DOI QR코드

DOI QR Code

Development of a Fault Detection Algorithm for Multi-Autonomous Driving Perception Sensors Based on FIR Filters

FIR 필터 기반 다중 자율주행 인지 센서 결함 감지 알고리즘 개발

  • Jae-lee Kim (Dept. of Electronic engineering Korea National Univ, of Transportation) ;
  • Man-bok Park (Dept. of Electronic engineering Korea National Univ, of Transportation)
  • 김재이 (한국교통대학교 전자공학과) ;
  • 박만복 (한국교통대학교 전자공학과 )
  • Received : 2023.05.31
  • Accepted : 2023.06.19
  • Published : 2023.06.30

Abstract

Fault detection and diagnosis (FDI) algorithms are actively being researched for ensuring the integrity and reliability of environment perception sensors in autonomous vehicles. In this paper, a fault detection algorithm based on a multi-sensor perception system composed of radar, camera, and lidar is proposed to guarantee the safety of an autonomous vehicle's perception system. The algorithm utilizes reference generation filters and residual generation filters based on finite impulse response (FIR) filter estimates. By analyzing the residuals generated from the filtered sensor observations and the estimated state errors of individual objects, the algorithm detects faults in the environment perception sensors. The proposed algorithm was evaluated by comparing its performance with a Kalman filter-based algorithm through numerical simulations in a virtual environment. This research could help to ensure the safety and reliability of autonomous vehicles and to enhance the integrity of their environment perception sensors.

자율주행 차량의 무결성과 내결함성을 보장하기 위한 환경 인식 센서의 결함 감지 및 격리(FDI) 알고리즘이 중요한 연구 주제로 다루어지고 있다. 본 논문에서는 자율주행 차량의 안전성 보장을 위한 레이다, 카메라, 라이다로 구성된 다중 인지 시스템 결함 검출 알고리즘을 제시하였다. 제안된 결함 감지 알고리즘은 FIR(Finite Impulse Response) 필터 추정치에 기반한 레지듀얼의 생성 및 분석으로 고장의 감지 및 격리를 수행한다. 알고리즘의 성능 검증을 위해 가상환경에서의 수치 시뮬레이션을 수행하여 알고리즘을 기존의 칼만 필터 기반 알고리즘과 비교 및 고찰하였다. 결과적으로 제안된 알고리즘은 인지 시스템의 강건성을 확보할 수 있음을 검증하였다. 본 연구는 자율주행 차량의 안전성과 신뢰성을 확보하기 위한 필수적인 연구로, 자율주행 차량의 환경 인지 센서의 무결성을 향상 시킬 것으로 판단된다.

Keywords

Acknowledgement

본 논문은 산업통상자원부가 지원한 '자율주행기술개발혁신사업'의 지원을 받아 수행된 연구 결과입니다 [과제명: Lv.4 자율주행시스템의 FailOperational 기술개발 / 과제번호: 20018055]. 또한 이 연구는 2023년도산업통상자원부가 지원한 '자동차산업기술개발'의 지원을 받아 수행된 연구 결과입니다. [과제명 : 수요응답형 자동발렛주차 및 서비스 기술 개발 / 과제번호 : 20018448]

References

  1. Bak, Y. S., Shin, S. G., Park, J., Lee, H. K., Eom, S., Cho, S. and Shin, J.(2021), "A Study on the Risk Analysis and Fail-safe Verification of Autonomous Vehicles Using V2X Based on Intersection Scenarios", Korea Institute Intelligent Transportation System, vol. 20, no. 6 pp.299-312. https://doi.org/10.12815/kits.2021.20.6.299
  2. Bak, Y. S., Shin, S. G., Park, J., Lee, H. K., Eom, S., Cho, S. and Shin, J.(2022), "Predicting Accident Vulnerable Situation and Extracting Scenarios of Automated Vehicle using Vision Transformer Method Based on Vision Data", Korea Institute Intelligent Transportation System, vol. 21, no. 5, pp.233-252. https://doi.org/10.12815/kits.2022.21.5.233
  3. Chen, K. Y., Chen, L. S., Chen, M. C. and Lee, C. L.(2011), "Using SVM based method for equipment fault detection in a thermal power plant", Computers in Industry, vol. 62, no. 1, pp.42-50. https://doi.org/10.1016/j.compind.2010.05.013
  4. Goelles, T., Schlager, B. and Muckenhuber, S.(2020), "Fault Detection, Isolation, Identification and Recovery (FDIIR) Methods for Automotive Perception Sensors Including a Detailed Literature Survey for Lidar," Sensors, vol. 20, no. 13, p.3662.
  5. Isermann, R.(2005), "Model-based fault-detection and diagnosis-status and applications", Annual Reviews in Control, vol. 29, no. 1, pp.71-75. https://doi.org/10.1016/j.arcontrol.2004.12.002
  6. Kim, P. S., Lee, E. H., Jang, M. S. and Song, K. S.(2014), "A Bank of Finite Memory Filters for Fault Detection and Adjusting Detection Latency", Applied Mechanics and Materials, vol. 575.
  7. Kobayashi, T. D. and Simon, L.(2005), "Enhanced bank of Kalman filters developed and demonstrated for in-flight aircraft engine sensor fault diagnostics", Research and Technology, NASA Glenn Research Center at Lewis Field, 213419, pp.25-26.
  8. Nemati, F., Hamami, S. M. S. and Zemouche, A.(2019), "A nonlinear observer-based approach to fault detection, isolation and estimation for satellite formation flight application", Automatica, vol. 107, pp.474-482. https://doi.org/10.1016/j.automatica.2019.06.007
  9. Pak, J. M., Ahn, C. K., Lim, M. T. and Song, M. K.(2014), "Horizon group shift FIR filter: Alternative nonlinear filter using finite recent measurements", Measurement, vol. 57, pp.33-45. https://doi.org/10.1016/j.measurement.2014.07.007
  10. Shmaliy, Y. S.(2010), "Linear optimal FIR estimation of discrete time-invariant state-space models", IEEE Transportation on Signal Processing, vol. 58, no. 6, pp.3086-3096. https://doi.org/10.1109/TSP.2010.2045422
  11. Tudoroiu, N. and Khorasani, K.(2007), "Satellite fault diagnosis using a bank of interacting Kalman filters", IEEE Transportation Aerospace Electronics System, vol. 43, pp.1334-1350. https://doi.org/10.1109/TAES.2007.4441743
  12. Wellington, S. J., Atkinson, J. K., Sion, R. P. and Liu, F.(2002), "Sensor validation and fusion using the Nadaraya-Watson statistical estimator", FUSION, IEEE Cat.No.02EX5997, pp.8-11.
  13. Yang, J. W., Lee, Y. D. and Koo, I. S.(2018), "Sensor Fault Detection Scheme based on Deep Learning and Support Vector Machine", The Journal of the Institute of Internet, Broadcasting and Communication(IIBC), vol. 18, no. 2, pp.185-195.
  14. Zhao, S., Shmaliy, Y. S., Huang, B. and Liu, F.(2015), "Minimum variance unbiased FIR filter for discrete time-variant systems", Automatica, vol. 53, no. 2, pp.355-361. https://doi.org/10.1016/j.automatica.2015.01.022