Vision-based omnidirectional indoor robots for autonomous navigation and localization in manufacturing industry

Area of Science:

• Robotics
• Artificial Intelligence
• Manufacturing Automation

Background:

• Traditional material transport in factories faces limitations in flexibility and efficiency.
• Integrating new robotic systems often requires significant facility modifications and process redefinitions.
• Automated Guided Vehicles (AGVs) offer potential but require sophisticated navigation and environmental interaction capabilities.

Purpose of the Study:

• To develop and present a new generation of omnidirectional AGVs (omniAGVs) for autonomous material transport within manufacturing facilities.
• To enable intelligent navigation and interaction with the environment, including people and other entities.
• To ensure seamless integration into existing factory settings with minimal disruption to current infrastructure and logistics.

Main Methods:

• Development and integration of advanced vision-based systems and cognitive robotics methods.
• Implementation of a localization module using visual and wheel odometry for precise robot pose estimation.
• Integration of an obstacle avoidance module with object recognition for adaptive navigation.
• Inclusion of a tag detection module to assist in cart picking and global localization.

Main Results:

• The proposed omniAGV demonstrates effective autonomous navigation within an industrial environment.
• Vision and perception modules successfully support robot navigation, localization, and obstacle avoidance.
• The system facilitates intelligent interaction with the dynamic factory environment.
• Extensive results validate the AGV's capability and effectiveness in real-world industrial settings.

Conclusions:

• The smart integration of vision and perception is critical for the successful deployment of AGVs in industrial contexts.
• The developed omniAGV offers a viable solution for enhancing material transport efficiency and flexibility in manufacturing.
• The system's ability to navigate autonomously and interact intelligently ensures minimal disruption to existing factory operations.