Research on the Industrial Robot Grasping Method Based on Multisensor Data Fusion and Binocular Vision

Area of Science:

• Robotics
• Computer Vision
• Automation Systems

Background:

• Traditional industrial robots rely on preprogramming, limiting production line flexibility.
• Flexible manufacturing demands adaptable material handling solutions.
• Existing systems struggle to meet dynamic production requirements.

Purpose of the Study:

• To develop an adaptive material handling system for industrial robots.
• To integrate computer binocular vision with a five-axis robot for enhanced flexibility.
• To improve the adaptability of production lines in material handling.

Main Methods:

• Implemented a binocular vision system with a five-axis industrial robot.
• Developed a high-precision binocular camera calibration method.
• Analyzed robot kinematics for precise target positioning and developed gripper schemes.
• Established Ethernet communication between robot systems and designed control software.

Main Results:

• Achieved accurate target positioning and grasping through integrated binocular vision.
• Demonstrated enhanced flexibility in material handling operations.
• Successfully fused multisensor data with binocular vision for advanced control.
• Validated two control software schemes for visual control.

Conclusions:

• The proposed binocular vision system significantly improves industrial robot flexibility.
• This approach addresses the limitations of traditional preprogrammed robots.
• The integrated system offers a viable solution for flexible material handling in manufacturing.