Modeling and simulation of vehicles can be improved by using actual road surface data acquired by Road Surface Measurement Systems. Due to inherent properties of the sensors used, the data acquired is often ridden with outliers. This work addresses the issue of identifying and removing outliers by extending the robust outlier rejection algorithm, Random Sampling and Consensus (RANSAC). Specifically, this work modifies the cost function utilized in RANSAC in such a way that it provides a smooth transition for the classification of points as inliers or outliers. The modified RANSAC algorithm is applied to neighborhoods of data points, which are defined as subsets of points that are close to each other based on a distance metric. Based on the outcome of the modified RANSAC algorithm in each neighborhood, a novel measure for determining the likelihood of a point being an outlier, defined in this work as its exogeny, is developed. The algorithm is tested on a simulated road surface dataset. In the future this novel algorithm will also be tested on real-world road surface datasets to evaluate its performance.