Volumetric compensation of accuracy errors in a multi-robot surgical platform

Area of Science:

• Robotics
• Neurosurgery
• Medical Engineering

Background:

• Open skull neurosurgery demands sub-millimeter accuracy for tool tracking and motion compensation.
• Individual robot calibration is insufficient for achieving high precision in multi-robot surgical systems.
• Inaccuracies in multi-robot platforms often stem from the calibration phase.

Purpose of the Study:

• To develop and evaluate a volumetric error compensation method for a multi-robot platform used in neurosurgery.
• To improve the end-to-end static and dynamic accuracy of robotic surgical systems.
• To reduce inaccuracies originating from the calibration of individual robotic arms.

Main Methods:

• A hybrid parallel kinematic machine combined with two KUKA LWR arms forms the multi-robot platform.
• An offline training phase computes a compensation transform for discretized workspace subregions.
• At runtime, compensation motion is applied to robots to achieve accurate targeting on anatomical parts.

Main Results:

• Achieved a median end-to-end static accuracy of 0.75 mm, with 95% of tests below 1 mm.
• Demonstrated a 1:36 reduction factor in positioning errors compared to initial conditions.
• Evaluated accuracy under dynamic conditions with mild oscillatory patterns.

Conclusions:

• Volumetric error compensation effectively enhances the accuracy of multi-robot surgical platforms.
• The developed method significantly improves precision for demanding neuro-surgical tasks.
• This approach offers a viable solution for achieving sub-millimeter accuracy in complex robotic surgeries.