A road grade estimation model which uses the curvature to express the rate of change in the grade is proposed in this paper. The assumption that the rate of change in the road grade equals zero is widely accepted in the field of online road grade estimation. This assumption is reasonable to some extent, but it results in an inevitable time lag in the rolling-hills situation. This paper offers a road curvature estimation method which can be used to express the rate of change in the road grade. The recursive least-squares algorithm is used to the estimate the curvature, and then the Kalman filter is employed to estimate the road grade from the other vehicle states. Field tests are performed on a highway in a mountainous area. The offline road grade is used to analyse the instantaneous error and the time lag. The field test results show that the model performs well in reducing the time lag, especially in periods where the gradient changes rapidly.
Piecewise Linear Road Grade Estimation
