اسماعيل فخيمي شريف

The primary objective of this research is to design an‎d develop an intelligent steering control model for autonomous vehicles. By leveraging attention mechanisms, the proposed architecture facilitates a deeper an‎d more contextual understan‎ding of the surrounding environment. This model is specifically optimized for the precise detection of obstacles an‎d pedestrians, thereby significantly enhancing driving safety stan‎dards. To address the challenges associated with navigating unknown environments an‎d the scarcity of labeled training data, a Reinforcement Learning (RL) approach is adopted. This method demonstrates high sample efficiency an‎d adaptability in learning complex control policies. Furthermore, the proposed model is rigorously eva‎luated against various baseline algorithms to establish an optimal trade-off between control accuracy, processing speed, an‎d reaction time.

ماشين خودران , مكانيزم توجه , يادگيري تقويتي , مدل هدايت فرمان

Autonomous Vehicle , Attention Mechanism , Reinforcement Learning , Steering Control Model

Esmaiel Fakhimi Sharif

Dr. Seyyed Hamid Haji Seyyed Javadi