HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation

Contributions à l'estimation de l'attitude et à la détection de l'hypovigilance au volant

Abstract : This thesis focuses on the estimation of the attitude of rigid bodies and on the detection of driver drowsiness. When estimating attitude, several practical problems may arise. A first problem concerns the estimation of attitude when the body is subject to a non-negligible external acceleration. We propose, in this context, observers that estimate the attitude of the considered object while limiting the effects of the external acceleration on the quality of the estimate. We have also addressed another practical problem: the gyroscope bias. Indeed, the structure of gyroscopes can lead to a bias in the measurements. This bias can be assimilated to an unknown parameter and must be estimated. For this purpose, we have designed an adaptive observer for a class of uncertain non-linear systems with multi-sampled outputs. These outputs, corresponding to several sensors, can be sampled irregularly and independently of each other. The proposed observer has been applied in particular to the problem of simultaneous estimation of attitude and bias. This thesis also deals with the detection of driver drowsiness, which is one of the many fields of application of attitude estimation. We have shown that the driver's level of drowsiness can be estimated from non-intrusive data. This issue was raised in order to conceptualize a reliable and non-intrusive device for detecting drowsiness.
Document type :
Complete list of metadata

Contributor : Abes Star :  Contact
Submitted on : Friday, December 11, 2020 - 10:38:27 AM
Last modification on : Saturday, May 21, 2022 - 3:25:32 AM
Long-term archiving on: : Friday, March 12, 2021 - 7:07:46 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03053718, version 1


Tristan Bonargent. Contributions à l'estimation de l'attitude et à la détection de l'hypovigilance au volant. Automatique / Robotique. Normandie Université, 2020. Français. ⟨NNT : 2020NORMC212⟩. ⟨tel-03053718⟩



Record views


Files downloads