Data-Driven Modeling of Haptic Properties

Researchers: Arsen Abdulali, Sunghoon Yim, and Seokhee Jeon

Pure data-driven haptic modeling and rendering is one of the emerging techniques in haptics. This approach records signals generated during manual or automated palpation of a target object — e.g., high-frequency vibrations through active stroking of a surface for haptic texture — and uses them in rendering to approximate the responses of the target surface under a given user interaction. It can effectively handle diverse and complex behaviors with less computational load and without prior knowledge of the object or system, making it one of the most relevant approaches for applications requiring high haptic realism. This research targets three haptic properties for data-driven modeling: stiffness, friction, and haptic texture.

Related Publications:

• Sunghoon Yim, Seokhee Jeon, and Seungmoon Choi, "Data-Driven Haptic Modeling and Rendering of Viscoelastic and Frictional Responses of Deformable Objects," IEEE Transactions on Haptics. (Stiffness and friction)
• Sunghoon Yim, Seokhee Jeon, and Seungmoon Choi, "Data-Driven Haptic Modeling and Rendering of Deformable Objects Including Sliding Friction," IEEE World Haptics Conference (WHC), pp. 305-312, 2015. (Stiffness and friction)
• Arsen Abdulali and Seokhee Jeon, "Data-Driven Modeling of Anisotropic Haptic Textures: Data Segmentation and Interpolation," EuroHaptics, pp. 228-239, 2016. (Haptic texture)
• Arsen Abdulali, Waseem Hassan, and Seokhee Jeon, "Stimuli-Magnitude-Adaptive Sample Selection for Data-Driven Haptic Modeling," MDPI Entropy, vol. 18, no. 222, 2016. (Object dynamics)

Abdulali, A., Hassan, W., & Jeon, S. (2016). Stimuli-Magnitude-Adaptive Sample Selection for Data-Driven Haptic Modeling. MDPI Entropy, vol. 18, no. 222. https://doi.org/10.3390/e18060222