SPAD: A Robust Algorithm to Enhance Precision in Line-Following Robots Navigation

Line-Following Robots (LFRs) have significant importance for their ability to automate transportation systems and decrease human effort. However, challenges such as complex navigation paths and dynamics environments can significantly affect their efficiency and reliability. This study introduces the Statistical Processing and Adaptive Decision (SPAD) algorithm for LFRs to increase accuracy of the decision-making of LFRs compared to conventional single-point analysis methods. The proposed algorithm processes dataset from multiple sensor points, ensuring precise trajectory identification and improved adaptability to complex tracks. Experimental results proves that the SPAD algorithm achieves an accuracy of approximately 98% in navigating intricate LFR paths, whereas conventional algorithms have an accuracy range of of 60-70%. However, machine learning-based algorithms are generally as accurate as SPAD but tend to be around 20% slower on the same hardware and more resource-hungry. These findings highlight the potential of the SPAD approach in optimizing LFR performance for real-world applications.