Path Planning and Decision Model for Intelligent Robots Based on Reinforcement Learning

Path planning, as a core component of mobile robot technology, aims to find a collision free path for robots to safely and efficiently reach the endpoint in complex and changing environments. In an idealized and simple known environment, traditional path planning algorithms have demonstrated good performance. However, when faced with the complex and unknown environments commonly present in the real world, these algorithms often appear inadequate, making it difficult to fully explore unknown areas and quickly make optimal decisions in dynamic changes, thereby limiting the efficiency and safety of robot path planning. To address this challenge, this paper innovatively proposes an intelligent robot path planning and decision-making model that integrates deep learning (DL) and reinforcement learning (RL). This model utilizes the self-learning and adaptive capabilities of deep reinforcement learning (DRL) to continuously optimize the path selection strategy of robots during their interaction with the environment. The experimental results show that the model exhibits significant advantages in improving path planning efficiency and enhancing environmental adaptability, providing strong technical support for autonomous navigation of mobile robots in complex unknown environments.