Inter-Arm Registration Using Image and Kinematic Velocity Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Establishing and maintaining accurate registration between computer-assisted devices with repositionable arms is challenging, especially when one device has an imaging device mounted to it, as the geometric relationships between instruments are difficult to determine, affecting teleoperation precision.
Innovation Solution
A control unit is used to receive images from an imaging device, determine the observed and expected velocities or orientations of an instrument, transform them to a common coordinate system using a registration transform, and update the transform based on errors to maintain accurate registration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If an imaging device is mounted to a repositionable arm to capture images of the workspace, then the operator can monitor the activity and change the field of view, but the geometric relationships (registrations) between the imaging device and other instruments become difficult to determine
Solution Approach 1:
The system continuously captures images of the instrument from the imaging device, determines observed velocities of features, compares them with expected velocities based on kinematic models, and uses the error between observed and expected velocities to iteratively refine the registration transform. This closed-loop feedback mechanism maintains accurate geometric relationships despite device repositioning.
Solution Approach 2:
The patent replaces traditional mechanical calibration methods with an image-based visual feedback system. Instead of relying on physical alignment tools or manual coordinate measurement, the system uses computer vision to track instrument features in images and automatically computes registration transforms through velocity comparison and error minimization.
2Extent of automation
If the operator uses input controls to provide movement commands for instruments mounted to repositionable arms, then teleoperation is enabled, but accurate translation of operator movements to instrument motions requires precise registration that is difficult to maintain
Solution Approach 1:
The system establishes a closed-loop control mechanism where the registration transform is continuously refined by comparing observed instrument velocities from images with expected velocities from kinematic models. This feedback ensures that teleoperation commands are accurately translated to instrument motions even as devices are repositioned during the procedure.
3Ease of manufacture
If traditional calibration methods are used to establish geometric relationships, then initial registration can be obtained, but the registration becomes inaccurate when devices are repositioned during use
Solution Approach 1:
The system transitions from static calibration to dynamic registration maintenance. Instead of performing a one-time calibration that becomes obsolete when devices move, the system continuously updates the registration transform during operation by comparing real-time observed velocities with expected velocities from kinematic models, ensuring accuracy is maintained throughout the procedure.
Solution Approach 2:
The patent implements continuous feedback-based registration refinement where the system monitors instrument positions and images, calculates velocity discrepancies, and iteratively adjusts the registration transform. This dynamic feedback loop maintains reliable geometric relationships even as devices are repositioned, overcoming the limitations of traditional static calibration methods.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods for inter-arm registration include a computer-assisted system having a control unit coupled to a repositionable arm of a computer-assisted device. The control unit is configured to: receive, from an imaging device, successive images of an instrument mounted to the repositionable arm; determine an observed velocity of a feature of the instrument; determine an expected velocity of the feature of the instrument based on kinematics of the repositionable arm; transform the observed velocity and/or the expected velocity to a common coordinate system using a registration transform; determine an error between directions of the observed and expected velocities in the common coordinate system; and update the registration transform based on the determined error. In some embodiments, the instrument is a medical instrument and the imaging device is an endoscope. In some embodiments, the control unit is further configured to control the instrument using the registration transform.