Adaptive Hybrid Vision-Force Controller for Robotic Grippers in Lab Automation
This study introduces a novel controller for manipulating squeeze bottles in laboratory settings.
Robotic automation in laboratories aims to enhance safety and efficiency by minimizing human involvement in hazardous tasks. This paper discusses the challenges of manipulating deformable objects, specifically squeeze bottles, and presents an adaptive hybrid vision-force controller designed for robotic grippers.
The controller utilizes both force and vision feedback to manage the interaction with the squeeze bottles, allowing for effective liquid dispensing while accommodating bottle deformation. Validation of the controller involved testing across various conditions, comparing its performance to a traditional force-only controller.
Results indicated that the hybrid controller significantly improved dispensing accuracy, achieving a mean absolute error of 6.54 mL, compared to 15.92 mL from the pure force controller, while closely mimicking human grip actions recorded through electromyography.