Mathieu Grossard

Object-agnostic grasp planning achieves 86.93% success on diverse objects without the need for object-specific training data.

Achieving zero-shot generalization for dexterous grasp synthesis, CoToGrasp outperforms traditional methods by leveraging contact topology without requiring annotated datasets.

By fusing fast tactile sensing with quadratic programming for internal force control, this work achieves impressively rapid (20ms) slip detection and reactive stabilization in multifingered robotic grasping.