Advanced Driver Assistance System (ADAS) can assist the driver to avoid the traffic accidents due to distractions. However, if the sensor fault occurs while measuring the vehicle dynamics, ADAS might make wrong decisions and cause undesired consequences. A model-based sensor fault detection algorithm is proposed in this paper to detect and isolate the faulty sensor of yaw rate, lateral acceleration, steering wheel angle, longitudinal speed, longitudinal acceleration, or wheel speed. Unlike the conventional algorithms, the proposed algorithm can detect the sensor faults for both longitudinal and lateral dynamics simultaneously. Since wheel speeds are employed to check some measurements of the vehicle dynamics, they are validated by calculating the wheel speed deviations using four wheel speed sensors. Kinetic models are employed to estimate yaw rate, lateral acceleration, and steering wheel angle. A kalman filter based on a point mass model is employed to estimate longitudinal speed and acceleration. The estimated vehicle dynamics and sensor measurements are used to calculate the residuals. Adaptive threshold values are employed to identify the abnormal increments of residuals. Recursive least square method is used to design the coefficients of the expressions for adaptive threshold values, such that the false alarms caused by model uncertainties can be prevented. Different combinations of estimations are employed to obtain 22 residuals. Unlike the conventional algorithms, the proposed algorithm can detect the faulty sensor without assuming that other sensors are normal. If a residual is larger than the adaptive threshold value for a certain time, the associated sensor is determined to be faulty. Multiple sensor faults can also be detected. The simulation result show that when the sensor fault occurs, the corresponding residuals will be abnormally increased and larger than the adaptive threshold values. The proposed algorithm can successfully detect the sensor fault for the test scenarios.