Robotics knowledge is increasing on a daily basis and it is expected that this knowledge will affect all aspects of humans lives. In the meanwhile, mobile robots have extra importance because they can perform both simple and complex tasks effectively. Extracting an accurate and complete map of the world is among the most difficult tasks in mobile robotics. Generally, there are three main tasks for a mobile robot: localization, mapping and path planning(motion planning). One of the most important tasks in a mobile robot is exploration which is a combination of mapping and path planning. Although the problem of exploration is the combination of mapping and path planning, but in practice, without localization, solving the problem is not feasible. As a result, the combination of exploration with simultaneous localization and mapping(SLAM) should always be considered to work on a real environment. This type of robot is able to navigate through an unknown environment without human supervision, scan the environment and create a map of it while avoiding obstacles. Mobile robot working environments are unstructured with high uncertainty. Generally, there are five sources of uncertainty including: environment, sensors, actuators, mathematical models and calculations. To deal with these uncertainties, probabilistic robotics is used in this research. Probabilistic robotics uses probability theory to consider uncertainty in perception and action. In this way, it can solve the three main tasks of the mobile robot. In this regard, enabling the mobile robot to learn the map of the world and choose the optimum destination to continue the autonomous scanning of the environment, is the topic of this thesis. For this purpose, the Rao-Blackwellized particle filter for occupancy grid maps is used to determine the pose of the robot and create the map of the world. Furthermore, a two layered architecture is used for motion planning, considering the path planning and obstacle avoidance. The fundamental of exploration problem is determining the optimum goal among the other potential goals. The main objective of this research is to gain the ability of selecting the goal without human intervention and creating the map of the world autonomously. Hence in this thesis, the frontier-based exploration strategy is applied to the robot for the purpose of selecting the optimum goal. For practical implementation of the exploration problem, Robot Operating System(ROS) framework is used as the software platform of the robot. Moreover, the area of the fifth floor of the mechanical engineering department of the Isfahan university of technology is chosen as the unknown environment to test the robot. As a result, the frontier-based approach turned to be an effective algorithm for exploring the large and cluttered environments. Keywords: Frontier-Based Exploration, SLAM, Exploration for SLAM, Autonomous Mobile Robot, Rao-Blackwellized Particle Filter, Occupancy Grid Map