One of the major challenges in the development of autonomous systems is designing intelligent mechanisms for parallel parking in complex an‎d highly congested urban environments. In this study, a framework based on path planning an‎d model predictive control is presented, developed with a focus on high accuracy an‎d adaptability to both static an‎d dynamic obstacles.Initially, two common path planning algorithms, RRT an‎d Hybrid A*, were reviewed an‎d compared. Experimental results showed that the RRT algorithm offers significantly faster performance in terms of execution time compared to other methods. Accordingly, its improved version, RRT*, was selec‎ted an‎d combined with Reeds-Shepp curves an‎d a Spline smoothing method to enhance performance in the presence of static obstacles.Numerical eva‎luations indicated that the combined RRT* structure outperforms other methods in terms of path smoothness, scalability, an‎d path length reduction—especially in environments with fixed obstacles. These results demonstrate that leveraging advanced kinematic an‎d geometric methods can play a crucial role in path optimization for real-world applications. For path tracking, Model Predictive Control (MPC) was employed. To improve control performance, the MPC parameters were tuned using a Bayesian optimization algorithm, which enhanced accuracy, stability, an‎d eliminated the need for trial-an‎d-error tuning. In dealing with dynamic obstacles, Reeds-Shepp-based paths were replaced by a motion model due to their limited responsiveness, enabling real-time path redesign an‎d equipping the system with higher reactivity. Simulation results showed that the proposed framework is capable of generating safe, smooth, an‎d executable paths in real time. By integrating classical an‎d modern techniques an‎d leveraging intelligent optimization, this approach represents an effective step toward the practical realization of autonomous parallel parking systems in the next generation of smart vehicles.

جعفر قيصري

يداله ذاكري حسين آبادي

حامد جلالي بيدگلي , فريد شيخ الاسلام