Electric vehicles are expected to dominate the vehicle fleet in the near future due to their zero emissions of pollutants, reduced fossil fuel reserves, comfort, and lightness. However, Battery Electric Vehicles (BEVs) suffer from gradual performance degradation caused by irreversible chemical and physical changes inside their batteries. Moreover, predicting the health and remaining useful life of BEVs is difficult due to various internal and external factors. In this paper, we propose an integrated data-driven framework for accurately predicting the Remaining Useful Life (RUL) of lithium-ion batteries used in BEVs. As part of this work, we conducted a comprehensive analysis and comparison of publicly available battery datasets, providing an up-to-date list of data sources for the community. Our framework relies on a novel feature extraction strategy that accurately characterizes the battery, leading to improved RUL predictions. Feature types can be divided into three groups: initial state features, current state features, and historic state features. Experimental results indicate that the proposed method performs very well on the NASA benchmark dataset with an average accuracy of 97% on batteries that are not used during the training phase, demonstrating the ability of our framework to operate with batteries being discharged in any conditions.

Bird recognition in computer vision poses two main challenges: high intra-class variance and low inter-class variance. High intra-class variance refers to the significant variation in the appearance of individual birds within the same species. Low inter-class variance refers to the limited visual differences between distinct bird species. In this paper, we propose a robust integrated framework for bird recognition using a dynamic machine learning-based technique. Our system is designed to identify over 11,000 species of birds based on multiple components. As part of this work, we propose two public datasets. The first one (E-Moulouya BDD) contains over 13k images of birds for detection tasks. While the second one (LaSBiRD) contains about 5M labelled images of 11k species. Our experiments yielded promising results, indicating the impressive performance of our system in detecting and classifying birds. With a mAP of 0.715 for detection and an accuracy rate of 96% for classification.

Facial recognition is widely used, but the reliability of the embeddings extracted by most computer vision-based approaches is a challenge due to the high similarity among human faces and the effect of facial expressions and lighting. Our proposed approach aggregates multiple embeddings to generate a more robust reference for facial embedding comparison and explores the distances metrics to use in order to optimize the comparison efficiency while preserving complexity. We also apply our method to the state-of-the-art algorithm that extracts embeddings from faces in an image. The proposed approach was compared with several approaches. It optimizes the Resnet accuracy to 99.77%, Facenet to 99.79%, and Inception-ResnetV1 to 99.16%. Our approach preserves the inference time of the model while increasing its reliability since the number of comparisons is kept at a minimum. Our results demonstrate that our proposed approach offers an effective solution for addressing facial recognition in real-world environments.

An effective Battery Management System (BMS) is compulsory for the safe and reliable operation of lithium-ion batteries, which are increasingly being used in Electric Vehicles (EVs). Accurate State of Charge estimation (SoC) is a cumbersome task, since lithium-ion batteries are highly influenced by such random factors including driving loads, operational conditions, and aging. This work presents a comparison study of various machine learning algorithms for SoC estimation. To do this, a 3Ah LGHG2 battery cell was put through a series of temperature and driving cycle tests. The proposed models are evaluated in terms of accuracy and robustness. The simulation results have shown that the Gaussian Process Regression (GPR) model outperforms the other algorithms achieving R 2 and RMSE values of (97 %, 1.3 %) and (95 %, 1.6 %) in normal conditions and in a noisy environment, respectively.

There has been significant interest in using Convolutional Neural Networks (CNN) based methods for Automated Vehicular Surveillance (AVS) systems. Although these methods provide high accuracy, they are computationally expensive. On the other hand, Background Subtraction (BS)-based approaches are lightweight but provide insufficient information for tasks such as monitoring driving behavior and detecting traffic rules violations. In this paper, we propose a framework to reduce the complexity of CNN-based AVS methods, where a BS-based module is introduced as a preprocessing step to optimize the number of convolution operations executed by the CNN module. The BS-based module generates image-candidates containing only moving objects. A CNN-based detector with the appropriate number of convolutions is then applied to each image-candidate to handle the overlapping problem and improve detection performance. Four state-of-the-art CNN-based detection architectures were benchmarked as base models of the detection cores to evaluate the proposed framework. The experiments were conducted using a large-scale dataset. The computational complexity reduction of the proposed framework increases with the complexity of the considered CNN model's architecture (e.g., 30.6% for YOLOv5s with 7.3M parameters; 52.2% for YOLOv5x with 87.7M parameters), without undermining accuracy.

Range anxiety remains one of the main hurdles to the widespread adoption of electric vehicles (EVs). To mitigate this issue, accurate energy consumption prediction is required. In this study, a hybrid approach is proposed toward this objective by taking into account driving behavior, road conditions, natural environment, and additional weight. The main components of the EV were simulated using physical and equation-based models. A rich synthetic dataset illustrating different driving scenarios was then constructed. Real-world data were also collected using a city car. A machine learning model was built to relate the mechanical power to the electric power. The proposed predictive method achieved an R2 of 0.99 on test synthetic data and an R2 of 0.98 on real-world data. Furthermore, the instantaneous regenerative braking power efficiency as a function of the deceleration level was also investigated in this study.

We propose a system for monitoring the driving maneuver at road intersections using rule-based reasoning and deep learning-based computer vision techniques. Along with detecting and classifying turning movements online, the system also detects violations such as ignoring STOP signs and failing to yield the right-of-way to other drivers. There is no distinction between temporarily and permanently stopped vehicles in the majority of frameworks proposed in the literature. Therefore, to conduct an accurate right-of-way study, permanently stopped vehicles should be excluded not to confound the results. Moreover, we also propose in this work a low-cost Convolutional Neural Network (CNN)-based object detection framework able to detect moving and temporally stopped vehicles. The detection framework combines the reasoning system with background subtraction and a CNN-based object detector. The obtained results are promising. Compared to the conventional CNN-based methods, the detection framework reduces the execution time of the object detection module by about 30% (i.e., 54.1 instead of 75ms/image) while preserving the same detection reliability. The accuracy of trajectory recognition is 95.32%, that of the zero-speed detection is 96.67%, and the right-of-way detection was perfect.

We propose a system for monitoring the headway and following distance using a roadside camera and deep learning-based computer vision techniques. The system is composed of a vehicle detector and tracker, a speed estimator and a headway estimator. Both motion-based and appearance-based methods for vehicle detection are investigated. Appearance-based methods using convolutional neural networks are found to be most appropriate given the high detection accuracy requirements of the system. Headway estimation is then carried out using the detected vehicles on a video sequence. The following distance estimation is carried out using the headway and speed estimations. We also propose methods to assess the performance of the headway and speed estimation processes. The proposed monitoring system has been applied to data that we have collected using a roadside camera. The root mean square error of the headway estimation is found to be around 0.045 seconds.

Driver distraction is a major cause of road accidents. Distracting activities while driving include text messaging and talking on the phone. In this paper, we propose a robust driver distraction detection system that extracts the driver's state from the recordings of an onboard camera using Deep Learning. We consider ten driving activities, which consist of one normal driving and nine distracted driving behaviors. Nine drivers were included in the experiments, and each one was asked to perform the ten activities in naturalistic and simulated driving situations. The main feature of the proposed solution is the extraction of the driver's body parts, using deep learning-based segmentation, before performing the distraction detection and classification task. Experimental results show that the segmentation module significantly improves the classification performance. The average accuracy of the proposed solution exceeds 96% on our dataset and 95% on the public AUC dataset.