This paper presents a novel Hybrid Fuzzy-Potential Field (HFPF) navigation algorithm for mobile robots operating in cluttered environments. Unlike traditional potential field approaches that rely on static gains, the proposed method dynamically adjusts both attractive and repulsive forces using an Interval Type-2 Fuzzy Inference System (IT2-FIS). Three normalized inputs—obstacle proximity, heading error, and robot velocity—are used to adaptively tune the navigation gains in real time. The resulting navigation force is then translated into linear and angular velocities through a PID-based converter, ensuring smooth and bounded motion suitable for differential-drive platforms. To validate the effectiveness of the proposed approach, extensive simulations were conducted using MATLAB. The HFPF algorithm was benchmarked against five state-of-the-art methods—Type-1 Potential Field (PF-T1), Dynamic Window Approach (DWA), RRT*, Deep Reinforcement Learning Navigation (DRL-Nav), and Fuzzy Sliding Mode Control (Fuzzy-SMC)—across five performance metrics: success rate, path smoothness, obstacle clearance, time to goal, and path efficiency. Results demonstrate that HFPF achieves superior trade-offs between safety and efficiency, with the Balanced configuration achieving a 98% success rate and over 91% path efficiency. The Adaptive Fuzzy architecture enables real-time context-aware decision-making, making HFPF a robust and versatile solution for autonomous navigation in uncertain and dynamic environments.