy that integrates directional probing, memory-guided search, and deterministic relocation to effectively balance exploration and exploitation. Unlike conventional population-based approaches, MIDO-GS avoids stochastic operators and instead relies on guided sensing and explicit memory structures, resulting in stable and reproducible search dynamics.The performance of the proposed algorithm is evaluated using two complementary benchmarking frameworks: the COCO/BBOB suite and the CEC2014 benchmark set. Comparative analyses against established methods, including Covariance Matrix Adaptation Evolution Strategy (CMA-ES), Differential Evolution (DE), and Particle Swarm Optimization (PSO), demonstrate that MIDO-GS achieves competitive performance. The algorithm shows particularly strong and stable behavior across structured optimization landscapes, outperforms PSO, and achieves performance comparable to DE, while remaining moderately inferior to CMA-ES in overall ranking.These results highlight the potential of deterministic, sensing-driven optimization as a robust alternative to stochastic metaheuristics, particularly for problems requiring stable and interpretable search behavior.