Phantom Force: Injecting Adversarial Tactile Perceptions into Embodied Intelligence via EMI

📅 2026-05-13
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
This study addresses a critical security gap in embodied intelligence systems by revealing the vulnerability of Hall-effect-based fingertip tactile sensors to electromagnetic interference (EMI). While prior work has largely overlooked the security of tactile perception channels, this paper demonstrates for the first time that adversarial attacks can be launched against such systems by remotely injecting modulated EMI signals, thereby inducing false force perceptions without physical contact. Experimental results show that the proposed method can amplify perceived force magnitude by over ninefold and shift its direction by up to 65 degrees, effectively degrading the performance of learning-based tactile classification models and significantly disrupting robotic manipulation tasks. This work thus extends adversarial attacks beyond the visual modality, exposing a new attack surface in multimodal embodied AI systems.
📝 Abstract
Embodied intelligent robots rely on tactile sensors to interact with the physical world safely. While the security of visual perception systems has been studied (e.g., adversarial samples), the integrity of the tactile sensory channel remains unexplored. This work explores a vulnerability in Hall-effect fingertip sensors, showing their susceptibility to intentional Electromagnetic Interference (EMI). We demonstrate that a targeted signal injection can induce strong ``phantom forces'', amplifying perceived force magnitude by over \textbf{9$\times$} and deviating the inferred force direction by \textbf{65$^\circ$}. Such perturbations can paralyze learning-based tactile classification models, seriously affecting robot movement. An attacker could exploit this vulnerability to coerce a robot hand into crushing fragile objects or dropping dangerous payloads.
Problem

Research questions and friction points this paper is trying to address.

tactile perception
adversarial attack
electromagnetic interference
embodied intelligence
Hall-effect sensor
Innovation

Methods, ideas, or system contributions that make the work stand out.

adversarial tactile perception
electromagnetic interference (EMI)
Hall-effect tactile sensor
embodied intelligence
phantom force
💼 Related Jobs
No related jobs found.
Z
Zirui Kong
The Hong Kong Polytechnic University
Y
Youqian Zhang
The Hong Kong Polytechnic University
Sze Yiu Chau
Sze Yiu Chau
The Chinese University of Hong Kong
Network SecuritySoftware SecurityApplied Cryptography