Research on NLOS error suppression in UWB based on RICT algorithm

• Addresses the sensitivity of NLOS errors in UWB indoor positioning systems using the Chan algorithm and proposes an enhanced approach known as the Residual Improved Chan-Taylor (RICT) algorithm, which aims to suppress the impact of NLOS errors on positioning accuracy. • The excellent positioning solution capability of the RICT algorithm proposed in this paper has been verified through simulation experiments and measured data. • Simulation results demonstrate that compared to the Chan algorithm, RICT significantly reduces positioning errors across different scenarios, with particularly notable improvements when the number of base stations is limited. Real-world tests show that RICT reduces positioning errors by 62.41 %, achieving an average error of approximately 10.3 cm, and ensuring that over 90 % of positioning points have errors less than 20 cm. • The RICT algorithm presents an effective solution to mitigate NLOS errors in UWB indoor positioning, offering low algorithmic complexity and wide applicability. Ultra-Wideband (UWB) technology has attracted considerable attention due to its high precision and strong multipath resistance. However, Non-Line-Of-Sight (NLOS) errors remain a significant obstacle to improving accuracy. This paper proposes an improved algorithm based on the sensitivity of Chan algorithm to NLOS errors in UWB indoor positioning, which named as the Residual Improved Chan-Taylor (RICT) algorithm. The RICT algorithm aims to suppress the impact of NLOS errors on positioning accuracy. The RICT algorithm operates through several key steps: initially obtaining a preliminary position estimate using the Chan algorithm, calculating distances between this estimate and base stations, computing residuals between these distances and observed pseudo-ranges, filtering out observations with big residuals, it can be considered that the filtered data is less affected by NLOS, and finally using an improved Chan-Taylor algorithm to determine the final position based on selected data. The effectiveness of the RICT algorithm is validated through simulations and real-world measurements. Simulation results demonstrate that compared to the Chan algorithm, RICT significantly reduces positioning errors across different scenarios, with particularly notable improvements when the number of base stations is limited, which is very meaningful in practical engineering. Real-world tests show that RICT reduces positioning errors by 62.41 %, achieving an average error of approximately 10.3 cm, the standard deviation has decreased by approximately 29.79 %, and ensuring that over 90 % of positioning points have errors less than 20 cm. The RICT algorithm provides an effective solution to the NLOS error problem in UWB indoor positioning, without the need for pre-processing of measurement data, with low algorithm complexity and wide applicability.