Session: 02-01-01: Product and Process Design 1

Paper Number: 166009

Intelligent Bio-Inspired Robotic Gripper With 3D Object Modeling and Adaptive Force Modulation 

Robotic grippers are essential in manufacturing, logistics, healthcare, agriculture, and space exploration, enabling automated handling of objects with diverse geometries, weights, and material properties. However, conventional grippers struggle with inconsistent force application, slippage, and unintended damage, particularly when handling delicate or irregularly shaped objects. While computer vision and machine learning have been employed to enhance gripping precision, their reliance on extensive training datasets and high computational costs limits their real-time responsiveness in dynamic manufacturing environments. We propose a novel scanner-based robotic gripping system that integrates real-time 3D object modeling with an adaptive gripping mechanism to address these challenges. This system leverages a six-degree-of-freedom (6-DOF) Universal Robot, which performs a 360-degree rotational scan to construct an accurate 3D model of the object of interest. The scanned data is processed using a Python-based algorithm, which analyzes the model and determines the optimal grasping position and orientation for a biomimetic, 3D-printed robotic gripper designed to resemble the human hand. A key innovation in this system is its pre-trained algorithm, which dynamically adjusts the applied grip force based on real-time object weight analysis. This ensures a firm grasp while mitigating the risk of slippage or structural damage. Unlike conventional grippers based on pre-defined grip settings, this adaptive approach allows for the exact handling of objects of varying fragility and structural integrity. The scanner-based gripping system offers several advantages. First, it significantly enhances grasp stability by precisely identifying optimal grip points, reducing the chances of mishandling or object damage. Second, the ability to model objects in real time eliminates the need for extensive pre-training, making the system adaptable to different environments and object types. Third, its lightweight and modular design makes it suitable for industrial automation applications, e-commerce packaging, and handling delicate items in unstructured environments, such as warehouses or medical facilities. Furthermore, integrating 3D scanning with adaptive gripping presents a promising alternative to vision-based gripping techniques. Unlike traditional computer vision systems that require complex image processing pipelines and are susceptible to environmental factors such as lighting conditions or occlusions, the scanner-based approach ensures reliable and precise object modeling in any setting. The adaptability of this system extends its usability to diverse applications, including automated sorting, robotic-assisted surgery, and space missions where precise object manipulation is critical. In conclusion, the proposed scanner-based robotic gripper provides a robust and effective solution for real-time, precision handling of diverse objects. By combining real-time 3D modeling with adaptive grip force control, this system overcomes the limitations of conventional grippers and machine learning-based approaches, making it a transformative innovation for industrial automation and beyond.

Presenting Author: Saquib Shahriar Prairie View A&M University

Presenting Author Biography: Saquib Shahriar is a graduate student in Mechanical Engineering at Prairie View A&M University, with a background in Mechatronics and early industry experience in Compressor Manufacturing.

Authors:

Saquib Shahriar Prairie View A&M University
Wenhua Yang Prairie View A&M University
Chang Duan Prairie View A&M University
Jaejong Park Prairie View A&M University