Abstract
Wheelbase preview control scheme including the estimation of state and road information and its performance limit are investigated. It is known that wheelbase preview control based on the conventional Kalman-Bucy filter shows considerable performance improvement but requires a computationally unaffordable estimation scheme whose observer gain needs to be updated online. Instead of the Kalman-Bucy filter, we propose to use state and input observer that does not require any knowledge of road input dynamics. It is found that if appropriate signals are measured to satisfy the existence condition of stable observer, computationally affordable estimation scheme can be derived. Several simulations of a half car model are performed to verify the performance of the proposed control law.