A vision-guided robot manipulator for surgical instrument singulation in a cluttered environment

The logistics of counting, sorting, sterilizing, and transporting surgical instruments is labor and capital intensive. Furthermore, infection due to improper sterilization is a critical safety hazard. To address these problems, we have developed a unique robotic manipulation system that is capable of accurately singulating surgical instruments in a cluttered environment. Our solution is comprised of two parts. First, we use a single-view vision algorithm for identifying surgical instruments from a pile and estimating their poses. Occlusion reasoning is performed to determine the next instrument to grip using a contrast invariant feature descriptor. Second, we design a compliant electromagnetic gripper that is capable of picking up the identified surgical instrument based on its estimated pose. We validate our solution through instrument singulation experiments demonstrating identification, localization accuracy, and robustness of occlusion reasoning as well as the flexibility of the electromagnetic gripper.