The Integration of the Image Sensor with a 3-DOF Pneumatic Parallel Manipulator.

Area of Science:

• Robotics and Automation
• Computer Vision
• Mechatronics

Background:

• Automated pick-and-place systems require precise object recognition and manipulator control.
• Integrating image sensing with robotic manipulators enhances operational capabilities.

Purpose of the Study:

• To develop an automated system for object pick-and-place using a three-axial pneumatic parallel manipulator.
• To integrate an image sensor and employ advanced algorithms for feature recognition and manipulator control.

Main Methods:

• Utilized the Speeded Up Robust Features (SURF) algorithm for object feature definition and matching.
• Employed the Random Sample and Consensus (RANSAC) method for accurate target localization and feature matching.
• Integrated an ASUS Xtion Pro Live depth camera for 3-D coordinate estimation and calibration.
• Implemented hand gesture recognition with skin detection and noise elimination for manipulator input.

Main Results:

• Successfully achieved accurate feature recognition of target objects.
• Demonstrated precise pick-and-place functionality of the parallel manipulator.
• Enabled manipulator pose adjustment based on recognized hand gestures and finger counts.

Conclusions:

• The integrated system effectively performs automated object recognition and manipulation.
• The combination of SURF, RANSAC, and depth sensing enhances robotic pick-and-place accuracy.
• Hand gesture recognition provides an intuitive control interface for the parallel manipulator.