1,721,072 research outputs found
A weighting approach to the shared-control of lateral vehicle dynamics
No full textThis work focuses on the shared-control of vehicle lateral dynamics, proposing a control architecture that can help the driver following the desired trajectory in dangerous situations. To do this, a novel shared-control formulation is presented, in which the assistance and the driver steering torques are adaptively weighted to dynamically change the control authority share between driver and controller. Simulations in the CarSim multibody simulation environment favorably witness the effectiveness of the overall approach
A new comprehensive monitoring and diagnostic approach for early detection of mechanical degradation in helicopter transmission systems
No full textHelicopters are complex and vulnerable due to single-load-path critical parts that transmit the engine's power to the rotors. A fault in even one single transmission's gear component may compromise the whole helicopter, involving high maintenance costs and safety hazards. In this work, we present an effective diagnosis and monitoring system for the early detection of the mechanical degradation in such components, also capable of providing insights on the damage's causes. The classification task is performed by an ensemble of two learners: a convolutional autoencoder and a distance&density-based unsupervised classifier that use as regressors specific Health Indexes (HIs) and flight parameters. The proposed approach leverages the autoencoder reconstruction error information to infer the most probable cause of each detected fault, and enacts post-processing filtering policies defined to reduce the number of false alarms. Extensive experimental validation witnesses the effectiveness and robustness of the proposed approach
Bias Score: Estimating Gender Bias in Sentence Representations
No full textThe ever-increasing number of applications based on semantic text analysis is making natural language understanding a fundamental task. Language models are used for a variety of tasks, such as parsing CVs or improving web search results. At the same time, concern is growing around embedding-based language models, which often exhibit social bias and lack of transparency, despite their popularity and widespread use. Word embeddings in particular exhibit a large amount of gender bias, and they have been shown to reflect social stereotypes. Recently, sentence embeddings have been introduced as a novel and powerful technique to represent entire sentences as vectors. However, traditional methods for estimating gender bias cannot be applied to sentence representations, because gender-neutral entities cannot be easily identified and listed. We propose a new metric to estimate gender bias in sentence embeddings, named bias score. Our solution, leveraging the semantic importance of individual words and previous research on gender bias in word embeddings, is able to discern between correct and biased gender information at sentence level. Experiments on a real-world dataset demonstrates that our novel metric identifies gender stereotyped sentences
Tire-Wear Control in Aircraft via Active Braking
Full text linkIn ground vehicles, tire consumption is, in general, mainly due to the mileage covered, and in fact, the life span of tires, at least in common situations, is rather long. In the aeronautical context, and for aircraft in particular, instead, tire consumption plays a crucial role in determining the maintenance costs. This is due to the fact that, in aircraft braking, nearly all maneuvers activate the antiskid controller, which remains in use for long time intervals. In ground vehicles, instead, anti-lock braking systems (ABS) are usually active for short time intervals that cover a part of the braking maneuvers only. Thus, tire consumption in the automotive context is usually studied under constant speed assumptions. In this article, we formulate a tire consumption model that encompasses explicitly the wheel acceleration/deceleration dynamics and show that tire wear can be directly related to the antiskid controller parameters. Based on this, a sensitivity analysis of tire consumption versus braking performance is carried out, showing that, using an appropriate antiskid control approach, one may directly formulate the braking problem as a tire consumption regulation one, being sure that the resulting braking performance will have an a priori guaranteed outcome. The tire-wear model is also validated in an experimental setting
Combining tire-wear and braking control for aeronautical applications
Full text linkIn ground vehicles, tire consumption is in general mainly due to the mileage covered, and in fact the life span of tires, at least in common situations, is rather long. In the aeronautical context, and for aircraft in particular, instead, tire consumption plays a crucial role in determining the maintenance costs. This is due to the fact that, in aircraft braking, nearly all maneuvers activate the anti-skid controller, which remains in use for long time intervals. In ground vehicles, instead ABS systems are usually active for short time intervals which cover a part of the braking maneuvers only. Thus, tire consumption in the automotive context is usually studied under constant speed assumptions. In this work, we formulate a tire consumption models that encompasses explicitly the wheel acceleration/deceleration dynamics, and we show that tire wear can be directly related to the anti-skid controller parameters. Based on this, a sensitivity analysis of tire-consumption versus braking performance is carried out, showing that the braking control problem can be reformulated as a tire consumption regulation one
Wheel-slip estimation for advanced braking controllers in aircraft: Model based vs. black-box approaches
No full textAs of today, most aircraft are endowed with Anti-lock Braking Systems that are active during landings and rejected take-off manoeuvres, ensuring the maximum exploitation of road-friction capability. Due to strict certification issues, the braking controller must function using only signals that are local to the landing gear, which is the aircraft sub-system hosting the braking actuators. In the most common scenarios, the only available signals are the wheel speed and the braking pressure. This limited set of information prevents the use of advanced braking control architectures which are now mainstream in the automotive sector, i.e., those based on the regulation of the wheel slip, so that typical aircraft Anti-lock Braking Systems are usually based on the regulation of the wheel deceleration. This paper investigates how a reliable wheel slip estimation can be achieved using wheel speed and braking pressure signals, analysing different wheel slip observers options. In particular, a model-based approach that uses a sliding-mode observer is proposed, together with a black-box approach based on Nonlinear Auto-Regressive models with eXogenous input, with a neural network implementation. Both approaches are tested and compared on experimental data, proving that the obtained estimation performance are adequate for use in closed-loop braking systems
Sistema di riconoscimento di manomissione del codice di un sistema di controllo mediante uso di processori con memoria trusted
No full text
- …
