Adaptive Kinematic Modeling for Improved Hand Posture Estimates Using a Haptic Glove

TL;DR

This work has added extra sensors to the commercially available Dexmo haptic glove by Dexta Robotics and applied kinematic models of the haptic glove and the user's hand to improve the accuracy of hand-posture measurements.

Abstract

Most commercially available haptic gloves compromise the accuracy of hand-posture measurements in favor of a simpler design with fewer sensors. While inaccurate posture data is often sufficient for the task at hand in biomedical settings such as VR-therapy-aided rehabilitation, measurements should be as precise as possible to digitally recreate hand postures as accurately as possible. With these applications in mind, we have added extra sensors to the commercially available Dexmo haptic glove by Dexta Robotics and applied kinematic models of the haptic glove and the user's hand to improve the accuracy of hand-posture measurements. In this work, we describe the augmentations and the kinematic modeling approach. Additionally, we present and discuss an evaluation of hand posture measurements as a proof of concept.

Figures (8)

• Figure 1: Data flow diagram of our model-based approach. The haptic gloves data is converted to fingertip positions by applying Forward Kinematics derived from a model of the glove. Based on a model of the user's hand, the application of Inverse Kinematics yields a valid joint configuration.
• Figure 2: Concept of additionally measured angles for each finger. B denotes the bend sensor of Dexmo, which measures the flexion angle $\alpha$. F and T denote the newly added sensors to measure angles $\beta$ and $\gamma$ to compute the fingertip's position. Inverse Kinematics then solves for the finger-joint angles $\sigma$, $\delta$, and $\phi$. 1. and 2. display two distinct configurations with different configurations of DIP and PIP, but the same values for angles $\alpha$ and $\beta$, as an example of the under-determination problem, when only measuring $\alpha$.
• Figure 3: Top: Labeled overview of the augmentation of Dexmo. Bottom: Augmented Dexmo in action.
• Figure 4: Schematic representation of the Augmented Dexmo model. Proximal, we used the original device Gu2016DAIDexmoLink and the original model DexmoPatent (gray joints Rotate, Split and Bend). Distal, we replaced the joints with our own augmented hardware (red joints F, T).
• Figure 5: Dorsal, schematic representation of bones and joints in the right hand in extracts based on Nanayakkara2017TRONierop2008ANHChen2013CSF and our simplified hand model.