Physiological motion compensation method and system for a robotic navigation system
By constructing a four-dimensional dynamic organ model and using reinforcement learning for active predictive control, the problem of target displacement and anatomical deformation caused by respiratory motion in bronchial interventional surgery was solved, achieving high-precision navigation compensation and improved safety.
Patent Information
- Application Number
- CN202610429152.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-02
- Publication Date
- 2026-07-17
AI Technical Summary
Existing navigation technologies are unable to compensate for target displacement and anatomical deformation caused by the respiratory movements of subjects in real time during bronchial interventional surgery, resulting in large positioning errors and low safety. Furthermore, existing robot control strategies are outdated and cannot achieve high-precision dynamic compensation.
A robot navigation system based on a four-dimensional dynamic organ model is adopted. By constructing a continuous spatiotemporal feature field and combining six-plane decomposition technology and active predictive control based on reinforcement learning, high-fidelity, real-time compensation of respiratory motion is achieved. The system uses parameterized primitives to represent organ features and physical properties and generates forward-looking compensation control commands.
It achieves high-precision target hitting, reduces dynamic tracking errors, improves surgical safety and operational smoothness, and ensures navigation accuracy and safety in complex environments.
Smart Images
Figure CN122398464A_ABST