Vision-Based Grasping System

A flexible tool based on visual markers able to guide the robotic arm to perform pick and place tasks.

This tool addresses the pregrasping planning problem, whose goal is to find a configuration that determines how and where the robotic arm must position and orientate itself, according to a distribution criterion (avoiding collisions with other agents). Artificial fiducial markers are used as reference pose and then the grasping action is performed. The use of a camera in a robot control loop can be performed with two types of architecture: eye-in-hand system (when rigidly mounted on the robot end-effector); eye- to-hand system(when it observes the robot within its work space).

- Artificial fiducial markers have high detection and low false-positive and false-negative rates.
- The use of fiducial markers is suitable when object recognition or pose determination is needed with high reliability and when the environment can be modified to affix them


INESC P&D Brasil