Traction control systems are a fundamental active safety equipment of vehicles; they control wheel slip when excessive torque is applied on driving wheels, helping the driver to bring the vehicle under control and improving handling and stability when starting or accelerating and especially under poor or slippery road conditions. The aim of this work is to develop a parameter estimation block for further development of an intelligent traction control system. To evaluate the performance of the proposed estimation algorithm, estimated variables are compared making use of BikeSim 2.0 ®. Parameter estimation was performed using an extended Kalman filter optimized using genetic algorithms. Using an artificial neural network, the slip that maximizes the tire-road friction coefficient is identified.