Localization is vital in autonomous driving, which an accurate vehicle position enables. There are many studies on different localization algorithms, like Montecarlo and Kalman filters, for sensor fusion and localization of the vehicle. Particle filters are one of the techniques that help accurately estimate the vehicle’s position, even in non-linearities. Such an algorithm is presented in this study for vehicle position estimation integrating data from GPS and IMU sensors, which is generated from a sensor, and simulation is carried out for demonstration. The algorithm differs from the available study in that it can create particles differently. The results demonstrate that increasing the number of particles in the filter significantly enhances accuracy and reduces error. This approach is particularly effective for autonomous navigation in complex and dynamic environments, making it an ideal candidate for future autonomous driving and Advanced Driver Assistance Systems (ADAS).