Industrial Self-Guided Vehicles (SGVs), a type of Autonomous Mobile Robots (AMRs) used for material handling tasks, have recently attracted increased attention. As logistics scenarios grow more complex, effectively using state-of-the-art navigation algorithms becomes more challenging but essential for enhancing industrial operational efficiency. In recent years, several path and motion planning algorithms have been developed for broad autonomous navigation applications. While classical navigation schemes might seem suitable for automating SGV movement, a thorough evaluation reveals potential performance bottlenecks, especially when multiple SGVs navigate in dynamically evolving manufacturing environments. Deploying a fleet of SGVs in a shared environment complicates navigation safety for different occupants. Therefore, this paper proposes an adaptive architecture suitable for heterogeneous types of SGVs, capable of characterizing and modeling navigation safety. An extensive evaluation of the proposed approach for the SGV-System, aimed at modeling navigation conflicts through the configuration space, is conducted. The assessment highlights safety and energy efficiency benchmarks across various scenarios, including avoiding unpredictable obstacles such as pedestrians and forklifts. Additionally, simulation tests with various local motion planners were conducted to test the reliability of the proposed strategy. The results suggest that adding an adaptive component to the navigation framework, alongside suitable motion planners, enhances the motion performance of the entire fleet of autonomous mobile platforms.