Attitude Estimation of Unmanned Vehicles Using Unscented Kalman Filter

무향 칼만 필터를 이용한 무인 운송체의 자세 추정

  • 송경섭 (조선대학교 대학원 제어계측공학과) ;
  • 고낙용 (조선대학교 전자공학과) ;
  • 최현승 (신한시스템즈(주))
  • Received : 2019.01.15
  • Accepted : 2019.02.15
  • Published : 2019.02.28


The paper describes an application of unscented Kalman filter(UKF) for attitude estimation of an unmanned vehicle(UV), which is equipped with a low-cost attitude heading reference system (AHRS). The roll, pitch and yaw required at the correction stage of the UKF are calculated from the measurements of acceleration and geomagnetic field. The roll and pitch are attributed to the measurement of acceleration, while yaw is calculated from the geomagnetic field measurement. Since the measurement of geomagnetic field is vulnerable to distortion by hard-iron and soft-iron effects, the calculated yaw has more uncertainty than the calculated roll and pitch. To reduce the uncertainty of geomagnetic field measurement, the proposed method estimates bias in the geomagnetic field measurement and compensates for the bias for more accurate calculation of yaw. The proposed method is verified through navigation experiments of a UV in a test pool. The results show that the proposed method yields more accurate attitude estimation; thus, it results more accurate location estimation.

KCTSAD_2019_v14n1_265_f0001.png 이미지

그림 1. 실험에 사용된 수조 및 무인 운송체의 궤적 Fig. 1 Test tank for experiment and trajectory of unmanned vehicle navigation

KCTSAD_2019_v14n1_265_f0002.png 이미지

그림 2. 시간에 따른 편차 추정 결과 Fig. 2 Estimated bias in time

KCTSAD_2019_v14n1_265_f0003.png 이미지

그림 3. xy평면에서의 자기장 측정값 편향 Fig. 3 Shift of magnetic field measurements in xy plane

KCTSAD_2019_v14n1_265_f0004.png 이미지

그림 4. xz평면에서의 자기장 측정값 편향 Fig. 4 Shift of magnetic field measurements in xz plane

KCTSAD_2019_v14n1_265_f0005.png 이미지

그림 5. 롤 추정 오차 Fig. 5 Error in estimated roll

KCTSAD_2019_v14n1_265_f0006.png 이미지

그림 6. 피치 추정 오차 Fig. 6 Error in estimated pitch

KCTSAD_2019_v14n1_265_f0007.png 이미지

그림 7. 요 추정 오차 Fig. 7 Error in estimated yaw

KCTSAD_2019_v14n1_265_f0008.png 이미지

그림 8. 추정된 xy평면 궤적 Fig. 8 Estimated trajectory in xy plane

KCTSAD_2019_v14n1_265_f0009.png 이미지

그림 9. 추정된 xz평면 궤적 Fig. 9 Estimated trajectory in xz plane

표 1. UKF방법을 사용한 편차 추정 알고리즘 Table 1. Bias estimation algorithm using UKF

KCTSAD_2019_v14n1_265_t0001.png 이미지

표 2. 실험에서 사용된 FOG센서의 사양 Table 2. Specification of FOG used in the experiment

KCTSAD_2019_v14n1_265_t0002.png 이미지

표 3. 실험에서 사용된 AHRS센서의 사양 Table 3. Specification of AHRS used in the experiment

KCTSAD_2019_v14n1_265_t0003.png 이미지

표 4. 측정된 자기장 비교 Table 4. Comparison of measured magnetic field

KCTSAD_2019_v14n1_265_t0004.png 이미지

표 5. 추정된 자세에 대한 오차 통계 Table 5. Error statistics of estimated attitude

KCTSAD_2019_v14n1_265_t0005.png 이미지

표 6. AHRS센서에서 측정된 자세에 대한 오차통계 Table 6. Error statistics of AHRS attitude

KCTSAD_2019_v14n1_265_t0006.png 이미지


Grant : 비행시간 증가를 위한 가솔린 발전기 내장형 멀티콥터 개발, 멀티콥터 위치 및 자세추정 기술개발

Supported by : 방위사업청


  1. J. Han, N. Ko, H. Choi, and C. Lee, "Simulation of Sensor Measurements for Location Estimation of an Underwater Vehicle," Korean Institute of Intelligent Systems, vol. 26, no. 3, June 2016, pp. 208-217.
  2. G. Troni, L. Whitcomb "Adaptive Estimation of Measurement Bias in Three-Dimensional Field Sensors with Angular-Rate Sensors Theory and Comparative Experimental Evaluation," Robotics Science and Systems 2013. Berlin, Germany, June 24-28, 2013.
  3. P. Martin, P. douglas, "Expression of a re-centering bias in saccade regulation by superior colliculus neurons," Experimental Brain Research, issue 3-4, Aug. 2001, pp. 354-368.
  4. D. Robinson, "Bias in a least square method of analysing decay data," NUCLEAR INSTRUMENTS AND METHODS. vol. 79, no 1, Mar 1970, pp 65-68.
  5. J. Simin, K. Jeffrey, "A New Extension of the Kalman Filter to Nonliear Systems," Proc. SPIE 3068, Signal Processing, Sensor Fusion, and Target Recognition VI, vol.3068, July 28, 1997.
  6. J. Hashmall, J. Deutschmann, "An Evaluation of Attitude-Independent Magnetometer-Bias Determination Methods - Flight Mechanics," Flight Mechanics Estimation Theory Symposium, NASA Goddard Space Flight Center; Greenbelt, MD United States, May 1, 1996. pp. 169-178.
  7. S. Thrun, W. Burgard, D. Fox, Probabilistic Robotics. Massachusetts: The MIT Press, 2006, pp. 220-223.
  8. S. Noh, N. Ko, and T. Kim, "Implementing Autonomous Navigation of a Mobile Robot Integrating Localization, Obstacle Avoidance and Path Planning," J. of the Korea Institute of Electronic Communication Sciences, vol. 6, no. 1, 2011, pp.148-156.
  9. M. Rhudy, Y. Gu, "Understanding Nonlinear Kalman Filters, Part II: An Implementation Guide," Interactive Robotics Letters(IRL). Morgantown: West Virginia University, 2013.
  10. T. Kim, N. Ko, S. Noh, Y. Lee, "Localization on an Underwater Robot Using Monte Carlo Localization Algorithm," J. of the Korea Institute of Electronic Communication Sciences, vol. 6, no. 2, 2011, pp.288-295.
  11. N. Ko, S. Jeong, S. Hwang, and J. Pyun, " Attitude Estimation Using Field Measurements and Bias Compensation," Sensors, vol.19, no.330, Jan 15, 2019.
  12. S. Noh, T. Kim, N. Ko, and Y. Bae. "Particle filter for Correction of GPS location data of a mobile robot," J. of the Korea Institute of Electronic Communication Sciences, vol.7, no.2, 2012, pp.381-389.
  13. Advanced Navigation. Spatial FOG Reference Manual; Version 2.2 27; Advanced Navigation: Sydney, Australia, 2016.
  14. World Magnetic Model 2015 Calculator. Available online: (accessed on 9 July 2018).