Zhiyuan Robotics Advances Robot Motion Control with New Patent

Shanghai Zhiyuan Xinchuang Technology Co., Ltd., an affiliated company of Zhiyuan Robotics, has successfully published a patent titled "Training Method and Program Product for a Robot's Motion Control Model" through the Zhiqicha APP platform. This development marks a significant step forward in robotic motion control technology.

Technical Innovation in Motion Control

The patented method introduces an advanced approach to refining robot movement by analyzing the interaction between robotic limbs and surfaces. Key components include:

• Actual Contact Sequence Analysis: Captures real-world contact patterns during foot actions generated by the motion control model
• Expected Contact Sequence Generation: Creates theoretical contact patterns based on phase variables and gait mapping relationships
• Swing Excitation Value Calculation: Determines adjustment parameters by comparing expected vs. actual contact sequences

How the System Works

The process involves continuous iterative improvements:

1. The system monitors the robot's actual surface contacts during movement
2. Compares these against mathematically modeled expected contacts
3. Generates swing excitation values inversely proportional to sequence differences
4. Automatically adjusts motion control parameters to minimize discrepancies

This closed-loop system enables robots to progressively refine their movements, particularly valuable for applications requiring precise foot placement such as search-and-rescue operations or industrial automation.

Industry Implications

The technology addresses a critical challenge in robotics - adaptive movement control in dynamic environments. By implementing this learning-based approach, Zhiyuan's solution could significantly improve:

• Stability on uneven terrain
• Energy efficiency through optimized movement patterns
• Adaptability to surface variations without manual reprogramming

Key Points:

• Patent Status: Published application through Chinese intellectual property system
• Technical Focus: Machine learning-enhanced motion control for robotic limbs
• Potential Applications: Industrial automation, service robotics, and emergency response systems
• Innovation Core: Self-adjusting algorithm that reduces need for manual calibration