- For the last two decades, intelligent traortation systems have emerged as an efficient way of improving the performance of traortation systems. One of the key components of these systems is vehicle control system with the aim of developing intelligent vehicles. Intensive researches are carried out in the automotive industry for development of these vehicles that can make the driving procedure easier and more secure. Two kinds of problems in automated control of vehicles are driving on road and automatic parking. Parking a car can be very cumbersome and difficult in many circumstances, so autonomous parking problems of the vehicle have attracted a great deal of attention recently from research organization. This study aims at developing an autonomous parallel parking controller for a vehicle. At first, this system is designed based on the path planning approach. In this approach, a feasible geometry path is planned in advance taking into account the environmental model as well as the nonholonomic constraints, and then control commands are generated to follow the reference path. In order to obtain zero velocity for the vehicle at the initial and goal points of the parking maneuver, as well as, at the point that direction of motion must change, two timing laws for forward and backward maneuvers are proposed and control laws are designed in timing laws-domain. For parking the vehicle in limited space, it is known that switching of moving direction is needed to execute the parking motion. During the motion, the vehicle has to follow the desired given path which may be a consecutive path after a reverse without well convergence for the former desired path. This may cause insufficient convergence to a final desired point which is inside of the parking space in the current problem. In order to overcome this inconvenience, the path tracking controller must be very accurate. A feedback linearization controller is proposed for the reference path tracking. By using this controller, the reference path can be followed even with initial tracking errors and bounded disturbances but, it doesn’t exhibit robustness against parameter variations. In order to construct a robust control structure for tracking the desired path, Sliding Mode Control structure is proposed and stability analysis and controller design theorem of this structure are obtained. The tracking error of these two controllers is negligible, so the vehicle is able to perform the parallel parking task with high accuracy. Also in this thesis, a method for park the vehicle without knowing the parking space dimension, based on RBF network and sliding mode and feedback linearization controller is developed. In this method, the reference path is generated by using the RBF network based on the sonar measurement. The algorithm is computationally efficient because there is no need of any offline path planner which reduces the computational time. Also in this method, due to change of the desired path and direction of motion based on the vehicle’s configuration, the vehicle can be parked in limited area. In each section, the simulation results show effectiveness and applicability of proposed approach. Key words: parallel parking, nonholonomic constraint, kinematic model, feedback