AI-Powered Personalized Path Planning Using User Behavioural Patterns and Semantic Context for Smart Navigation Systems

Smart navigation systems are advancing towards personalized and context-aware solutions that adjust to dynamic environments and individual user requirements. Traditional path planning algorithms, such as A*, demonstrate effectiveness on static maps; however, they fail to incorporate behavioral patterns or semantic layers, including traffic, road conditions, or user preferences. This paper presents a novel Semantic-Aware Behavioral Routing Framework (SBRF) that improves path planning through the integration of adaptive, modular AI components. The framework’s foundation is the Adaptive Context-Aware A* (ACA*) algorithm, which employs a personalized cost grid based on user behavior and semantic overlays to facilitate real-time navigation decisions. The proposed framework comprises seven implementation phases, which include modeling user mobility patterns, integrating semantic data layers, and assessing routing outcomes across various grid environments. Experimental validation on 20 × 20, 30 × 30, and 50 × 50 grid maps indicated that ACA* achieved consistent path cost reductions of 18.18% across grid-based experiments (20 × 20, 30 × 30, 50 × 50), with up to 19.6% improvement observed in real-world network validation. Planning time showed mixed results: 33.33% improvement on 20 × 20 grids, but 27–30% degradation on larger grids due to semantic processing overhead. The algorithm generates routes that better align with user preferences and enhance safety, despite occasional increases in complexity due to context-based detours. Statistical validation across 30 trials per configuration confirms significant improvements (p < 0.001) with ablation studies demonstrating the individual contribution of behavioral and semantic components.