Skip to content
Preprint

AuthGuard-R: Safety-Compliant Mission Hijacking and Dual-Gate Defense for LLM-Controlled Robots

Sep 2026 · 1 citation · 21 references
Computer Science

TL;DR

MissionPAIR, an adaptive attack framework that searches for executable plans that pass a safety gate while violating an authenticated mission, and AuthGuard-R, a deterministic authorization layer that binds every executable action to a signed mission, robot identity, object and region scope, current state, time, and input provenance are proposed.

Abstract

Large language models are increasingly used as high-level planners for mobile robots, robot manipulators, and autonomous vehicles. Recent studies show that these systems can be influenced through malicious text, speech, visual instructions, retrieved documents, and poisoned sensory context. Most defenses ask whether a proposed action is physically safe. This paper studies a different problem: an action may be physically safe and still violate the mission authorized by the user. An attacker may redirect a delivery robot, replace an approved object, extend a robot's operating region, activate an unnecessary sensor, or delay a mission without creating an immediate physical hazard. We call this attack \emph{safety-compliant mission hijacking}. We propose MissionPAIR, an adaptive attack framework that searches for executable plans that pass a safety gate while violating an authenticated mission. We also propose AuthGuard-R, a deterministic authorization layer that binds every executable action to a signed mission, robot identity, object and region scope, current state, time, and input provenance. AuthGuard-R operates with an independent safety gate, giving a dual-gate architecture. We formalize mission policies over robot traces, define security games, and prove authorization soundness, mission non-escalation, replay resistance, robot binding, provenance separation, threshold-approval security, audit-log tamper evidence, and trace-level composition. We report a preliminary cross-model evaluation with Claude Haiku~4.5 and the open-source Qwen2.5~7B planner. Across 240 live attack trials, the planners followed an injected mission deviation in 109 trials; AuthGuard-R rejected all 109 resulting unauthorized actions. A separate hand-constructed suite of eleven protocol- and policy-level attacks was also blocked completely.

View source

Similar papers

#artificial intelligence Preprint Open access Sep 2026

Silent Sabotage: Internal State Triggered Backdoor Attacks on LLM-Powered Robotic Systems

It is demonstrated that an attacker can embed a stealthy backdoor into an LLM-based robot controller by manipulating its instructions, triggered not by an external cue, but by a specific, rare sequence of the robot's own past actions.

Doniyorkhon Obidov, Shivayogi Akki, Cheng-Qiu Tan et al. · 2 citations
#artificial intelligence Conference Open access Sep 2026

Autonomy in Check: Governor-Mediated Adaptive Security at the Edge

This paper proposes a split-control architecture in which an untrusted planner emits typed security intents, a deterministic governor checks each intent against safety, resource, temporal-stability, and proportionality invariants, and only admitted actions are bound to signed receipts and compiled into pre-installed eB...

Ijaz Ahmad, Ijaz Ahmad, Flavio Esposito et al. · 0 citations
Preprint Oct 2026

Mission-Aware Attestation Envelopes for Time-Critical Autonomous Action: A Hardware-in-the-Loop V2I Study

An autonomous system that asks for a privileged physical action is usually gated on integrity evidence: a platform proves what it is running, and the request is granted or refused on that basis. Such a gate is normally treated as a predicate, yet the evidence behind it has an age, the decision that consumes it has a la...

D. Nikou, Nikolaos Kekatos, Sophia G. Petridou et al. · 0 citations
Preprint Oct 2026

Refuse the Harm, Complete the Task via Graph-of-Rules Planning for LLM-Enabled Robots

Large Language Models (LLMs) are increasingly used in embodied AI for robotic manipulation and planning, but their deployment in physical systems exposes vulnerabilities such as jailbreak attacks that induce unsafe behavior. Existing inference-time defenses are mainly evaluated on short instructions containing a single...

Ali Taheri, Alireza Taban, Qi-Zhou Wang et al. · 0 citations
Open access Oct 2026

IIoT-RoboAuth: PUF-Based Command-Bound Authentication for Industrial Robotic Control Networks

Industrial robotic controllers require identity assurance and command-level integrity within bounded control intervals. This study redesigns a fog-assisted three-factor scheme as IIoT-RoboAuth, a PUF-rooted protocol in which the session key is derived from the three participant nonces, an ephemeral P-256 secret, the po...

H. Byeon · 0 citations
Conference Sep 2026

RAPTOR: Role-Based Access Policy Tooling for Operating Robots

Secure ROS 2 (SROS2) provides cryptographic authentication and access control for ROS 2 deployments, but requires administrators to manually author per-node permission XML with no support for roles, inheritance, or visual audit. This flat, per-enclave model is systematically error-prone: a static scan of 124 public SRO...

K. T. Selvam, Shamik Sural · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.