Robotics Researchers Develop Deep Learning-Based Grasping Model for Cluttered Fasteners

Strategic Insights -

Researchers from Zhengzhou University of Aeronautics have made significant advancements in robotics with the development of a deep learning-based cascaded grasping model for cluttered fasteners. This breakthrough addresses the challenges of grasping cluttered fasteners and the high cost of manual annotation. The model uses an improved YOLO v8n to detect graspable fasteners and an enhanced Generative Residual Convolutional Neural Network (GRCNN) to estimate the optimal grasping pose. The synthetic data generation pipeline leverages physics-based simulation with domain randomization to create synthetic cluttered fastener datasets, effectively reducing the reliance on manual annotation.

Key Takeaways:

  • The deep learning-based cascaded grasping model achieves an average grasping success rate of 92% across different fastener types and varying levels of clutter.
  • The model uses an improved YOLO v8n to detect graspable fasteners and an enhanced Generative Residual Convolutional Neural Network (GRCNN) to estimate the optimal grasping pose.
  • The synthetic data generation pipeline leverages physics-based simulation with domain randomization to create synthetic cluttered fastener datasets.
  • The research was funded by the National Natural Science Foundation of China (NSFC), General Projects of Humanities and Social Sciences Research of Ministry of Education, Henan Provincial Science and Technology Research Project, and Key Scientific Research Projects of Colleges and Universities in Henan Province.
  • The research has been peer-reviewed and published in The International Journal of Advanced Manufacturing Technology.

Statistics:

  • 92% average grasping success rate of the proposed cascaded grasping model across different fastener types and varying levels of clutter.
  • 50.3% increase in detection accuracy and robustness of the proposed model compared to existing methods.
  • The research was funded by four organizations with a total funding amount not specifically mentioned.

Sources:

  • NewsRx LLC, "New Robotics Findings Has Been Reported by Investigators at Zhengzhou University of Aeronautics (Robot Arm Grasping for Cluttered Fasteners Based On Deep Learning With Synthetic Data Augmentation)", Information Technology Newsweekly, November 4, 2025, p 503.
  • The International Journal of Advanced Manufacturing Technology, "Robot Arm Grasping for Cluttered Fasteners Based On Deep Learning With Synthetic Data Augmentation", 2025.
  • Zhengzhou University of Aeronautics, School of Management Engineering, China.
  • National Natural Science Foundation of China (NSFC).
  • General Projects of Humanities and Social Sciences Research of Ministry of Education, People's Republic of China.
  • Henan Provincial Science and Technology Research Project.
  • Key Scientific Research Projects of Colleges and Universities in Henan Province.