End-Effector Positioning Control With Predictive Feedback Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current techniques for continuous positioning control of interventional devices, particularly in clinical settings, face challenges due to the complexity of continuum robot structures and sensitivity to environmental changes and manufacturing inaccuracies, leading to inaccurate kinematic predictions and control issues.
Innovation Solution
A predictive model-based approach for feed-forward and feedback positioning control, utilizing forward and inverse kinematics, and data collection techniques to improve the accuracy and consistency of positioning control, incorporating imaging data and shape sensing for real-time adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a static model with nonlinear differential equations is used to account for lance deformation, then the positioning control of the end-effector can be achieved, but the accuracy is susceptible to environmental changes and manufacturing inaccuracies
Solution Approach 1:
The patent implements feedback control by continuously measuring the actual position of the end-effector using imaging devices and shape sensors, comparing it with the predicted position from the predictive model, and adjusting the control inputs to minimize positioning errors. This closed-loop feedback mechanism compensates for environmental changes and manufacturing inaccuracies, maintaining consistent positioning accuracy.
Solution Approach 2:
The patent employs a predictive model that is trained offline using collected data to predict the relationship between control inputs and end-effector position before actual operation. This preliminary action allows the system to pre-compensate for known environmental factors and manufacturing variations, improving positioning accuracy without requiring real-time complex calculations.
2Adaptability or versatility
If the degree of freedom of the robot is increased to become more maneuverable, then the robot can navigate complex anatomical environments, but the kinematics becomes more complicated
Solution Approach 1:
The patent introduces a predictive model as an intermediary between the control inputs and the robot's motion. Instead of directly computing complex kinematics for high-degree-of-freedom continuum robots, the predictive model learns the mapping from control inputs to end-effector position, simplifying the control problem while maintaining maneuverability in complex anatomical environments.
Solution Approach 2:
The patent replaces traditional mechanical kinematic modeling with a data-driven predictive model. By using machine learning to approximate the complex kinematic relationships, the system achieves maneuverability in complex environments without the computational burden of exact kinematic models for high-degree-of-freedom continuum structures.
3Ease of operation
If continuous control is implemented for continuum robots, then smooth positioning can be achieved, but the complexity of modeling and actuator control increases significantly
Solution Approach 1:
The patent creates a simplified virtual model (predictive model) that copies the essential input-output behavior of the complex continuum robot system. This virtual model allows continuous control to be implemented by manipulating the simplified model rather than the full complex system, achieving smooth positioning with reduced control complexity.
Solution Approach 2:
The patent transforms the control problem by changing parameters from direct manipulation of multiple actuator inputs to controlling a reduced set of parameters through the predictive model. This parameter transformation simplifies continuous control while maintaining smooth positioning performance.
Data Source
AI summary
A positioning controller (50) including an imaging predictive model (80) and inverse control predictive model (70). In operation, the controller (50) applies the imaging predictive model (80) to imaging data generated by an imaging device (40) to render a predicted navigated pose of the imaging device (40), and applies the control predictive model (70) to error positioning data derived from a differential aspect between a target pose of the imaging device (40) and the predicted navigated pose of the imaging device (40) to render a predicted corrective positioning motion of the imaging device (40) (or a portion of the interventional device associated with this imaging device) to the target pose. From the predictions, the controller (50) further generates continuous positioning commands controlling a corrective positioning by the interventional device (30) of the imaging device (40) (or said portion of interventional device) to the target pose based on the predicted corrective positioning motion of the interventional device (30).