The functional elements of decision making system are fuzzy, adaptive and self-learning for intelligent ground vehicles. As is well-known, operating environments of unmanned ground vehicles (UGVs) are complex, unknown and dynamic. And in the meanwhile, modeling of the exact dynamics is relatively difficult. However, the changing of special dynamic parameters and the man-made driving laws of velocities and running direction are easily available. Therefore, this paper attempts to provide an approach based on fuzzy Q-learning algorithm for studying autonomous navigation and control system’s design under multi-constraints, which aims to make the navigation and control system of unmanned vehicles adaptive and robust under complex and changing environment. The presented approach utilizes the human drivers’ empirical knowledge in order to obtain fuzzy control laws. Fuzzy inference system introduces the human beings’ successful experiences into the system, and Q-learning mainly pays more attention to the interaction between the robot and the environment and thus keeps on learning until achieving the goal. Through this model, autonomous navigation and control system can be designed accordingly. In order to facilitate the application of the proposed method, this paper used a type of the nonholonomic robotic system for the computational experiments so as to verify the algorithm, which only considers many candidate conclusions. The final simulation results under different conditions show the validity of the designed algorithm. The presented algorithm is not dependent on the dynamic model, and only has to design in terms of the special model parameters. Therefore, the provided approach based fuzzy Q-learning has strong versatility and transplantable, which can be easily used for penetration maneuver strategies and autonomous maneuver of other type of autonomous vehicles such as unmanned aerial vehicles (UAVs) or autonomous underwater vehicles (AUVs).