Computer Vision Positioning for Independently Mobile Surgical Robots
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining the relative positioning of robotic arms in surgical robotic systems is challenging when they are mounted on separate carts that can be independently moved, as kinematic methods are inadequate for such configurations.
Innovation Solution
Utilizing computer vision with cameras and markers/fiducials to determine the relative positions of robotic arms and patient beds by processing images and applying computer vision algorithms to recognize markers associated with the arms, incorporating machine learning for enhanced accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If robotic arms are mounted on separate independently moveable carts, then the system provides flexibility and adaptability in positioning, but the ability to determine relative positioning between arms deteriorates because kinematic methods are no longer sufficient
Solution Approach 1:
The patent introduces markers as intermediary objects attached to each robotic arm and cart. These markers serve as mediators between the independently moveable carts and the camera system, enabling the vision system to detect and calculate relative positions. The markers provide visual reference points that bridge the gap between separate carts, allowing the system to determine relative positioning despite the lack of mechanical connection.
Solution Approach 2:
The patent replaces mechanical/kinematic positioning determination with an optical vision-based system. Instead of relying on mechanical linkages or encoders to determine relative positions, the system uses cameras to capture images of markers and computationally determines positions. This substitution enables relative positioning measurement in systems where mechanical methods are insufficient due to independent cart movement.
2Measurement precision
If cameras and markers are introduced to determine relative positions, then measurement precision of relative positioning improves, but device complexity increases due to additional components
Solution Approach 1:
The camera system serves multiple functions: it captures images for relative positioning determination, tracks marker positions for orientation calculation, and provides visual feedback for system calibration. By making the camera system multi-functional, the patent reduces the need for separate dedicated components for each function, thereby limiting the increase in overall device complexity while maintaining high measurement precision.
3Measurement precision
If computer vision algorithms and machine learning are used to process images and determine positions, then measurement precision and automation improve, but loss of time for image processing and computation increases
Solution Approach 1:
The patent implements preliminary calibration procedures where the system captures images of markers at known positions and pre-computes transformation matrices and coordinate system relationships. This preliminary action stores calibration data that can be rapidly applied during actual surgical procedures, reducing the computational time required during real-time operation while maintaining high measurement precision.
Data Source
AI summary
A robot-assisted surgical system comprises a plurality of robotic system components independently positionable relative to a patient in a room, at least one of the robotic system components being a robotic manipulator configured for robotic positioning of a surgical instrument in a body cavity. Other robotic system components may include a second robotic manipulator and/or a patient bed. A camera is positioned to generate an image of a portion of the room, including the robotic manipulator, or an instrument carried by the robotic manipulator as it is moved within the body. Image processing is used to detect the robotic system components on the image captured by the camera, or the tracked portion of the instrument. Once the robotic components or tracked portion of the instruments are detected in the image, the relative positions of the bases within the room may be determined.