The rapid advancement of machine learning (ML) has led to its increasing integration into cyber-physical systems (CPS) across diverse domains. While CPS offer powerful capabilities, incorporating ML components introduces significant safety and assurance challenges. Among ML techniques, reinforcement learning (RL) is particularly suited for CPS due to its capacity to handle complex, dynamic environments where explicit models of interaction between system and environment are unavailable or difficult to construct. However, in safety-critical applications, this learning process must not only be effective but demonstrably safe. Safe-RL methods aim to address this by incorporating safety constraints during learning, yet they fall short in providing systematic assurance across the RL lifecycle. The AMLAS methodology offers structured guidance for assuring the safety of supervised learning components, but it does not directly apply to the unique challenges posed by RL. In this paper, we adapt AMLAS to provide a framework for generating assurance arguments for an RL-enabled system through an iterative process; AMLAS-RL. We demonstrate AMLAS-RL using a running example of a wheeled vehicle tasked with reaching a target goal without collision.

This study evaluates the performance of a compression method, called CompactifAI, developed by Multiverse Computing, applied to the large language model Llama 3.1 8B\cite{llama}. The evaluation focused on model efficiency (in terms of energy consumption) and accuracy using respectively the frameworks Codecarbon\cite{codecarbon} and Ragas\cite{ragas}. A comparison was performed between the model compressed with CompactifAI\cite{compactifai}\cite{compactifai2} and its full-size version. Our findings reveal that the compressed model using CompactifAI not only significantly reduced the computational resources but also maintained the model accuracy, making the model more efficient, scalable and cost-effective.

Vision-Language Models (VLMs) offer promising capabilities for mobile devices, but their deployment faces significant challenges due to computational limitations and energy inefficiency, especially for real-time applications. This study provides a comprehensive survey of deployment frameworks for VLMs on mobile devices, evaluating llama.cpp, MLC-Imp, and mllm in the context of running LLaVA-1.5 7B, MobileVLM-3B, and Imp-v1.5 3B as representative workloads on a OnePlus 13R. Each deployment framework was evaluated on the OnePlus 13R while running VLMs, with measurements covering CPU, GPU, and NPU utilization, temperature, inference time, power consumption, and user experience. Benchmarking revealed critical performance bottlenecks across frameworks: CPU resources were consistently over-utilized during token generation, while GPU and NPU accelerators were largely unused. When the GPU was used, primarily for image feature extraction, it was saturated, leading to degraded device responsiveness. The study contributes framework-level benchmarks, practical profiling tools, and an in-depth analysis of hardware utilization bottlenecks, highlighting the consistent overuse of CPUs and the ineffective or unstable use of GPUs and NPUs in current deployment frameworks.

Cloud and cloud shadow masking is a crucial preprocessing step in hyperspectral satellite imaging, enabling the extraction of high-quality, analysis-ready data. This study evaluates various machine learning approaches, including gradient boosting methods such as XGBoost and LightGBM as well as convolutional neural networks (CNNs). All boosting and CNN models achieved accuracies exceeding 93%. Among the investigated models, the CNN with feature reduction emerged as the most efficient, offering a balance of high accuracy, low storage requirements, and rapid inference times on both CPUs and GPUs. Variations of this version, with only up to 597 trainable parameters, demonstrated the best trade-off in terms of deployment feasibility, accuracy, and computational efficiency. These results demonstrate the potential of lightweight artificial intelligence (AI) models for real-time hyperspectral image processing, supporting the development of on-board satellite AI systems for space-based applications.

Steering vectors are a lightweight method for controlling text properties by adding a learned bias to language model activations at inference time. So far, steering vectors have predominantly been evaluated in multiple-choice settings, while their effectiveness in free-form generation tasks remains understudied. Moving "Beyond Multiple Choice," we thoroughly evaluate the effectiveness of steering vectors in adaptively controlling topical focus, sentiment, toxicity, and readability in abstractive summaries of the NEWTS dataset. We find that steering effectively controls the targeted summary properties, but high steering strengths consistently degrade both intrinsic and extrinsic text quality. Compared to steering, prompting offers weaker control, while preserving text quality. Combining steering and prompting yields the strongest control over text properties and offers the most favorable efficacy-quality trade-off at moderate steering strengths. Our results underscore the practical trade-off between control strength and text quality preservation when applying steering vectors to free-form generation tasks.

Federated Learning has emerged as a leading paradigm for decentralized, privacy-preserving learning, particularly relevant in the era of interconnected edge devices equipped with sensors. However, the practical implementation of Federated Learning faces three primary challenges: the need for human involvement in costly data labelling processes for target adaptation, covariate shift in client device data collection due to environmental factors affecting sensors, leading to discrepancies between source and target samples, and the impracticality of continuous or regular model updates in resource-constrained environments due to limited data transmission capabilities and technical constraints on channel availability and energy efficiency. To tackle these issues, we expand upon an efficient and scalable Federated Learning framework tailored for real-world client adaptation in industrial settings. This framework leverages a pre-trained source model comprising a deep backbone, an adaptation module, and a classifier running on a powerful server. By freezing the backbone and classifier during client adaptation on resource-constrained devices, we allow the domain adaptive linear layer to handle target domain adaptation, thus minimizing overall computational overhead. Furthermore, this setup, designated as FedAcross+, is extended to encompass the processing of streaming data, thereby rendering the solution suitable for non-stationary environments. Extensive experimental results demonstrate the effectiveness of FedAcross+ in achieving competitive adaptation on low-end client devices with limited target samples, successfully addressing the challenge of domain shift. Moreover, our framework accommodates sporadic model updates within resource-constrained environments, ensuring practical and seamless deployment.

--Unmanned Aerial V ehicles (UA Vs) are vital for public safety, particularly in wildfire monitoring, where early detection minimizes environmental impact. In UA V-Assisted Wildfire Monitoring (UA WM) systems, joint optimization of sensor transmission scheduling and velocity is critical for minimizing Age of Information (AoI) from stale sensor data. Deep Reinforcement Learning (DRL) has been used for such optimization; however, its limitations such as low sampling efficiency, simulation-to-reality gaps, and complex training render it unsuitable for time-critical applications like wildfire monitoring. This paper introduces a new online Flight Resource Allocation scheme based on LLM-Enabled In-Context Learning (FRSICL) to jointly optimize the UA V's flight control and data collection schedule along the trajectory in real time, thereby asymptotically minimizing the average AoI across ground sensors. In contrast to DRL, FRSICL generates data collection schedules and controls velocity using natural language task descriptions and feedback from the environment, enabling dynamic decision-making without extensive retraining. Simulation results confirm the effectiveness of the proposed FRSICL compared to Proximal Policy Optimization (PPO) and Nearest-Neighbor baselines. Nowadays, Unmanned Aerial V ehicles (UA Vs) have a wide range of applications in public safety [1], energy [2], and environmental monitoring [3]. Public safety UA Vs serve critical roles in emergency operations, including search and rescue (SAR), wildfire surveillance, and disaster management.

Accurate inference of human intent enables human-robot collaboration without constraining human control or causing conflicts between humans and robots. We present GUIDER (Global User Intent Dual-phase Estimation for Robots), a probabilistic framework that enables a robot to estimate the intent of human operators. GUIDER maintains two coupled belief layers, one tracking navigation goals and the other manipulation goals. In the Navigation phase, a Synergy Map blends controller velocity with an occupancy grid to rank interaction areas. Upon arrival at a goal, an autonomous multi-view scan builds a local 3D cloud. The Manipulation phase combines U2Net saliency, FastSAM instance saliency, and three geometric grasp-feasibility tests, with an end-effector kinematics-aware update rule that evolves object probabilities in real-time. GUIDER can recognize areas and objects of intent without predefined goals. We evaluated GUIDER on 25 trials (five participants x five task variants) in Isaac Sim, and compared it with two baselines, one for navigation and one for manipulation. Across the 25 trials, GUIDER achieved a median stability of 93-100% during navigation, compared with 60-100% for the BOIR baseline, with an improvement of 39.5% in a redirection scenario (T5). During manipulation, stability reached 94-100% (versus 69-100% for Trajectron), with a 31.4% difference in a redirection task (T3). In geometry-constrained trials (manipulation), GUIDER recognized the object intent three times earlier than Trajectron (median remaining time to confident prediction 23.6 s vs 7.8 s). These results validate our dual-phase framework and show improvements in intent inference in both phases of mobile manipulation tasks.

This paper presents a novel framework for whole-body torque control of humanoid robots without joint torque sensors, designed for systems with electric motors and high-ratio harmonic drives. The approach integrates Physics-Informed Neural Networks (PINNs) for friction modeling and Unscented Kalman Filtering (UKF) for joint torque estimation, within a real-time torque control architecture. PINNs estimate nonlinear static and dynamic friction from joint and motor velocity readings, capturing effects like motor actuation without joint movement. The UKF utilizes PINN-based friction estimates as direct measurement inputs, improving torque estimation robustness. Experimental validation on the ergoCub humanoid robot demonstrates improved torque tracking accuracy, enhanced energy efficiency, and superior disturbance rejection compared to the state-of-the-art Recursive Newton-Euler Algorithm (RNEA), using a dynamic balancing experiment. The framework's scalability is shown by consistent performance across robots with similar hardware but different friction characteristics, without re-identification. Furthermore, a comparative analysis with position control highlights the advantages of the proposed torque control approach. The results establish the method as a scalable and practical solution for sensorless torque control in humanoid robots, ensuring torque tracking, adaptability, and stability in dynamic environments.

-- This work presents a vision-based underwater exploration and inspection autonomy solution integrated into Ariel, a custom vision-driven underwater robot. Ariel carries a 5 camera and IMU based sensing suite, enabling a refraction-aware multi-camera visual-inertial state estimation method aided by a learning-based proprioceptive robot velocity prediction method that enhances robustness against visual degradation. Furthermore, our previously developed and extensively field-verified autonomous exploration and general visual inspection solution is integrated on Ariel, providing aerial drone-level autonomy underwater . The proposed system is field-tested in a submarine dry dock in Trondheim under challenging visual conditions. The field demonstration shows the robustness of the state estimation solution and the generalizability of the path planning techniques across robot embodiments.