Continuously updated knowledge of the friction potential and the friction demand can help to improve manoeuvrability and thereby safety of vehicles under slippery road conditions. An on line estimation method, based on recognition of the pneumatic trail, is presented using a simple brush type tyre model. The method is verified by outdoor experiments using a tyre test trailer. Disturbances such as camber, residual forces and moments, combined slip of the tyre and dynamic behaviour of the tyre are considered. Neural networks have been used both for modelling and identification. Finally, a low-cost implementation of the system in the front wheel suspension of a passenger car is discussed.