mlsanityData to 14 Oct 2026

Closed-loop Analysis of Vision-based Autonomous Systems: A Case Study

arXiv 2302.04634 ↗First posted 6 Feb 2023cs.CV cs.AI cs.FLcited by 18

Not on the map. Why?

Abstract (arXiv)

Deep neural networks (DNNs) are increasingly used in safety-critical autonomous systems as perception components processing high-dimensional image data. Formal analysis of these systems is particularly challenging due to the complexity of the perception DNNs, the sensors (cameras), and the environment conditions. We present a case study applying formal probabilistic analysis techniques to an experimental autonomous system that guides airplanes on taxiways using a perception DNN. We address the above challenges by replacing the camera and the network with a compact probabilistic abstraction built from the confusion matrices computed for the DNN on a representative image data set. We also show how to leverage local, DNN-specific analyses as run-time guards to increase the safety of the overall system. Our findings are applicable to other autonomous systems that use complex DNNs for perception.

Key facts. Cited by 18 papers in the map (data through 14 Oct 2026). Builds on 21 of its 41 references that are in the map. First posted on arXiv 6 Feb 2023.

Builds on

21 of 41 references.

References · 21

  1. 2017On Calibration of Modern Neural Networkscited by 4,960
  2. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,340
  3. 2015Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)cited by 1,070
  4. 2019A Survey of Deep Learning Techniques for Autonomous Drivingcited by 403
  5. 2018On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Modelscited by 328
  6. 2018Certified Defenses against Adversarial Examplescited by 306
  7. 2016Safety Verification of Deep Neural Networkscited by 263
  8. 2022Safe Control with Learned Certificates: A Survey of Neural Lyapunov, Barrier, and Contraction methodscited by 151
  9. 2021Globally-Robust Neural Networkscited by 108
  10. 2018Verisig: verifying safety properties of hybrid systems with neural network controllerscited by 84
  11. 2020The Probabilistic Model Checker Stormcited by 66
  12. 2019Deep Learning for Large-Scale Traffic-Sign Detection and Recognitioncited by 34
9 more
  1. 2021Verification of Image-based Neural Network Controllers Using Generative Modelscited by 27
  2. 2022Learning Safe, Generalizable Perception-based Hybrid Control with Certificatescited by 24
  3. 2020Fast Geometric Projections for Local Robustness Certificationcited by 21
  4. 2017DeepSafe: A Data-driven Approach for Checking Adversarial Robustness in Neural Networkscited by 20
  5. 2021Verifying Controllers with Convolutional Neural Network-based Perception: A Case for Intelligible, Safe, and Precise Abstractionscited by 19
  6. 2022NNLander-VeriF: A Neural Network Formal Verification Framework for Vision-Based Autonomous Aircraft Landingcited by 19
  7. 2022Discrete-Event Controller Synthesis for Autonomous Systems with Deep-Learning Perception Componentscited by 5
  8. 2019Counterexample-Guided Synthesis of Perception Models and Controlcited by 4
  9. 2022Software Performability Analysis Using Fast Parametric Model Checkingcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 5; 2024: 6; 2025: 5; 2026: 2.

18 papers.

Sort:

By area:

  1. 2024Marabou 2.0: A Versatile Formal Analyzer of Neural Networkscited by 81
  2. 2024Vehicle: Bridging the Embedding Gap in the Verification of Neuro-Symbolic Programscited by 14
  3. 2025Four Principles for Physically Interpretable World Modelscited by 10
  4. 2023Bridging Dimensions: Confident Reachability for High-Dimensional Controllerscited by 5
  5. 2024System Safety Monitoring of Learned Components Using Temporal Metric Forecastingcited by 4
  6. 2025Scenario-based Compositional Verification of Autonomous Systems with Neural Perceptioncited by 3
  7. 2025Conformal Safety Shielding for Imperfect-Perception Agentscited by 3
  8. 2024AssistTaxi: A Comprehensive Dataset for Taxiway Analysis and Autonomous Operationscited by 2
  9. 2025ScenicProver: A Framework for Compositional Probabilistic Verification of Learning-Enabled Systemscited by 2
  10. 2023Assumption Generation for the Verification of Learning-Enabled Autonomous Systemscited by 1
  11. 2023Is Certifying ℓp Robustness Still Worthwhile?cited by 1
  12. 2023Refining Perception Contracts: Case Studies in Vision-based Safe Auto-landingcited by 1
6 more
  1. 2023Automatic Generation of Scenarios for System-level Simulation-based Verification of Autonomous Driving Systemscited by 1
  2. 2024Pre-Trained Vision Models as Perception Backbones for Safety Filters in Autonomous Drivingcited by 1
  3. 2025Formal Control for Uncertain Systems via Contract-Based Probabilistic Surrogates (Extended Version)cited by 1
  4. 2024Using Formal Models, Safety Shields and Certified Control to Validate AI-Based Train Systemscited by 0
  5. 2026Interval POMDP Shielding for Imperfect-Perception Agentscited by 0
  6. 2026Correcting Learning-based Perception for Safetycited by 0