面向软组织器官（如肝脏、肾脏、前列腺等）的定位穿刺是外科诊断、治疗中常用的手段。机器人辅助穿刺手术中穿刺精度是成败的关键，影响精度的因素突出体现在两个方面：一是由于穿刺力导致软组织器官变形所引起的肿瘤靶点的漂移；二是由于呼吸运动导致肿瘤靶点逃出计划靶区使得穿刺靶点与术前规划产生较大偏差。事实上，软组织变形和呼吸运动可使肝脏在头尾方向上的运动幅度达12～75mm，这将大大降低穿刺定位精度，严重影响治疗的效果。 本课题以实现机器人高精度软组织靶向穿刺为目标，围绕穿刺针-软组织交互建模、呼吸运动预测与补偿、机器人自动靶点跟踪三个关键科学问题，深入研究基于质量-弹簧和有限元模型基础上的软组织变形建模；研究综合视觉靶点跟踪和加速度计阵列方法的呼吸运动四维建模，实现肿瘤靶点的超前预测；通过将仿真预测结果应用到机器人规划，控制穿刺针位姿预测补偿与靶点自动跟踪，实现对软组织肿瘤靶点的一次性精准穿刺。

Locating puncture for sofe tissue organ (liver, kidney,prostate etc.) is a common method of surgical diagnose and treatment. The insertion accuracy is directly related to the robot-assisted insertion surgery success or failure of an important part.The insertion accuracy is badly affected by two key factors.One is the drift of tumor target caused by soft tissue deformation ;The other is that the tumor target go out of the planning region caused by respiratory motion which causes great error compared with preoperative planning. In the fact, the soft tissue deformation and respiratory motion cause liver ranges of motion 12～75mm front to back which badly reduces the insertion accuracy and affects treatment. The goal of the project is to achieve robot-assisted high accurately insertion to soft tissue target. These are three key scientific problems that are interactive modeling between needle and soft tissue, predictive compensation of respiratory motion and robot tracking the target automatically. we will build the model of sofe tissue deformation based on the anlaysis and synthesis of mass-spring model and finite element model; build the four-dimensional dynamic model and forecast location of tumor target based on anlaysis and synthesis of the data from tracking vision target method and accelerometer array method. The control algorithm of the robotic system for microwave ablation is designed with the emulational and predictive results which makes robot tracking the target automatically. The location and posture of puncture needle are compensated which makes the needle precision puncture to the tumor target just once.