Surgical Robot Path Mapping for Predictive Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Surgical robots face challenges with slow navigation updates and collisions during procedures due to inadequate prediction and adaptation to user movements in the operating environment.
Innovation Solution
A system that predicts user movements using machine learning models and imaging sensors to dynamically adjust the navigation path of a robotic arm, creating 3D maps with high and low probability zones to avoid collisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional navigation systems are used without prediction capabilities, then the system is simpler and faster to implement, but collisions occur and navigation updates are slow
Solution Approach 1:
The system performs preliminary actions by predicting future positions of the surgical tool and surgeon's hand before collisions occur. The machine learning model forecasts object positions at future time points, allowing the navigation system to proactively adjust the robotic arm's path to avoid predicted collision zones, rather than reactively responding after collisions occur.
Solution Approach 2:
The patent introduces an intermediary prediction layer between the current sensor data and the navigation decisions. The machine learning model acts as a mediator that transforms current positional data into predicted future positions, which then inform the collision avoidance navigation. This intermediary processing layer enables more reliable collision avoidance while managing system complexity through specialized prediction algorithms.
2Reliability
If real-time prediction and dynamic path adjustment are implemented, then collision avoidance improves, but computational requirements and processing time increase
Solution Approach 1:
The system implements periodic action by updating predictions and adjusting navigation paths at regular time intervals or frame rates. Rather than continuously calculating infinite future positions, the machine learning model predicts positions at discrete future time points, and the navigation system adjusts paths periodically. This approach maintains navigation safety while managing processing time through structured, periodic updates.
Solution Approach 2:
The prediction model performs preliminary calculations of future positions before navigation decisions are required. By pre-computing predicted positions at multiple future time points based on current sensor data, the system prepares collision avoidance information in advance, reducing the processing time required during actual navigation adjustments and maintaining real-time performance.
3Adaptability or versatility
If the robotic arm follows a static navigation path, then the path planning is simpler, but it cannot adapt to user movements and collisions occur
Solution Approach 1:
The patent transforms the static navigation path into a dynamic system that continuously adapts to user movements. The machine learning model tracks the surgeon's hand position and predicts future positions, allowing the robotic arm's navigation path to dynamically adjust in real-time. This dynamic adaptation enables the system to respond to user movements while managing complexity through efficient prediction algorithms and selective path updates.
Solution Approach 2:
The system implements feedback by continuously monitoring the surgeon's hand position through sensors and using this information to update predictions and adjust the navigation path. The predicted future positions feed back into the navigation system, which modifies the robotic arm's path accordingly. This closed-loop feedback mechanism enables adaptability to user movements while managing complexity through efficient feedback processing and prediction updates.
Data Source
AI summary
A system according to at least one embodiment of the present disclosure includes a processor; and a memory coupled with the processor and including data stored thereon that, when processed by the processor, enables the processor to: predict, at a first time, a motion of an object during a surgical procedure and at a second time following the first time; and update, based on the predicted motion of the object, a surgical navigation path of a robotic arm.