This document summarizes a probabilistic algorithm for online 3D mobile robot mapping presented by Noha Quan Ravi. The algorithm uses an online Expectation-Maximization approach to build 3D maps from laser rangefinder and camera sensor data in real-time. It models the environment as planar surfaces and represents each surface with nine parameters. The algorithm iterates between estimating surface correspondences to measurements (E-step) and re-estimating the surface parameters (M-step). It processes a constant number of new measurements at each time step to allow real-time mapping performance as the robot moves through its environment. Experimental results demonstrate it can successfully build 3D maps of indoor environments in real-time.