Why Look at It at All?: Vision-Free Multifingered Blind Grasping Using Uniaxial Fingertip Force Sensing

2026-02-07

AI Summary

This work addresses the challenge of achieving low-cost, highly reliable multi-finger blind grasping without visual input and with minimal tactile sensing. The authors propose a grasping strategy that relies solely on uniaxial fingertip force feedback and joint proprioception. Leveraging a teacher–student framework, they distill a reinforcement learning–based teacher policy trained in simulation into a deployment-friendly Transformer-based student policy. This study demonstrates for the first time that uniaxial force sensing alone is sufficient to support high-success-rate multi-finger blind grasping, achieving a 98.3% success rate across 18 real-world objects—including out-of-distribution samples—while significantly simplifying the perceptual system without compromising generalization or robustness.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

Grasping under limited sensing remains a fundamental challenge for real-world robotic manipulation, as vision and high-resolution tactile sensors often introduce cost, fragility, and integration complexity. This work demonstrates that reliable multifingered grasping can be achieved under extremely minimal sensing by relying solely on uniaxial fingertip force feedback and joint proprioception, without vision or multi-axis/tactile sensing. To enable such blind grasping, we employ an efficient teacher-student training pipeline in which a reinforcement-learned teacher exploits privileged simulation-only observations to generate demonstrations for distilling a transformer-based student policy operating under partial observation. The student policy is trained to act using only sensing modalities available at real-world deployment. We validate the proposed approach on real hardware across 18 objects, including both in-distribution and out-of-distribution cases, achieving a 98.3~$\%$ overall grasp success rate. These results demonstrate strong robustness and generalization beyond the simulation training distribution, while significantly reducing sensing requirements for real-world grasping systems.

Problem

• blind grasping
• minimal sensing
• multifingered manipulation
• force sensing
• vision-free

Innovation

• blind grasping
• uniaxial force sensing
• teacher-student policy distillation
• transformer-based policy
• vision-free manipulation