Contact Surface

Contact surface research focuses on understanding and accurately modeling the interaction between objects at their points of contact, aiming to improve robotic manipulation, medical imaging analysis, and other applications requiring precise geometric and physical characterization. Current research employs diverse approaches, including neural networks (e.g., DeepSDF) for geometry representation, particle filters for dynamic estimation, and contrastive learning for enhancing the resolution of tactile sensor data. These advancements enable more accurate estimations of contact dynamics, force distributions, and friction coefficients, leading to improved robotic dexterity, more precise medical diagnoses (e.g., tumor-organ contact area estimation), and enhanced understanding of complex interactions in various fields.