Title: Monitoring System Design for Estimating the Longitudinal and Lateral Vehicle

300~500 words): The current automotive industry is confronted with ever-stringent demand for active safety systems since their capabilities such as preventing fatal accidents through artificial intervention are increasingly recognized. To implement Anti-lock Braking Systems (ABS), Traction Control Systems (TCS), etc, the availability of an accurate measure of the longitudinal velocity is the decisive factor and the accuracy is the key for a reliable computation of the longitudinal tire slip, which is the main control variable for designing braking and traction control laws. Information of the lateral velocity in a vehicle is also important factor in many active vehicle safety systems such as yaw stability control, Electronic Stability Control (ESC) and Active Rollover Prevention (ARP). However, the estimation of the longitudinal and lateral velocities is determined not only by estimation algorithms, but also by driving environment such as road bank angle, slope angle and roll angle. The lateral velocity can be estimated through direct integration of gyro signal, but the estimated velocity may diverge in case of banked road since the bank angle acts as a biased element to true lateral acceleration. The estimation of the longitudinal velocity is similar for the case of sloped road. It is believed that there exist few velocity estimation methods which are accurate and cheap enough to be implemented to passenger cars In this study, an Extended Kalman Filter is designed to estimate longitudinal and lateral velocities by considering the road bank angle and roll angle. In order to overcome the aforementioned difficulties, the estimation algorithm is based on the proper combination of the physical model and sensor kinematics for identifying road bank angle and roll angle. The availability of their angles information improves the estimation of the longitudinal and lateral velocities and also prevents vehicle stability systems from being activated unnecessarily. The estimation scheme is given by the block diagram as shown in figure. The Extended Kalman filter is based on longitudinal, lateral, and yaw motions, and the tire force model is combined Dugoff with Burckhardt/Kiencke model. The performance of the proposed estimation methods is evaluated through vehicle simulator software and by experimental field tests in different driving and road conditions. Combined Estimation , , , , i f x y w a a r δ Kinematic-Based Estimation Model-Based Estimation (EKF+ Burckhardt/Kiencke) Tire Model) ˆˆ, x y v v