Details

Autor(en) / Beteiligte

• Zhou, Zebo
• Wu, Jin

Titel

Cascaded Indirect Kalman Filters for Land-Vehicle Attitude Estimation With MARG Sensors and GNSS Observations

Ist Teil von

• IEEE transactions on vehicular technology, 2021-04, Vol.70 (4), p.3267-3282

Ort / Verlag

New York: IEEE

Erscheinungsjahr

2021

Quelle

IEEE Electronic Library (IEL)

Beschreibungen/Notizen

• This paper proposes a novel quaternion-based attitude estimation method for land-vehicle applications by fusing the low-cost magnetic, angular rate, and gravity (MARG) sensors and global navigation satellite systems measured velocity (GNSS-V). A structure of three-layer cascaded indirect Kalman filter (IKF) is devised to minimize the interactions of heterogeneous error sources on attitude quaternion solutions. It sequentially fuses the Earth's gravity vector, magnetic vector, and GNSS-V vector pairs with quaternion state constraints from Euler angles in each local filter. To further ensure the robustness of quaternion-based attitude solution, the external acceleration term is adaptively compensated within a sliding window of GNSS-V information. Meanwhile, independent and cross check procedures for observation vectors are also established for excluding individual sensor fault by taking land-vehicle dynamics into account. It is of significant importance that, in this contribution, the GNSS measured velocity is fully utilized in all potential aspects of aiding attitude determination with MARG sensors. Real-world vehicular experiments are carried out to evaluate the performance of our proposed method. The presented results verify the validity and efficiency thus the developed method may be promising for implemention in land-vehicle applications.