arXiv subjects

Barak Or

Publications and source records attributed to Barak Or.

At least 19 recordsLinked to original sources

arXiv2026

Can Predicted Dynamics Exist in the Physical World?

Can learned state-action proposals exist in the physical world? To filter infeasible commands before execution, policies are often wrapped in a runtime monitor. However, aggregating diverse diagnostic signals obscures whether a proposal violates dynamic transitions or merely departs from recorded behavior. We formalize this prediction-control interface and prove that the all-pairs displacement term is redundant within a max-aggregated composite. We evaluate these monitors on 700 nominal and 5,250 synthetically perturbed 32-transition PushT windows, monitoring only planar pusher positions and goals. A simple transition-RMSE baseline (AUC 0.982) outperforms a heterogeneous max-aggregated monitor (AUC 0.957). We conclude that physical transition checks must be strictly separated from empirical logs.

cs.RO↗

arXiv2026

Silent Failures in Physical AI: A Literature Review of Runtime Action Authorization for Autonomous Systems

Physical AI systems increasingly map multimodal observations, language instructions, and learned world representations into physically consequential actions. Robotics foundation models, vision-language-action models, and world-model-based autonomous systems can condition decisions that move vehicles, robots, drones, and industrial machines. This transition exposes a safety problem that is not fully captured by conventional AI content moderation or by classical robot safety alone: a black-box model may issue a physically consequential action while appearing confident, plausible, and semantically aligned. The resulting failure can be silent, arising from sensor drift, occlusion, state-estimation error, distribution shift, hallucinated affordances, or invalid physical assumptions before downstream hardware controllers detect a violation. Across embodied foundation models, world models, robotics simulation, embodied safety benchmarks, safe control, runtime assurance, uncertainty estimation, verification, and guardrail evaluation, model capability and safety mechanisms have advanced along largely separate technical tracks. A recurring gap synthesized here is that no single stream surveyed in this review supplies a complete runtime authorization boundary between black-box Physical AI models and physical execution. The resulting analysis develops a bounded problem formulation, a definition of silent physical-action failure, a taxonomy of runtime guardrail functions, and evaluation requirements for comparing guardrails as Physical AI assurance mechanisms.

cs.RO↗

arXiv2026

Quantifying Prior Dominance in RAG Systems

Retrieval-Augmented Generation (RAG) grounds Large Language Models in external knowledge, yet current evaluations rely on discrete heuristics that suffer from ''epistemic blindness'' - failing to distinguish genuine contextual information extraction from parametric memory recall. To address this, we introduce the Normalized Context Utilization (NCU) metric, leveraging continuous token log-probabilities across zero-shot, oracle, and adversarial conditions to strictly quantify contextual information gain. Evaluating architectures ranging from 1.5B to 72B parameters alongside a proprietary commercial API reveals that for strict factual extraction (without Chain-of-Thought reasoning), traditional scaling laws exhibit extreme diminishing returns: highly efficient Small Language Models (SLMs) match or outperform high-capacity architectures. Furthermore, we demonstrate that ``Prior Dominance'' correlates with model scale and proprietary alignments. The evaluated commercial API not only overrode explicit external evidence in nearly half of adversarial conflicts, but also frequently suffered from systemic confidence collapse (Negative Transfer) when its parametric priors were contradicted. Our findings highlight the structural epistemic advantage and superior contextual adherence of SLMs in strict extraction workflows.

cs.CL↗

arXiv2026

Automatic Stability and Recovery for Neural Network Training

Training modern neural networks is increasingly fragile, with rare but severe destabilizing updates often causing irreversible divergence or silent performance degradation. Existing optimization methods primarily rely on preventive mechanisms embedded within the optimizer, offering limited ability to detect and recover from instability once it occurs. We introduce a supervisory runtime stability framework that treats optimization as a controlled stochastic process. By isolating an innovation signal derived from secondary measurements, such as validation probes, the framework enables automatic detection and recovery from destabilizing updates without modifying the underlying optimizer. We provide theoretical runtime safety guarantees that formalize bounded degradation and recovery. Our implementation incurs minimal overhead and is compatible with memory-constrained training settings.

cs.LG↗

arXiv2026

Kalman-Inspired Runtime Stability and Recovery in Hybrid Reasoning Systems

Hybrid reasoning systems that combine learned components with model-based inference are increasingly deployed in tool-augmented decision loops, yet their runtime behavior under partial observability and sustained evidence mismatch remains poorly understood. In practice, failures often arise as gradual divergence of internal reasoning dynamics rather than as isolated prediction errors. This work studies runtime stability in hybrid reasoning systems from a Kalman-inspired perspective. We model reasoning as a stochastic inference process driven by an internal innovation signal and introduce cognitive drift as a measurable runtime phenomenon. Stability is defined in terms of detectability, bounded divergence, and recoverability rather than task-level correctness. We propose a runtime stability framework that monitors innovation statistics, detects emerging instability, and triggers recovery-aware control mechanisms. Experiments on multi-step, tool-augmented reasoning tasks demonstrate reliable instability detection prior to task failure and show that recovery, when feasible, re-establishes bounded internal behavior within finite time. These results emphasize runtime stability as a system-level requirement for reliable reasoning under uncertainty.

cs.LG↗

arXiv2025

Selection-Induced Contraction of Innovation Statistics in Gated Kalman Filters

Validation gating is a fundamental component of classical Kalman-based tracking systems. Only measurements whose normalized innovation squared (NIS) falls below a prescribed threshold are considered for state update. While this procedure is statistically motivated by the chi-square distribution, it implicitly replaces the unconditional innovation process with a conditionally observed one, restricted to the validation event. This paper shows that innovation statistics computed after gating converge to gate-conditioned rather than nominal quantities. Under classical linear--Gaussian assumptions, we derive exact expressions for the first- and second-order moments of the innovation conditioned on ellipsoidal gating, and show that gating induces a deterministic, dimension-dependent contraction of the innovation covariance. The analysis is extended to NN association, which is shown to act as an additional statistical selection operator. We prove that selecting the minimum-norm innovation among multiple in-gate measurements introduces an unavoidable energy contraction, implying that nominal innovation statistics cannot be preserved under nontrivial gating and association. Closed-form results in the two-dimensional case quantify the combined effects and illustrate their practical significance.

stat.ME↗

arXiv2025

MTTR-A: Measuring Cognitive Recovery Latency in Multi-Agent Systems

Reliability in multi-agent systems (MAS) built on large language models is increasingly limited by cognitive failures rather than infrastructure faults. Existing observability tools describe failures but do not quantify how quickly distributed reasoning recovers once coherence is lost. We introduce MTTR-A (Mean Time-to-Recovery for Agentic Systems), a runtime reliability metric that measures cognitive recovery latency in MAS. MTTR-A adapts classical dependability theory to agentic orchestration, capturing the time required to detect reasoning drift and restore coherent operation. We further define complementary metrics, including MTBF and a normalized recovery ratio (NRR), and establish theoretical bounds linking recovery latency to long-run cognitive uptime. Using a LangGraph-based benchmark with simulated drift and reflex recovery, we empirically demonstrate measurable recovery behavior across multiple reflex strategies. This work establishes a quantitative foundation for runtime cognitive dependability in distributed agentic systems.

cs.MA↗

arXiv2025

Improving Requirements Classification with SMOTE-Tomek Preprocessing

This study emphasizes the domain of requirements engineering by applying the SMOTE-Tomek preprocessing technique, combined with stratified K-fold cross-validation, to address class imbalance in the PROMISE dataset. This dataset comprises 969 categorized requirements, classified into functional and non-functional types. The proposed approach enhances the representation of minority classes while maintaining the integrity of validation folds, leading to a notable improvement in classification accuracy. Logistic regression achieved 76.16%, significantly surpassing the baseline of 59.85%. These results highlight the applicability and efficiency of machine learning models as scalable and interpretable solutions.

cs.SE↗

arXiv2024

Learning-Based Speed Estimation from Accelerometer-Only Inertial Sensing

The proposed model, CarSpeedNet, estimates scalar vehicle speed from a window of three-axis smartphone acceleration, without gyroscope, wheel-odometry, vehicle-bus, or positioning input at inference. The reported experiment comprises 13.2 hours of on-road driving. Beyond the network comparison, a finite-context analysis treats window length as part of the sensing problem. For nested histories, the minimum Bayes mean-square error is non-increasing with context; a complementary cue-coverage relation links the same window to the amount of speed-dependent vibration presented to the network and to the observation horizon. At a 1-s input, CarSpeedNet is compared with five temporal-network alternatives; the effect of temporal context is then measured over six window lengths. On the 0.5-hour holdout, a 4-s window yielded a root-mean-square error (RMSE) of 1.8 m/s and a mean absolute error (MAE) of 0.72 m/s, compared with 2.9 and 1.3 m/s at 1 s. The 178,169-parameter network requires about 0.68 MiB for 32-bit weights.

cs.LG↗

arXiv2023

Learning Position From Vehicle Vibration Using an Inertial Measurement Unit

This paper presents a novel approach to vehicle positioning that operates without reliance on the global navigation satellite system (GNSS). Traditional GNSS approaches are vulnerable to interference in certain environments, rendering them unreliable in situations such as urban canyons, under flyovers, or in low reception areas. This study proposes a vehicle positioning method based on learning the road signature from accelerometer and gyroscope measurements obtained by an inertial measurement unit (IMU) sensor. In our approach, the route is divided into segments, each with a distinct signature that the IMU can detect through the vibrations of a vehicle in response to subtle changes in the road surface. The study presents two different data-driven methods for learning the road segment from IMU measurements. One method is based on convolutional neural networks and the other on ensemble random forest applied to handcrafted features. Additionally, the authors present an algorithm to deduce the position of a vehicle in real-time using the learned road segment. The approach was applied in two positioning tasks: (i) a car along a 6[km] route in a dense urban area; (ii) an e-scooter on a 1[km] route that combined road and pavement surfaces. The mean error between the proposed method's position and the ground truth was approximately 50[m] for the car and 30[m] for the e-scooter. Compared to a solution based on time integration of the IMU measurements, the proposed approach has a mean error of more than 5 times better for e-scooters and 20 times better for cars.

cs.RO↗

arXiv2023

Surface Recognition for e-Scooter Using Smartphone IMU Sensor

In recent years, as the use of micromobility gained popularity, technological challenges connected to e-scooters became increasingly important. This paper focuses on road surface recognition, an important task in this area. A reliable and accurate method for road surface recognition can help improve the safety and stability of the vehicle. Here a data-driven method is proposed to recognize if an e-scooter is on a road or a sidewalk. The proposed method uses only the widely available inertial measurement unit (IMU) sensors on a smartphone device. deep neural networks (DNNs) are used to infer whether an e-scooteris driving on a road or on a sidewalk by solving a binary classification problem. A data set is collected and several different deep models as well as classical machine learning approaches for the binary classification problem are applied and compared. Experiment results on a route containing the two surfaces are presented demonstrating the DNNs ability to distinguish between them.

eess.SP↗

arXiv2022

ProNet: Adaptive Process Noise Estimation for INS/DVL Fusion

Inertial and Doppler velocity log sensors are commonly used to provide the navigation solution for autonomous underwater vehicles (AUV). To this end, a nonlinear filter is adopted for the fusion task. The filter's process noise covariance matrix is critical for filter accuracy and robustness. While this matrix varies over time during the AUV mission, the filter assumes a constant matrix. Several models and learning approaches in the literature suggest tuning the process noise covariance during operation. In this work, we propose ProNet, a hybrid, adaptive process, noise estimation approach for a velocity-aided navigation filter. ProNet requires only the inertial sensor reading to regress the process noise covariance. Once learned, it is fed into the model-based navigation filter, resulting in a hybrid filter. Simulation results show the benefits of our approach compared to other models and learning adaptive approaches.

cs.RO↗

arXiv2022

Deep Learning for Inertial Sensor Alignment

Accurate alignment of a fixed mobile device equipped with inertial sensors inside a moving vehicle is important for navigation, activity recognition, and other applications. Accurate estimation of the device mounting angle is required to rotate the inertial measurement from the sensor frame to the moving platform frame to standardize measurements and improve the performance of the target task. In this work, a data-driven approach using deep neural networks (DNNs) is proposed to learn the yaw mounting angle of a smartphone equipped with an inertial measurement unit (IMU) and strapped to a car. The proposed model uses only the accelerometer and gyroscope readings from an IMU as input and, in contrast to existing solutions, does not require global position inputs from global navigation satellite systems (GNSS). To train the model in a supervised manner, IMU data is collected for training and validation with the sensor mounted at a known yaw mounting angle, and a range of ground truth labels is generated by applying a random rotation in a bounded range to the measurements. The trained model is tested on data with real rotations showing similar performance as with synthetic rotations. The trained model is deployed on an Android device and evaluated in real-time to test the accuracy of the estimated yaw mounting angle. The model is shown to find the mounting angle at an accuracy of 8 degrees within 5 seconds, and 4 degrees within 27 seconds. An experiment is conducted to compare the proposed model with an existing off-the-shelf solution.

cs.CV↗

arXiv2022

A Hybrid Adaptive Velocity Aided Navigation Filter with Application to INS/DVL Fusion

Autonomous underwater vehicles (AUV) are commonly used in many underwater applications. Usually, inertial sensors and Doppler velocity log readings are used in a nonlinear filter to estimate the AUV navigation solution. The process noise covariance matrix is tuned according to the inertial sensors' characteristics. This matrix greatly influences filter accuracy, robustness, and performance. A common practice is to assume that this matrix is fixed during the AUV operation. However, it varies over time as the amount of uncertainty is unknown. Therefore, adaptive tuning of this matrix can lead to a significant improvement in the filter performance. In this work, we propose a learning-based adaptive velocity-aided navigation filter. To that end, handcrafted features are generated and used to tune the momentary system noise covariance matrix. Once the process noise covariance is learned, it is fed into the model-based navigation filter. Simulation results show the benefits of our approach compared to other adaptive approaches.

cs.RO↗

arXiv2022

Adaptive Step Size Learning with Applications to Velocity Aided Inertial Navigation System

Autonomous underwater vehicles (AUV) are commonly used in many underwater applications. Recently, the usage of multi-rotor unmanned autonomous vehicles (UAV) for marine applications is receiving more attention in the literature. Usually, both platforms employ an inertial navigation system (INS), and aiding sensors for an accurate navigation solution. In AUV navigation, Doppler velocity log (DVL) is mainly used to aid the INS, while for UAVs, it is common to use global navigation satellite systems (GNSS) receivers. The fusion between the aiding sensor and the INS requires a definition of step size parameter in the estimation process. It is responsible for the solution frequency update and, eventually, its accuracy. The choice of the step size poses a tradeoff between computational load and navigation performance. Generally, the aiding sensors update frequency is considered much slower compared to the INS operating frequency (hundreds Hertz). Such high rate is unnecessary for most platforms, specifically for low dynamics AUVs. In this work, a supervised machine learning based adaptive tuning scheme to select the proper INS step size is proposed. To that end, a velocity error bound is defined, allowing the INS/DVL or the INS/GNSS to act in a sub-optimal working conditions, and yet minimize the computational load. Results from simulations and field experiment show the benefits of using the proposed approach. In addition, the proposed framework can be applied to any other fusion scenarios between any type of sensors or platforms.

cs.RO↗

arXiv2022

A Hybrid Model and Learning-Based Adaptive Navigation Filter

The fusion between an inertial navigation system and global navigation satellite systems is regularly used in many platforms such as drones, land vehicles, and marine vessels. The fusion is commonly carried out in a model-based extended Kalman filter framework. One of the critical parameters of the filter is the process noise covariance. It is responsible for the real-time solution accuracy, as it considers both vehicle dynamics uncertainty and the inertial sensors quality. In most situations, the process noise is covariance assumed to be constant. Yet, due to vehicle dynamics and sensor measurement variations throughout the trajectory, the process noise covariance is subject to change. To cope with such situations, several adaptive model-based Kalman filters were suggested in the literature. In this paper, we propose a hybrid model and learning-based adaptive navigation filter. We rely on the model-based Kalman filter and design a deep neural network model to tune the momentary system noise covariance matrix, based only on the inertial sensor readings. Once the process noise covariance is learned, it is plugged into the well-established, model-based Kalman filter. After deriving the proposed hybrid framework, field experiment results using a quadrotor are presented and a comparison to model-based adaptive approaches is given. We show that the proposed method obtained an improvement of 25% in the position error. Furthermore, the proposed hybrid learning method can be used in any navigation filter and also in any relevant estimation problem.

eess.SY↗

arXiv2022

Learning Vehicle Trajectory Uncertainty

A novel approach for vehicle tracking using a hybrid adaptive Kalman filter is proposed. The filter utilizes recurrent neural networks to learn the vehicle's geometrical and kinematic features, which are then used in a supervised learning model to determine the actual process noise covariance in the Kalman framework. This approach addresses the limitations of traditional linear Kalman filters, which can suffer from degraded performance due to uncertainty in the vehicle kinematic trajectory modeling. Our method is evaluated and compared to other adaptive filters using the Oxford RobotCar dataset, and has shown to be effective in accurately determining the process noise covariance in real-time scenarios. Overall, this approach can be implemented in other estimation problems to improve performance.

cs.RO↗

arXiv2022

Learning Car Speed Using Inertial Sensors for Dead Reckoning Navigation

A deep neural network (DNN) is trained to estimate the speed of a car driving in an urban area using as input a stream of measurements from a low-cost six-axis inertial measurement unit (IMU). Three hours of data was collected by driving through the city of Ashdod, Israel in a car equipped with a global navigation satellite system (GNSS) real time kinematic (RTK) positioning device and a synchronized IMU. Ground truth labels for the car speed were calculated using the position measurements obtained at the high rate of 50 Hz. A DNN architecture with long short-term memory layers is proposed to enable high-frequency speed estimation that accounts for previous inputs history and the nonlinear relation between speed, acceleration and angular velocity. A simplified aided dead reckoning localization scheme is formulated to assess the trained model which provides the speed pseudo-measurement. The trained model is shown to substantially improve the position accuracy during a 4 minutes drive without the use of GNSS position updates.

cs.LG↗