Abstract
In board tire/road friction estimation is of current interest in two different frameworks:
- Optimization of driver assistance systems efficiency: antilock braking system, electronic stability program, adaptive cruise control, lane departure control, advanced automatic driving etc.
- Driver instantaneous warning about the available friction and limits for his possible driving actions.
This subject has been the objective of many research programs throughout the world. Four main methods have been investigated and developed: instrumented hubs, sensors embedded in tires and two other approaches named "effect based" and "cause based". In the "effect based" method the friction is estimated from command and dynamic response of the vehicle measurements and in the "cause based" method, the friction is estimated from a predictive model based on road parameter influence knowledge.
This paper comments the three first methods and discuss in a deeper manner the cause based method. This discussion is founded on the LCPC (French Road and Bridges Central Laboratory) researches on road factors influencing tire road friction being conducted in France since 1996. In the framework of European, national and in-house research programs, investigations were conduced at LCPC to analyze the effect of road surface condition on available friction and to investigate and model the effect of factors such as asperities shapes, levels, distributions in the two texture scales (microtexture, macrotexture). Regarding friction/road characteristics relationships significant progresses on knowledge have been accomplished. Predictive parametric models were successfully built up for key factors of longitudinal friction forces (longitudinal stiffness, peak friction and locked wheel friction) from texture parameters for a given surface condition. The more general predictive parametric model requires variables describing asperities angularities, asperities envelopes angles and their distribution at the microtexture scale (few micrometres) and variable describing the macro contact of tire blocks on pavement.
Possibilities to get in board texture data and surface condition are considered. With respect to road condition in board sensor are capable to estimate different states of wetness, but additional information (surface type and texture) are needed to interpret this information in a quantitative manner. With regard to texture, available sensors are expensive but recent research have shown that high resolution digital image processing, a promising solution, could be applicable in medium term.
Getting in board a good estimate of the current tire/road friction, in due time with regard to safety is a question unsatisfactory worked out until now. For the time being, it seems that exploiting routine measurements of texture and friction carried out to assist road authorities for maintenance should be used as reference data to correct locally from in board estimation based on both tire slip assessment and low cost sensor (road condition and possibly macrotexture).
Keywords:tire road friction, slip rate, wetness, road texture