The impact of external conditions on quadruped robot behavior: modeling using fuzzy logic and machine learning

This article presents research on modeling the behavior of quadruped robots, specifically Spot from Boston Dynamics, under various external conditions using fuzzy logic and machine learning. A two-stage fuzzy inference model was developed to analyze the impact of environmental factors (temperature, precipitation, visibility) and operational factors (route duration, population density) on navigation disruptions and the probability of failure. Advanced machine learning algorithms, such as Gaussian Process Regression (GPR) and neural networks, were used for precise prediction of navigation failures. The proposed model, developed using MATLAB tools, represents a significant advancement in improving navigation precision and operational efficiency of assistive robots in complex environments. These findings hold promise for applications aiding individuals with special needs.