摘 要
随着智能交通系统的快速发展，自动驾驶技术已成为研究热点，其中路径规划与避障算法是实现车辆自主导航的核心问题。本研究旨在设计一种高效、安全的路径规划与避障算法，以应对复杂动态环境中的驾驶挑战。为此，提出了一种融合全局路径规划与局部避障优化的综合算法框架，结合改进的A*算法和模型预测控制（MPC）方法，分别用于静态环境下的最优路径生成和动态障碍物条件下的实时轨迹调整。在全局路径规划中，通过引入多权重因子优化路径选择策略，显著提升了路径平滑性和计算效率；而在局部避障阶段，采用基于深度强化学习的决策机制，增强了算法对不确定环境的适应能力。实验结果表明，所提算法能够在多种典型场景下实现高精度路径规划，并有效规避动态障碍物，同时保持较低的计算开销。此外，该算法在复杂交通流中的避障成功率达到了95%以上，验证了其可靠性和实用性。本研究的主要贡献在于提出了兼顾全局与局部优化的综合算法，并通过创新性的多模态融合策略提升了自动驾驶车辆的环境感知与决策能力，为未来智能交通系统的发展提供了重要技术支持。
关键词：自动驾驶；路径规划；避障算法；模型预测控制；深度强化学习
Abstract
With the rapid development of intelligent transportation systems, autonomous driving technology has become a research hotspot, where path planning and obstacle avoidance algorithms are core issues for achieving vehicle autonomous navigation. This study aims to design an efficient and safe path planning and obstacle avoidance algorithm to address driving challenges in complex dynamic environments. To this end, a comprehensive algorithm fr amework integrating global path planning with local obstacle avoidance optimization is proposed, combining an improved A* algorithm and Model Predictive Control (MPC) method for optimal path generation in static environments and real-time trajectory adjustment under dynamic obstacle conditions, respectively. In global path planning, the introduction of multiple weighting factors optimizes the path selection strategy, significantly enhancing path smoothness and computational efficiency; in the local obstacle avoidance phase, a decision-making mechanism based on deep reinforcement learning is employed, strengthening the algorithm's adaptability to uncertain environments. Experimental results demonstrate that the proposed algorithm can achieve high-precision path planning in various typical scenarios while effectively avoiding dynamic obstacles, maintaining low computational overhead. Additionally, the obstacle avoidance success rate in complex traffic flows exceeds 95%, validating its reliability and practicality. The primary contribution of this study lies in proposing a comprehensive algorithm that balances global and local optimization, and through innovative multimodal fusion strategies, enhances the environmental perception and decision-making capabilities of autonomous vehicles, providing critical technical support for the future development of intelligent transportation systems.
Keywords: Autonomous Driving；Path Planning；Obstacle Avoidance Algorithm；Model Predictive Control；Deep Reinforcement Learning
目 录
摘 要 I
Abstract II
一、绪论 1
（一）自动驾驶路径规划与避障研究背景 1
（二）路径规划与避障算法的研究意义 1
（三）国内外研究现状综述 2
二、路径规划算法基础理论 2
（一）路径规划的核心概念与分类 2
（二）常见路径规划算法分析 3
（三）算法性能评估指标体系 3
三、避障算法关键技术研究 4
（一）障碍物检测与建模方法 4
（二）动态避障算法设计思路 4
（三）避障算法的优化策略 5
四、路径规划与避障算法集成应用 5
（一）自动驾驶场景下的算法融合 5
（二）实验平台搭建与测试方案 6
（三）算法性能验证与结果分析 7
结 论 7
致 谢 9
参考文献 10
