【摘要】 针对基于目标偏向的RRT算法在复杂环境下易陷入局部搜索与搜索效率低的不足，在ROS中分别基于Dobot模型和DIY模型，对算法做出进一步的改进。在创建机械臂模型基础上，利用MoveIt!对其进行配置，并通过简化模型来提高碰撞检测效率；在基于目标偏向RRT算法的基础上，分别引入扩展目标变更策略和可变步长对算法进行改进；在ROS中分别进行了不同形态障碍物环境下的避障运动规划仿真；在样机上进行了避障和避障抓取实验。仿真及实验结果表明，提出的改进RRT算法相比基于目标偏向的RRT算法，规划成功率和效率分别可提高42.9%和29.2%。更多还原

【Abstract】 Aiming at the fact that the RRT algorithm based on the target bias is easy to fall into local search and has a low search efficiency in complex environment,further improvement is made in the ROS based on the Dobot model and the DIY model.On the basis of creating the manipulator model,the Move It!is used for configuration and the model is simplified to improve the efficiency of the collision detection.The extended target change strategy and variable step size are introduced to improve the algorithm which is based on the target biased RRT algorithm.The simulation of the manipulator obstacle avoidance motion planning under different forms of obstacles is carried out in the ROS.The experiments of the obstacle avoidance and obstacle avoidance grasping are carried out on the prototype.Simulation and experimental results show that the proposed improved RRT algorithm can improve the planning success rate and efficiency by 42.9%and 29.2%respectively compared with the RRT algorithm based on target bias.更多还原