2.5D vehicle odometry estimation

It is well understood that in ADAS applications, a good estimate of the pose vehicle required. This paper proposes metaphorically named 2.5D odometry, whereby planar odometry derived from yaw rate sensor and four wheel speed sensors augmented by linear model suspension. While core already literature, this fitting quadratic to incoming signals, enabling interpolation, extrapolation, finer integration position. shown, experimental results with DGPS/IMU reference, provides highly accurate estimates, compared existing methods. Utilising return change height reference points changing suspension configurations, defined, thus augmenting model. An framework evaluations criteria presented which goodness evaluated has been designed support low-speed surround-view camera systems are well-known. Thus, some application show performance boost for viewing computer vision applications using proposed presented.