Article Details

Abstract

Elevator floor recognition is of great significance in intelligent building management and elevator safety monitoring. Traditional methods based on encoder or limit switch are complex to install and high in maintenance cost, while methods based on single barometric pressure sensor or Inertial Measurement Unit (IMU) are susceptible to noise and environmental interference. To address these problems, this paper proposes an adaptive Kalman filter method fusing IMU and barometric pressure sensor. This method introduces a residual statistics-driven covariance self-updating mechanism in the filtering process to improve the robustness of height estimation; meanwhile, it designs an online clustering strategy to realize automatic modeling of floor height, and combines direction awareness constraint to suppress start-stop misjudgment. The measured results show that this method can achieve stable and reliable recognition in elevators in different buildings, the height estimation error is controlled within the sub-meter range, and the recognition accuracy is significantly improved compared with traditional fixed-parameter filtering and single-sensor method. The research shows that this method has good versatility and engineering application value.

Keywords

elevator floor recognition; adaptive Kalman filter; barometric pressure sensor; IMU; intelligent building

References

1. [1] Li Zhimin, Zhang Jikun Design of a new type of absolute floor recognition device for elevators [J]. China Elevator, 2023, 34 (8): 20-22
2. [2] Zhou Yaohong Research on Key Technologies for Safety Assessment and Renovation Inspection of Old Elevators [C]//Academic Seminar on Artificial Intelligence and Economic Engineering Development Proceedings of the Academic Symposium on Artificial Intelligence and Economic Engineering Development (III) [Place of publication unknown]: Academic Symposium on Artificial Intelligence and Economic Engineering Development, 2025: 1316-1319
3. [3] Sun Jia, Gui Yugui, Wu Jie, etc Measurement Error Analysis and Correction Method of Barometric Altimeter [J]. Internet of Things Technology, 2016, 6 (12): 19-21
4. [4] He Shaowei Research on Vibration Error Analysis and Real time Compensation of Micro Inertial Measurement Unit [D]. Taiyuan: North China University, 2024
5. [5] Apply, Jin Jun, Jin Yinde, etc Research on Speed Estimation Method for Elevator Safety Component Testing Based on Kalman Filter [J]. Journal of Jinhua Vocational and Technical College, 2015, 15 (6): 48-51
6. [6] Yang Yu, Wang Qianshan, Li Xuepeng, etc A non-invasive intelligent monitoring method for elevator operation status [J]. Journal of Taiyuan University of Technology, 2021, 52 (6): 920-927
7. [7] BASHIR M A，MALIK F M，AKBAR Z A，et al. Kalman Filter Based Sensor Fusion for Altitude Estimation of Aerial Vehicle [J/OL].IOP Conference Series Materials Science and Engineering，2020，853（1）：012034[2025-08-05].https://iopscience.iop.org/article/10.1088/1757-899X/853/1/012034.
8. [8] KALMAN R E. A New Approach to Linear Filtering and Prediction Problems [J].Journal of Fluids Engineering，1960，82（1）：35-45.
9. [9] AKHLAGHI S，ZHOU N，HUANG Z Y. Adaptive Adjustment of Noise Covariance in Kalman Filter for Dynamic State Estimation [J/OL].arXiv:1702.00884[2025-08-11].[cs.SY].https://doi.org/10.48550/arXiv.1702.00884.
10. [10] Chen Yongsen, Liu Jingbin, Lei Jietao, etc Indoor and outdoor elevation estimation and indoor floor recognition method using GNSS/barometer fusion [J/OL]. Journal of Wuhan University (Information Science Edition), 2024:1-13 [2026-03-03] https://doi.org/10.13203/j.whugis20230299.

About the authors

Wan Xiaoheng (September 2002-), male, Han ethnicity, from Suzhou, Jiangsu, is currently pursuing a bachelor's degree. His research direction is multi-sensor fusion positioning; Wang Huaide (October 1998-), male, Han nationality, from Lianyungang, Jiangsu, engineer, master's degree, research direction: multi-sensor fusion positioning.