Visuo-Tactile Object Pose Estimation for a Multi-Finger Robot Hand with Low-Resolution In-Hand Tactile Sensing

2025-03-25

AI Summary

To address the poor robustness of robotic in-hand object 3D pose estimation under severe occlusion and high visual noise, this paper proposes a tightly coupled visual–proprioceptive–tactile factor graph optimization method. Innovatively, sparse binary tactile signals—only one sensor per phalanx—are explicitly modeled and integrated with RGB-D imagery and joint encoder measurements, enabling joint optimization of the object’s full 6-DoF pose under a robust cost function. Compared to vision-only approaches, the method significantly improves pose accuracy and stability in highly occluded scenarios. Implemented in C++ within the GTSAM framework, it achieves a 42% reduction in average pose estimation error across 17 YCB objects in simulation. Real-world experiments demonstrate real-time performance at 13.3 Hz on physical hardware, validating both the efficacy and practicality of the low-cost tactile sensing scheme.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Biometrics, Face, Gesture & Pose
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Accurate 3D pose estimation of grasped objects is an important prerequisite for robots to perform assembly or in-hand manipulation tasks, but object occlusion by the robot's own hand greatly increases the difficulty of this perceptual task. Here, we propose that combining visual information and proprioception with binary, low-resolution tactile contact measurements from across the interior surface of an articulated robotic hand can mitigate this issue. The visuo-tactile object-pose-estimation problem is formulated probabilistically in a factor graph. The pose of the object is optimized to align with the three kinds of measurements using a robust cost function to reduce the influence of visual or tactile outlier readings. The advantages of the proposed approach are first demonstrated in simulation: a custom 15-DoF robot hand with one binary tactile sensor per link grasps 17 YCB objects while observed by an RGB-D camera. This low-resolution in-hand tactile sensing significantly improves object-pose estimates under high occlusion and also high visual noise. We also show these benefits through grasping tests with a preliminary real version of our tactile hand, obtaining reasonable visuo-tactile estimates of object pose at approximately 13.3 Hz on average.

Problem

• Estimating 3D object pose under hand occlusion using low-resolution tactile sensing
• Combining visual, proprioceptive, and binary tactile data for pose estimation
• Improving pose accuracy in high-occlusion and high-noise scenarios

Innovation

• Combines vision, proprioception, and binary tactile sensing
• Uses factor graph for probabilistic pose estimation
• Robust cost function reduces outlier influence