Development of an Artificial Vision for a Parallel Manipulator Using Machine-to-Machine Technologies

Area of Science:

• Robotics
• Computer Vision
• Machine-to-Machine Communication

Background:

• Flexible delta robot manipulators require advanced vision systems for precise real-time operation.
• Integrating machine-to-machine (M2M) communication can enhance device interaction and system performance.
• Optimizing real-time device interaction is crucial for manufacturing processes like assembly and sorting.

Purpose of the Study:

• To develop and integrate an artificial vision system with M2M communication for a flexible delta robot manipulator.
• To improve the speed and overall performance of the robotic system through enhanced real-time interaction.
• To enable accurate real-time detection, recognition, and localization of targets within limited spaces for manufacturing tasks.

Main Methods:

• Utilized RGB images as input for the MASK-R-CNN algorithm for object detection and recognition.
• Adapted MASK-R-CNN output data for the delta robot control system, considering its specific characteristics.
• Integrated M2M communication to facilitate rapid robot arm responses to environmental changes.

Main Results:

• The integrated system demonstrated high accuracy in real-time target detection and recognition.
• The M2M communication enabled the robot arm to react swiftly to dynamic changes in object position.
• The system proved effective for manufacturing processes such as sorting and packing, even with moving objects.

Conclusions:

• The combination of an artificial vision system and M2M communication significantly enhances the capabilities of flexible delta robots.
• This integration optimizes real-time device interaction, leading to increased speed and improved performance in manufacturing.
• The developed system shows high reliability and potential for real-world applications in automated assembly and sorting.