TactiGraph: An Asynchronous Graph Neural Network for Contact Angle Prediction Using Neuromorphic Vision-Based Tactile Sensing
Author:
Sajwani Hussain12, Ayyad Abdulla2, Alkendi Yusra3ORCID, Halwani Mohamad2ORCID, Abdulrahman Yusra23, Abusafieh Abdulqader24, Zweiri Yahya23ORCID
Affiliation:
1. UAE National Service & Reserve Authority, Abu Dhabi, United Arab Emirates 2. Advanced Research and Innovation Center (ARIC), Khalifa University, Abu Dhabi 127788, United Arab Emirates 3. Department of Aerospace Engineering, Khalifa University, Abu Dhabi 127788, United Arab Emirates 4. Research and Development, Strata Manufacturing PJSC, Al Ain 86519, United Arab Emirates
Abstract
Vision-based tactile sensors (VBTSs) have become the de facto method for giving robots the ability to obtain tactile feedback from their environment. Unlike other solutions to tactile sensing, VBTSs offer high spatial resolution feedback without compromising on instrumentation costs or incurring additional maintenance expenses. However, conventional cameras used in VBTS have a fixed update rate and output redundant data, leading to computational overhead.In this work, we present a neuromorphic vision-based tactile sensor (N-VBTS) that employs observations from an event-based camera for contact angle prediction. In particular, we design and develop a novel graph neural network, dubbed TactiGraph, that asynchronously operates on graphs constructed from raw N-VBTS streams exploiting their spatiotemporal correlations to perform predictions. Although conventional VBTSs use an internal illumination source, TactiGraph is reported to perform efficiently in both scenarios (with and without an internal illumination source) thus further reducing instrumentation costs. Rigorous experimental results revealed that TactiGraph achieved a mean absolute error of 0.62∘ in predicting the contact angle and was faster and more efficient than both conventional VBTS and other N-VBTS, with lower instrumentation costs. Specifically, N-VBTS requires only 5.5% of the computing time needed by VBTS when both are tested on the same scenario.
Funder
STRATA Manufacturing PJSC
Subject
Electrical and Electronic Engineering,Biochemistry,Instrumentation,Atomic and Molecular Physics, and Optics,Analytical Chemistry
Reference84 articles.
1. Huang, X., Muthusamy, R., Hassan, E., Niu, Z., Seneviratne, L., Gan, D., and Zweiri, Y. (2020). Neuromorphic Vision Based Contact-Level Classification in Robotic Grasping Applications. Sensors, 20. 2. Slip Detection with a Biomimetic Tactile Sensor;James;IEEE Robot. Autom. Lett.,2018 3. Dong, S., Jha, D., Romeres, D., Kim, S., Nikovski, D., and Rodriguez, A. (June, January 30). Tactile-RL for Insertion: Generalization to Objects of Unknown Geometry. Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA), Xi’an, China. 4. Kim, S., and Rodriguez, A. (2022, January 23–27). Active Extrinsic Contact Sensing: Application to General Peg-in-Hole Insertion. Proceedings of the 2022 International Conference on Robotics and Automation (ICRA), Philadelphia, PA, USA. 5. A review on sensory perception for dexterous robotic manipulation;Xia;Int. J. Adv. Robot. Syst.,2022
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
|
|