arXiv:2511.14755v1 Announce Type: cross 
Abstract: As perception-based controllers for autonomous systems become increasingly popular in the real world, it is important that we can formally verify their safety and performance despite perceptual uncertainty. Unfortunately, the verification of such systems remains challenging, largely due to the complexity of the controllers, which are often nonlinear, nonconvex, learning-based, and/or black-box. Prior works propose verification algorithms that are based on approximate reachability methods, but they often restrict the class of controllers and systems that can be handled or result in overly conservative analyses. Hamilton-Jacobi (HJ) reachability analysis is a popular formal verification tool for general nonlinear systems that can compute optimal reachable sets under worst-case system uncertainties; however, its application to perception-based systems is currently underexplored. In this work, we propose RoVer-CoRe, a framework for the Robust Verification of Controllers via HJ Reachability. To the best of our knowledge, RoVer-CoRe is the first HJ reachability-based framework for the verification of perception-based systems under perceptual uncertainty. Our key insight is to concatenate the system controller, observation function, and the state estimation modules to obtain an equivalent closed-loop system that is readily compatible with existing reachability frameworks. Within RoVer-CoRe, we propose novel methods for formal safety verification and robust controller design. We demonstrate the efficacy of the framework in case studies involving aircraft taxiing and NN-based rover navigation. Code is available at the link in the footnote.

يتناول المقال تطوير RoVer-CoRe، وهو إطار مصمم لتحسين التحقق من وحدات التحكم المعتمدة على الإدراك لأنظمة مستقلة في ظل عدم اليقين في الحالة. مع تزايد شعبية هذه الوحدات في التطبيقات الواقعية، يصبح ضمان سلامتها وأدائها وسط عدم اليقين الإدراكي أمرًا بالغ الأهمية. غالبًا ما تكون طرق التحقق التقليدية غير كافية بسبب قيودها في التعامل مع الأنظمة المعقدة وغير الخطية. يوفر تحليل الوصولية لهاملتون-جاكوب نهجًا واعدًا لمعالجة هذه التحديات.

El artículo aborda el desarrollo de RoVer-CoRe, un marco diseñado para mejorar la verificación de controladores basados en la percepción para sistemas autónomos bajo incertidumbre de estado. A medida que estos controladores ganan popularidad en aplicaciones del mundo real, garantizar su seguridad y rendimiento ante la incertidumbre perceptual se vuelve crucial. Los métodos de verificación tradicionales a menudo son insuficientes debido a sus limitaciones para manejar sistemas complejos y no lineales. El análisis de alcanzabilidad de Hamilton-Jacobi ofrece un enfoque prometedor para abordar est…

L'article traite du développement de RoVer-CoRe, un cadre conçu pour améliorer la vérification des contrôleurs basés sur la perception pour les systèmes autonomes en cas d'incertitude d'état. Alors que ces contrôleurs gagnent en popularité dans les applications réelles, garantir leur sécurité et leur performance face à l'incertitude perceptuelle devient crucial. Les méthodes de vérification traditionnelles sont souvent insuffisantes en raison de leurs limitations à traiter des systèmes complexes et non linéaires. L'analyse de portée de Hamilton-Jacobi offre une approche prometteuse pour releve…

The article discusses the development of RoVer-CoRe, a framework designed to enhance the verification of perception-based controllers for autonomous systems under state uncertainty. As these controllers gain traction in real-world applications, ensuring their safety and performance amidst perceptual uncertainty becomes crucial. Traditional verification methods often fall short due to their limitations in handling complex, nonlinear systems. The Hamilton-Jacobi reachability analysis offers a promising approach to address these challenges.

Robust Verification of Controllers under State Uncertainty via Hamilton-Jacobi Reachability Analysis

arXiv:2511.18606v1 Announce Type: cross 
Abstract: Latent safety filters extend Hamilton-Jacobi (HJ) reachability to operate on latent state representations and dynamics learned directly from high-dimensional observations, enabling safe visuomotor control under hard-to-model constraints. However, existing methods implement "least-restrictive" filtering that discretely switch between nominal and safety policies, potentially undermining the task performance that makes modern visuomotor policies valuable. While reachability value functions can, in principle, be adapted to be control barrier functions (CBFs) for smooth optimization-based filtering, we theoretically and empirically show that current latent-space learning methods produce fundamentally incompatible value functions. We identify two sources of incompatibility: First, in HJ reachability, failures are encoded via a "margin function" in latent space, whose sign indicates whether or not a latent is in the constraint set. However, representing the margin function as a classifier yields saturated value functions that exhibit discontinuous jumps. We prove that the value function's Lipschitz constant scales linearly with the margin function's Lipschitz constant, revealing that smooth CBFs require smooth margins. Second, reinforcement learning (RL) approximations trained solely on safety policy data yield inaccurate value estimates for nominal policy actions, precisely where CBF filtering needs them. We propose the LatentCBF, which addresses both challenges through gradient penalties that lead to smooth margin functions without additional labeling, and a value-training procedure that mixes data from both nominal and safety policy distributions. Experiments on simulated benchmarks and hardware with a vision-based manipulation policy demonstrate that LatentCBF enables smooth safety filtering while doubling the task-completion rate over prior switching methods.

دراسة حديثة تقدم نهجًا مبتكرًا لفلاتر الأمان الكامنة التي تعزز من إمكانية الوصول وفقًا لأسلوب هاملتون-جاكوب، مما يمكّن من التحكم البصري الحركي بشكل آمن تحت قيود معقدة. تبرز الأبحاث القيود المفروضة على الأساليب الحالية التي تعتمد على التحول بين السياسات بشكل متقطع، مما قد يهدد الأداء في البيئات عالية الأبعاد.

Un estudio reciente presenta un enfoque novedoso para los filtros de seguridad latentes que mejoran la alcanzabilidad de Hamilton-Jacobi, permitiendo un control visuomotor seguro bajo restricciones complejas. La investigación destaca las limitaciones de los métodos actuales que dependen de un cambio discreto de políticas, lo que puede comprometer el rendimiento en entornos de alta dimensión.

Une étude récente présente une approche novatrice des filtres de sécurité latents qui améliorent la portée de Hamilton-Jacobi, permettant un contrôle visuomoteur sûr sous des contraintes complexes. La recherche met en évidence les limites des méthodes actuelles qui reposent sur un changement de politique discret, ce qui peut compromettre les performances dans des environnements de haute dimension.

A recent study introduces a novel approach to latent safety filters that enhance Hamilton-Jacobi reachability, enabling safe visuomotor control under complex constraints. The research highlights the limitations of current methods that rely on discrete policy switching, which may compromise performance in high-dimensional environments.

How to Train Your Latent Control Barrier Function: Smooth Safety Filtering Under Hard-to-Model Constraints

arXiv:2511.18702v1 Announce Type: cross 
Abstract: General Visual Inspection is a manual inspection process regularly used to detect and localise obvious damage on the exterior of commercial aircraft. There has been increasing demand to perform this process at the boarding gate to minimise the downtime of the aircraft and automating this process is desired to reduce the reliance on human labour. Automating this typically requires estimating a camera's pose with respect to the aircraft for initialisation but most existing localisation methods require infrastructure, which is very challenging in uncontrolled outdoor environments and within the limited turnover time (approximately 2 hours) on an airport tarmac. Additionally, many airlines and airports do not allow contact with the aircraft's surface or using UAVs for inspection between flights, and restrict access to commercial aircraft. Hence, this paper proposes an on-site method that is infrastructure-free and easy to deploy for estimating a pan-tilt-zoom camera's pose and localising scan images. This method initialises using the same pan-tilt-zoom camera used for the inspection task by utilising a Deep Convolutional Neural Network fine-tuned on only synthetic images to predict its own pose. We apply domain randomisation to generate the dataset for fine-tuning the network and modify its loss function by leveraging aircraft geometry to improve accuracy. We also propose a workflow for initialisation, scan path planning, and precise localisation of images captured from a pan-tilt-zoom camera. We evaluate and demonstrate our approach through experiments with real aircraft, achieving root-mean-square camera pose estimation errors of less than 0.24 m and 2 degrees for all real scenes.

تقترح دراسة جديدة طريقة تعتمد على الشبكات العصبية التلافيفية (CNN) لتقدير وضع الكاميرا وتحديد مواقع الصور الممسوحة ضوئيًا لفحص الطائرات بصريًا، مما يعالج التحديات التي تواجه الفحوصات اليدوية التي غالبًا ما تكون محدودة بالوقت والظروف البيئية. يهدف هذا النهج الخالي من البنية التحتية إلى تحسين الكفاءة في اكتشاف الأضرار على الطائرات التجارية عند بوابات الصعود.

Un nuevo estudio propone un método basado en CNN para estimar la pose de la cámara y localizar imágenes escaneadas para la inspección visual de aeronaves, abordando los desafíos de las inspecciones manuales que a menudo están limitadas por el tiempo y las condiciones ambientales. Este enfoque sin infraestructura busca mejorar la eficiencia en la detección de daños en aeronaves comerciales en las puertas de embarque.

Une nouvelle étude propose une méthode basée sur le CNN pour estimer la pose de la caméra et localiser les images scannées pour l'inspection visuelle des avions, répondant aux défis des inspections manuelles souvent limitées par le temps et les conditions environnementales. Cette approche sans infrastructure vise à améliorer l'efficacité dans la détection des dommages sur les avions commerciaux aux portes d'embarquement.

A new study proposes a CNN-based method for estimating camera pose and localizing scan images for visual inspection of aircraft, addressing the challenges of manual inspections that are often limited by time and environmental conditions. This infrastructure-free approach aims to enhance efficiency in detecting damage on commercial aircraft at boarding gates.

Robust Verification of Controllers under State Uncertainty via Hamilton-Jacobi Reachability Analysis

Was this article worth reading? Share it

AI Contract Review

ISMS Copilot

Octofy