2.5D vehicle odometry estimation
Ciarán Eising,Leroy‐Francisco Pereira,Jonathan Horgan,Anbuchezhiyan Selvaraju,John McDonald,Paul Moran
DOI: https://doi.org/10.1049/itr2.12143
IF: 2.7
2021-11-24
IET Intelligent Transport Systems
Abstract:It is well understood that in ADAS applications, a good estimate of the pose of the vehicle is required. This paper proposes a metaphorically named 2.5D odometry, whereby the planar odometry derived from the yaw rate sensor and four wheel speed sensors is augmented by a linear model of suspension. While the core of the planar odometry is a yaw rate model that is already understood in the literature, this is augmented by fitting a quadratic to the incoming signals, enabling interpolation, extrapolation, and a finer integration of the vehicle position. It is shown, by experimental results with a DGPS/IMU reference, that this model provides highly accurate odometry estimates, compared with existing methods. Utilising sensors that return the change in height of vehicle reference points with changing suspension configurations, a planar model of the vehicle suspension is defined, thus augmenting the odometry model. An experimental framework and evaluations criteria is presented by which the goodness of the odometry is evaluated and compared with existing methods. This odometry model has been designed to support low‐speed surround‐view camera systems that are well‐known. Thus, some application results that show a performance boost for viewing and computer vision applications using the proposed odometry are presented.
engineering, electrical & electronic,transportation science & technology