Abstract: Reinforcement Learning (RL) has shown promise in robotic navigation tasks, yet applying it to real-world environments remains challenging due to dynamic complexities and the need for ...