Estimation of lateral motorcycle dynamics and rider action with Luenberger observer

Abstract

Abstract-This paper deals with lateral dynamics estimation of powered two-wheeled (PTW) vehicles. A linear parametervarying (LPV) model of the motorcycle is presented and then transformed into a Takagi-Sugeno (TS) form in order to design the observer under the variation of the longitudinal velocity. This paper presents a new way to estimate every motorcycle dynamic states including unknown input (UI) by keeping a simple observer structure with a Luenberger observer in TS form. The observer convergence study is based on the Lyapunov theory associated with LMI tools and L2-gain to guaranty boundedness of the state estimation error. The effectiveness of the proposed solution is illustrated for a nominal and an uncertain simulation cases.

Steering and Lateral Motorcycle Dynamics Estimation: Validation of Luenberger LPV Observer Approach

“…In 2008, precursor work was published about state estimation for motorcycles in [14]. In a first time, researchers have commonly addressed the estimation of the lean angle [15], [16], [17] and rapidly the aim was to estimate the whole dynamics states [18], [19] and more recently in [20] and [21]. These works dealing with P2WV dynamics estimation highlight the strong potential of these techniques.…”

Rider weight consideration for observer design with an application to the estimation of the lateral motorcycle dynamics and rider's action

“…The development of the proposed method is based on our previous work [4]. In this paper, an observer is synthesized to estimate the lateral motorcycle dynamics states and the rider's action on the handlebar.…”

“…Notice that here the L 2 -gain technic is not used to handle the model's uncertainties but in order to guarantee the error convergence of the augmented system. In [4] the effectiveness of the proposed nominal observer has been tested for a simulation scenario with two successive double lane changes (DLC) at variable speed. In addition, a simple DLC scenario at constant speed 100 km/h is simulated with the nominal observer and then with rider weight's uncertainty in the model.…”

“…The present section is based on our previous works. The main steps of the model's derivation are summarized here but for more detail please refer to [19] and [4]. Lateral dynamics of the motorcycle involves the roll inclination, the yaw rotation, and the steering and lateral motions of the bike.…”

“…Among all the litterature, one can cite [1] and [2] where authors adressed longitudinal dynamic estimation whereas in [3], [4], [5], [6], [7] and [8] authors designed observers for lateral dynamics. An important difficulty during the observer's design for PTWV is the uncertainty of some parameters like the tire's properties (camber stiffness, cornering stiffness, radius of the wheel, etc.…”