Abstract: Aiming at the problem of the path planning for flexible needle insertion into soft tissue, the planning algorithm based on environment properties and random method is proposed. In this method, the unicycle model is adopted, and the sample paths are planned using the Rapidly-exploring Random Tree algorithm. Then the optimal path is selected by the path evaluation function. The function is composed of normalized sub-functions relating to the path length, the number of arcs, the safety degree and the risk grade of obstacles. For different environments and applications, the weight coefficient of each sub-function and the risk level of each obstacle can be adjusted according to surgeon’s clinical experience. Finally, the simulation of the path planning for flexible needle is performed in complicated environments. The results show that the proposed method has better environmental adaptability and provides more interaction, and then has potential for practical application.

Key words: flexible needle, path planning, environment properties, rapidly-exploring random tree

摘要： 针对柔性针软组织穿刺路径规划问题，提出了一种基于环境特征和随机方法的规划算法。该方法采用单轮车模型，运用快速搜索随机树算法得到样本路径，再基于路径长度、弧段数、安全度及障碍物风险分级相结合的路径评估函数，得到最优路径。在路径生成过程中，医生可根据临床经验设置评估函数的权重系数及判断障碍物的风险等级，以适用不同组织的穿刺环境。进行复杂环境下柔性针软组织穿刺路径规划仿真，结果表明该方法环境适用性强，算法交互性好，具有较好的应用潜力。

关键词: 柔性针, 路径规划, 环境特征, 快速搜索随机树