Research on Parameter Compensation Method and Control Strategy of Mobile Robot Dynamics Model Based on Digital Twin.

Area of Science:

• Robotics and Automation
• Artificial Intelligence
• Industrial Safety

Background:

• Inspection robots are crucial for safety in high-risk industries (power, petrochemical, battery factories).
• Limited positional accuracy hinders the broad application of current inspection robots.
• Accurate kinematic models and control systems are needed for reliable navigation on unstructured surfaces.

Purpose of the Study:

• To develop a visualization monitoring and control system framework for intelligent factories.
• To improve the positional accuracy and inspection efficiency of mobile robots.
• To address challenges in virtual-real integration and real-time data management for industrial robots.

Main Methods:

• Utilized a virtual engine and digital twins to create a visualization monitoring and control system framework.
• Developed a neural-network-based compensation technique for nonlinear dynamic model parameters of outdoor mobile robots.
• Implemented and experimented with a physical prototype for online control and monitoring.

Main Results:

• The proposed system demonstrated effective online control and monitoring of outdoor mobile robots with high real-time performance and visualization.
• The precise kinematic model improved robot positional accuracy by 18% during obstacle navigation.
• The visualization system enables comprehensive, multi-method, real-time inspections in hazardous environments.

Conclusions:

• The developed visualization monitoring and control system significantly enhances robot positional accuracy and inspection efficiency.
• The system ensures safe and stable operations in high-risk environments like new energy battery factories.
• This framework provides a foundation for advanced, data-driven robotic inspections in intelligent manufacturing.