【摘要】 针对快速随机扩展树(Rapidly-Exploring Random Trees, RRT)算法在规划路径中随机性较大、扩展效率较低且规划的路径不利于机器人移动等缺点，提出一种改进的RRT算法。首先，加入目标偏向策略和自适应步长策略，减小RRT的随机性，增强路径规划的鲁棒性和算法的探索能力；其次，引入改进的人工势场法，使算法扩展方向更偏向目标点，提高算法的搜索效率；然后，去除改进RRT算法规划路径的冗余节点，使得生成的路径更加简短和高效；最后，使用n阶贝塞尔曲线对路径进行平滑处理，使规划的路径更有利于机器人移动。实验结果表明，与传统RRT算法、RRT*算法和另外一种已有的改进方法相比，改进的RRT算法在路径规划时间、路径长度、路径节点以及迭代次数等方面效果都更好，路径更加平滑且更短。更多还原

【Abstract】 Aiming at the shortcomings of the Rapidly-Exploring Random Trees(RRT) algorithm in the planning path, such as large randomness, low expansion efficiency, and the planned path is not conducive to robot movement, an improved RRT algorithm is proposed. Firstly, the target bias strategy and adaptive step size strategy are added to reduce the randomness of RRT and enhance the robustness of path planning and the exploration ability of the algorithm. Secondly, the improved artificial potential field method is introduced to make the expansion direction of the algorithm more biased towards the target point and improve the search efficiency of the algorithm. Then, the redundant nodes of the path planned by the improved RRT algorithm are removed to make the generated path shorter and more efficient. Finally, the nth-order Bezier curve is used to smooth the path, so that the planned path is more conducive to robot movement. The experimental results show that compared with the traditional RRT algorithm, RRT* algorithm and another existing improved method, the improved RRT algorithm has better effects in path planning time, path length, path nodes and iteration times, and the path is smoother and shorter.更多还原