Over the last decades, an increasing interest has been considered in improving vehicle performances by using advanced control systems. Nevertheless, the implementation of these systems can be a source of faults. To ensure vehicle stability without performance degradation, problems of both estimation and tracking control design are studied for vehicle dynamics in presence of faults and parametric uncertainties. The first part of this paper focuses on sideslip angle, yaw rate and fault estimations via a Proportional Integral observer (PIO). Then, a PIO based tracking control design method, explained via a block diagram, is proposed not only to maintain vehicle stability but also to track the desired trajectories despite the fault effects. The observer and controller gains are computed by solving convex optimization problems under Linear Matrix Inequality (LMI) constraints. Finally, simulation results are presented to prove the effectiveness of the proposed process.

The retina is affected by glaucoma and diabetic retinopathy (DR). Glaucoma must be detected early because it is irreversible and one of the leading causes of blindness. A delayed diagnosis will result in permanent vision loss. It is characterized primarily by ganglion cell dysfunction, which changes the thickness of the retinal nerve fiber layer and the shape of the optic nerve head. As a result, early detection of glaucoma is critical for preventing vision loss. This study employs a hybrid approach to glaucoma diagnosis by combining its powerful Multichannel Gabor filtering and Principal Component Analysis (PCA) capabilities with various transfer learning architectures such as MobileNet, MobileNetV2, and NASNet. These classifiers divide the source retinal images into two groups: glaucoma and non-glaucoma. The suggested approaches are used and evaluated on a dataset of Retinal Fundus Images. For the glaucoma diagnostic method, this strategy yields 99% Precision, 97% Recall, and 98% Accuracy.

International audience

Accurately estimating the position of static objects, such as traffic lights, from the moving camera of a self-driving car is a challenging problem. In this work, we present a system that improves the localization of static objects by jointly-optimizing the components of the system via learning. Our system is comprised of networks that perform: 1) 5DoF object pose estimation from a single image, 2) association of objects between pairs of frames, and 3) multi-object tracking to produce the final geo-localization of the static objects within the scene. We evaluate our approach using a publicly-available data set, focusing on traffic lights due to data availability. For each component, we compare against contemporary alternatives and show significantly-improved performance. We also show that the end-to-end system performance is further improved via joint-training of the constituent models. Code is available at: https://github.com/MedChaabane/Static_Objects_Geolocalization.

In this paper, we present a novel, end-to-end 6D object pose estimation method that operates on RGB inputs. Our approach is composed of 2 main components: the first component classifies the objects in the input image and proposes an initial 6D pose estimate through a multi-task, CNN-based encoder/multi-decoder module. The second component, a refinement module, includes a renderer and a multi-attentional pose refinement network, which iteratively refines the estimated poses by utilizing both appearance features and flow vectors. Our refiner takes advantage of the hybrid representation of the initial pose estimates to predict the relative errors with respect to the target poses. It is further augmented by a spatial multi-attention block that emphasizes objects' discriminative feature parts. Experiments on three benchmarks for 6D pose estimation show that our proposed pipeline outperforms state-of-the-art RGB-based methods with competitive runtime performance.

In this paper, we present an end-to-end future-prediction model that focuses on pedestrian safety. Specifically, our model uses previous video frames, recorded from the perspective of the vehicle, to predict if a pedestrian will cross in front of the vehicle. The long term goal of this work is to design a fully autonomous system that acts and reacts as a defensive human driver would - predicting future events and reacting to mitigate risk. We focus on pedestrian-vehicle interactions because of the high risk of harm to the pedestrian if their actions are miss-predicted. Our end-to-end model consists of two stages: the first stage is an encoder/decoder network that learns to predict future video frames. The second stage is a deep spatiotemporal network that utilizes the predicted frames of the first stage to predict the pedestrian's future action. Our system achieves state-of-the-art accuracy on the Joint Attention for Autonomous Driving (JAAD) dataset on both future frames prediction, with a pixel-wise prediction l1error of 1.12, and pedestrian behavior prediction with an average precision of 86.7.

MOTIVATION: Over the past two decades, a circular form of RNA (circular RNA), produced through alternative splicing, has become the focus of scientific studies due to its major role as a microRNA (miRNA) activity modulator and its association with various diseases including cancer. Therefore, the detection of circular RNAs is vital to understanding their biogenesis and purpose. Prediction of circular RNA can be achieved in three steps: distinguishing non-coding RNAs from protein coding gene transcripts, separating short and long non-coding RNAs and predicting circular RNAs from other long non-coding RNAs (lncRNAs). However, the available tools are less than 80 percent accurate for distinguishing circular RNAs from other lncRNAs due to difficulty of classification. Therefore, the availability of a more accurate and fast machine learning method for the identification of circular RNAs, which considers the specific features of circular RNA, is essential to the development of systematic annotation. RESULTS: Here we present an End-to-End deep learning framework, circDeep, to classify circular RNA from other lncRNA. circDeep fuses an RCM descriptor, ACNN-BLSTM sequence descriptor and a conservation descriptor into high level abstraction descriptors, where the shared representations across different modalities are integrated. The experiments show that circDeep is not only faster than existing tools but also performs at an unprecedented level of accuracy by achieving a 12 percent increase in accuracy over the other tools. AVAILABILITY AND IMPLEMENTATION: https://github.com/UofLBioinformatics/circDeep. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Over the past two decades, researchers have discovered a special form of alternative splicing that produces a circular form of RNA. Although these circular RNAs (circRNAs) have garnered considerable attention in the scientific community for their biogenesis and functions, the focus of current studies has been on the tissue-specific circRNAs that exist only in one tissue but not in other tissues or on the disease-specific circRNAs that exist in certain disease conditions, such as cancer, but not under normal conditions. This approach was conducted in the relative absence of methods that analyze a group of common circRNAs that exist in both conditions, but are more abundant in one condition relative to another (differentially expressed). Studies of differentially expressed circRNAs (DECs) between two conditions would serve as a significant first step in filling this void. Here, we introduce a novel computational tool, seekCRIT (seek for differentially expressed CircRNAs In Transcriptome), that identifies the DECs between two conditions from high-throughput sequencing data. Using rat retina RNA-seq data from ischemic and normal conditions, we show that over 74% of identifiable circRNAs are expressed in both conditions and over 40 circRNAs are differentially expressed between two conditions. We also obtain a high qPCR validation rate of 90% for DECs with a FDR of < 5%. Our results demonstrate that seekCRIT is a novel and efficient approach to detect DECs using rRNA depleted RNA-seq data. seekCRIT is freely downloadable at https://github.com/UofLBioinformatics/seekCRIT. The source code is licensed under the MIT License. seekCRIT is developed and tested on Linux CentOS-7.

MOTIVATION: Deep learning architectures have recently demonstrated their power in predicting DNA- and RNA-binding specificity. Existing methods fall into three classes: Some are based on convolutional neural networks (CNNs), others use recurrent neural networks (RNNs) and others rely on hybrid architectures combining CNNs and RNNs. However, based on existing studies the relative merit of the various architectures remains unclear. RESULTS: In this study we present a systematic exploration of deep learning architectures for predicting DNA- and RNA-binding specificity. For this purpose, we present deepRAM, an end-to-end deep learning tool that provides an implementation of a wide selection of architectures; its fully automatic model selection procedure allows us to perform a fair and unbiased comparison of deep learning architectures. We find that deeper more complex architectures provide a clear advantage with sufficient training data, and that hybrid CNN/RNN architectures outperform other methods in terms of accuracy. Our work provides guidelines that can assist the practitioner in choosing an appropriate network architecture, and provides insight on the difference between the models learned by convolutional and recurrent networks. In particular, we find that although recurrent networks improve model accuracy, this comes at the expense of a loss in the interpretability of the features learned by the model. AVAILABILITY AND IMPLEMENTATION: The source code for deepRAM is available at https://github.com/MedChaabane/deepRAM. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

This brief studies the problem of state and fault estimation for Takagi-Sugeno (T-S) fuzzy systems with time delay subject to simultaneously sensor/actuator faults and external disturbances. By extending the sensor faults as an auxiliary state vector, the original system is transformed into an augmented descriptor system. Adaptive Observer is proposed to achieve a simultaneous estimation of descriptor system states, actuator and sensor faults using the constrained H∞norm optimization. Based on the Lyapunov functional method and delay-partitioning technique, delay-dependent design conditions of the fuzzy observer are derived in terms of linear matrix inequalities (LMIs). An example is finally presented to validate our findings.

This paper deals with Wind Energy Conversion System (WECS) control equipped by Doubly-Fed Induction Generator (DFIG). Based on variable speed wind turbine and aerodynamic torque estimation, high order sliding mode controller is developed to improve the dynamic performances of WECS by producing the required electromagnetic torque generator. The desired DFIG torque is directly tracked using the proposed robust control to extract maximum power. The performance and the effectiveness of the proposed robust controller are compared to PI controller in presence of high wind speed variations.

This work concerns different stability approaches for Takagi-Sugeno (T-S) fuzzy systems. We present some recent approaches based on the idea of multiple Lyapunov functions and slack matrices to reduce the conservatism of stability analysis conditions. Then, new stability conditions are proposed in order to more reduce the conservatism using another upper bound of time-derivative of membership functions, fuzzy Lyapunov functions and slack matrices. Consequently, a large stability domain is obtained. Finally, examples are provided to illustrate the effectiveness of the proposed approaches in stability analysis

This paper is concerned with fault estimation design for continuous-time Takagi-Sugeno (T-S) fuzzy systems with time delay by using adaptive fault diagnostic observer. Through constructing an appropriate type of Lyapunov function, criteria is established to reduce the conservatism of the design procedure. Based on appropriate type of Lyapunov function, new sufficient delay-dependent conditions are given to guarantee the stability of error dynamics in terms of Linear Matrix Inequalities (LMI). Finally, the effectiveness of the proposed approach is illustrated through a simulation example.

In this paper, a method of robust H∞reliable control for uncertain Takagi-Sugeno (T-S) fuzzy systems with unavailable states and time-varying delay is developed. In terms of linear-matrix inequalities (LMIs), we derive sufficient conditions for the existence of the fuzzy observer and the reliable fuzzy controller. The single-step LMI conditions, depending on the size of the delay, are proposed for the observer-based H∞reliable controller design. This overcomes the drawback of two-step LMI approach in literature. Moreover, the designed fuzzy controller is reliable in the sense that the stability and the satisfactory performance of the closed-loop system are achieved not only under normal operation but in the presence of some actuator faults as well. The result is given without the assumption of the rate of change of time-delay function (i.e., |τ̊| <; 1), which makes it applicable to both slow and fast time-varying delayed systems. Without considering reliability, the proposed result is much less conservative than most of the existing results. Finally, several examples are given to show the effectiveness and advantages of our results.

This paper deals with the Hinfinestimation of both system state and faults for T-S (Takagi-Sugeno) fuzzy model with bounded input disturbances. Based on the descriptor technique, the sensor faults are considered as an auxiliary state variable. Then a descriptor observer for the obtained augmented system is designed and the observer gains are determined in LMI (Linear Matrix Inequalities) formulation. This method has the advantage to estimate the state variables and the sensor faults simultaneously with Hinfinapproach. Numerical example shows the efficiency of the proposed method.