mlsanityData to 14 Oct 2026

SpecGuard: Specification Aware Recovery for Robotic Autonomous Vehicles from Physical Attacks

arXiv 2408.15200 ↗First posted 27 Aug 2024cs.RO cs.CR cs.SYcited by 2

On the map: End-to-end autonomous driving → Adversarial scenarios and safety testing

Abstract (arXiv)

Robotic Autonomous Vehicles (RAVs) rely on their sensors for perception, and follow strict mission specifications (e.g., altitude, speed, and geofence constraints) for safe and timely operations. Physical attacks can corrupt the RAVs' sensors, resulting in mission failures. Recovering RAVs from such attacks demands robust control techniques that maintain compliance with mission specifications even under attacks to ensure the RAV's safety and timely operations. We propose SpecGuard, a technique that complies with mission specifications and performs safe recovery of RAVs. There are two innovations in SpecGuard. First, it introduces an approach to incorporate mission specifications and learn a recovery control policy using Deep Reinforcement Learning (Deep-RL). We design a compliance-based reward structure that reflects the RAV's complex dynamics and enables SpecGuard to satisfy multiple mission specifications simultaneously. Second, SpecGuard incorporates state reconstruction, a technique that minimizes attack induced sensor perturbations. This reconstruction enables effective adversarial training, and optimizing the recovery control policy for robustness under attacks. We evaluate SpecGuard in both virtual and real RAVs, and find that it achieves 92% recovery success rate under attacks on different sensors, without any crashes or stalls. SpecGuard achieves 2X higher recovery success than prior work, and incurs about 15% performance overhead on real RAVs.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Adversarial scenarios and safety testing, one of 9 directions in End-to-end autonomous driving. Builds on 20 of its 81 references that are in the map. First posted on arXiv 27 Aug 2024.

Builds on

20 of 81 references.

References · 20

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2016Concrete Problems in AI Safetycited by 1,642
  3. 2017Sim-to-Real Transfer of Robotic Control with Dynamics Randomizationcited by 672
  4. 2017Robust Adversarial Reinforcement Learningcited by 451
  5. 2017Safe Reinforcement Learning via Shieldingcited by 449
  6. 2021Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learningcited by 371
  7. 2022The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Modelscited by 328
  8. 2019End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Taskscited by 289
  9. 2020Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observationscited by 238
  10. 2017Robust Deep Reinforcement Learning with Adversarial Attackscited by 154
  11. 2017Control of a Quadrotor with Reinforcement Learningcited by 92
  12. 2017Vulnerability of Deep Reinforcement Learning to Policy Induction Attackscited by 91
8 more
  1. 2020Adaptive Reward-Poisoning Attacks against Reinforcement Learningcited by 86
  2. 2019Adversarial Objects Against LiDAR-Based Autonomous Driving Systemscited by 63
  3. 2021Autonomous Drone Racing with Deep Reinforcement Learningcited by 60
  4. 2015Robust Online Monitoring of Signal Temporal Logiccited by 39
  5. 2019Learning Visuomotor Policies for Aerial Navigation Using Cross-Modal Representationscited by 19
  6. 2020AirSim Drone Racing Labcited by 18
  7. 2018Injected and Delivered: Fabricating Implicit Control over Actuation Systems by Spoofing Inertial Sensorscited by 10
  8. 2022Diagnosis-guided Attack Recovery for Securing Robotic Vehicles from Sensor Deception Attackscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 1.

2 papers.

  1. 2025ARMOR: Robust Reinforcement Learning-based Control for UAVs under Physical Attackscited by 1
  2. 2026ASGARD: Action-Space Guard for UAV Resilience via Reinforcement Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022Security of Deep Reinforcement Learning for Autonomous Driving: A Survey
  2. 2025Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial Attacks
  3. 2026CRRL: A Causality-Based Reinforcement Learning Framework for Autonomous System Recovery
  4. 2026Demonstration-Guided Observation Attacks on Black-Box Safe Reinforcement Learning Controllers for Robotic Systems
  5. 2026RoAd-RL: A Unified Library and Benchmark for Robust Adversarial Reinforcement Learning
  6. 2024Realizable Continuous-Space Shields for Safe Reinforcement Learning
  7. 2025Evaluating Robustness of Deep Reinforcement Learning for Autonomous Surface Vehicle Control in Field Tests
  8. 2024SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning