Explore algorithmic motion planning for minimally-invasive robotic surgery in this comprehensive lecture from the Robotics Research Colloquium. Delve into the potential of steerable needles and concentric-tube robots to revolutionize common medical procedures, reducing patient recovery time and scarring. Examine the challenges of manual control and the need for automatic planning methods with provable guarantees. Learn about efficient planning capabilities for medical robots, combining techniques from computational geometry, graph theory, and machine learning. Discover how to measure distances between paths, follow reference paths, optimize designs, and minimize shearing in robotic surgeries. Investigate planning for steerable needles and site inspection for CRISP robots, including graph inspection planning challenges and implementation details. Gain insights from Oren Salzman, Assistant Professor at the Technion - Israel Institute of Technology, as he shares his research on addressing computational challenges in robot motion planning.
Overview
Syllabus
Intro
Medical procedures
Why medical robots? (cont.)
Autonomous planning for medical robots
Motivation & Problem definition
Measuring distances between paths
Following reference paths-Take (1)
Domain-Specific challenges following reference
Key challenge-obstacles along the reference path
Following Paths in Task Space
Following reference paths-Extending the optimization
Design or setup optimization
Design optimization via global optimization
Existing optimization criteria
Pop quiz
Planning to minimize shearing
Evaluation
What types of guarantees do planners offer?
Planning for steerable needles
Site inspecting for CRISP robots
Algorithmic approach
Graph inspection planning-challenges
(Optimal) Graph inspection planning
Using approadmate dominance
Properties and implementation detais
Referenced papers
Taught by
Paul G. Allen School
