This paper proposes a Long Short-Term Memory (LSTM) based approach for accurate State Of Charge (SOC) and State Of Health (SOH) estimation in lithium-ion batteries, which is critical for improving the safety and longevity of electric vehicles. The nonlinear dynamics of batteries, influenced by factors such as temperature, voltage, current, and SOC, poses significant challenges to traditional estimation methods. Using the Long-Term Degradation dataset, our LSTM model captures temporal dependencies and complex electrochemical interactions to predict SOC and SOH under varying operating conditions. The experimental results demonstrate robust performance, with mean squared errors as low as 5.3121×10−5for the estimation of SOC and 5.572×10−5for the estimation of SOH for different current profiles. The proposed framework provides a scalable solution for real-time battery management systems, reducing the reliance on manual feature extraction and enabling generalization across different battery technologies.

This paper investigates the classification of driving styles using unsupervised learning techniques applied to recorded driving data. The study focuses on identifying two primary driving styles: calm and aggressive. The importance of lane change scenarios in discriminating between these styles is highlighted, using features such as lateral speeds and yaw angles. Using spectral clustering and K-means algorithms, a driving style detection method is proposed. The obtained results indicate that K-means outperforms spectral clustering in effectively classifying drivers based on their behaviour, particularly in lane change situations. This research contributes to a deeper understanding of driver behaviour on the road and provides insights into the potential applications of unsupervised learning in driving style recognition.

The need to achieve high performance with very high control accuracy in an electrical machine is paramount and has led to increased interest in modern motor control techniques. Modern motor techniques often require position sensors to be integrated into electrical machines. The position estimation approach offers a cost advantage within acceptable error results over the position measurement approach, making the position estimation approach more suitable. This paper deals with sensorless position control of a synchronous reluctance machine (SynRM) by applying the load torque of an electric vehicle drive train based on an extended Kalman filter (EKF). The developed observer could be an excellent replacement for the analogue position sensor. This work is done in two steps: Firstly, the EKF observer is synthesised and secondly, two types of controllers are used for velocity and currents; the PI and the super-twisting controllers. The simulation results obtained in this work confirm the accuracy and reliability of the developed observer with the super-twisting controllers.

This work aims at solving the torque ripple effect problem minimization in synchronous reluctance machines. The objective is realized by controlling the currents in the synchronous reluctance machine in respect to the reference current and as well, leads to the control of the torque of the machine thereby; reducing the torque ripple effect. In the first, Field -Oriented Control (FOC) is used to control the current of the machine in respect to the reference current. In the second, the Maximum Torque Per Ampere MTPA as well as the Optimal Current control approach are applied for the same purpose. The obtained results are analyzed and the best architecture is selected with the use of a Proportional Integral (PI) controller.

This review discusses the importance of detecting driver emotions to improve driving safety and user experience. The article presents recent literature on emotion recognition in the context of driving, reviewing different models of emotion representation, recent public databases for driver emotion recognition, and various uni-modal and multi-modal methods to detect driver emotions. The study shows that detecting the driver's emotional state and its intensity is vital to improving driving safety and the user experience, particularly if the emotion is disconnected from or not related to the driving task.

A Passive Fault Tolerant Control (PFTC) based on deep Reinforcement Learning (RL) approach has been proposed in this work. The RL agent is responsible for controlling a heating system in order to regulate the indoor temperature of an area. During operation, faults may occur in the heating system, particularly at the level of the internal resistance of the boiler and the pump responsible for the flow of water injected into the pipes to heat the area. The RL agent has been trained without the presence of a fault either in a holy environment and the goal is to see his behavior when the fault appears. The agent manages even without prior training to adapt and propose a PFTC to guarantee an interior temperature faithful to the set temperature while ensuring minimizing the energy consumed and good use of the heating system to guarantee a long equipment life.

The main objective of this work is to apply rule-based energy management algorithm between two sources and a load using two interleaved floating DC/DC converters dedicated to vehicular applications; the first is a two-way converter, the second is a one-way converter. These two converters are combined to interface the hybrid power which is composed of a fuel cell as the primary source for the vehicle and a storage system (supercapacitor) as a secondary source. This article provides a modelling of the two proposed DC-DC converters as well as a control strategy for each of them. However, the studied system being powered by two sources, it is essential to apply an energy management algorithm which must ensure the distribution of the energy flows between the sources and the load in order to satisfy the required power. A rule based algorithm is applied as well to generate source power references considering state-of-charge limitation to avoid overcharging and deep discharging in order to decrease the lifetime of the supercapacitor. To evaluate the performance of the studied converters, simulation results of the hybrid system with the energy management algorithm are presented.

Energy consumption in buildings has become a real concern for scientists and seeking to reduce this consumption is essential. Heating, ventilation, and air conditioning (HVAC) systems account for more than 50% of this consumption. One of the solutions to reduce this excessive consumption is to detect and diagnose faults that can appear instantaneously and quickly with fault diagnostic detection systems (FDD) based on artificial intelligence. The paper presents a strategy based on a data-driven approach for the detection and diagnosis of sensor faults that may appear in the Air Handling Unit (AHU) systems. A Decision Tree, Random Forest and SVM algorithm were used to detect and diagnose temperature sensor faults occurring in the AHU. The comparison between these methods shows that the Random Forest gives the best result with 96% accuracy.

Nowadays, electric vehicles represent one of the most significant chances to reduce the pollution production rate. Unfortunately, in electric motors, the efficiency decreases by the relationship between speed proposed by the driver and the torque required by the vehicle. Those parameters can be estimated in order to make an efficiency optimization based on present and future road/weather conditions. Regrettably, this kind of control (optimal control) requires a model with low compilation time. Since bicycle motorcycle has nonlinearities, in this article, a state reduced linear dynamic model able to reproduce the behavior of a TWEV by more than one minute will be proposed. The model is oriented to an optimal controller in energetic field, for this reason, the most significant states are longitudinal speed to be used with the input torque to calculate the efficiency of the electric motor and the Yaw angle to creates constraints over the trajectory that has to be covered. After the model is proposed, its accuracy is tested by a comparison with a numerical iteration software. Finally, a sensitivity test is made in order to determine the behavior of the error according to the friction coefficients of the rear and front pneumatics.

This paper deals with a global study of numerical control implementation applied to a DC-DC buck converter with two techniques, classical control technique (PI control) and robust control technique (H-infinity loop-shaping control). The H-infinity loop-shaping control is applied to improve the performance and the stability robustness of this power converter based on pulse-width-modulation (PWM) techniques. The experimental results are very promising. They show that H-infinity loop-shaping control performs better than PI control. H-infinity loop-shaping control reject disturbance and noise produce by the DC-DC power converter. In addition to experimental validation, the other added value of this paper is to demonstrate the feasibility of easily implementing robust numerical controls allowing rejecting disturbances without real impact on computing time, which is very important for embedded power converter applications.

The studies dedicated to the battery open-circuit-voltage (OCV) online estimation are not as much as the research efforts on the state-of-charge (SOC) determination and the parameter identification such as capacity and resistance. However, as an important term that represents the distinct characteristic of different Lithium-Ion batteries, OCV should also be estimated. Four estimation algorithms, namely, Luenberger observer, Kalman filter, Recursive least-square with forgetting factor and Recursive least-square with variable forgetting factor are selected and compared in terms of estimation accuracy, computational cost, parameter tuning and robustness to model parameters variations. Simulation results have shown that observer-based methods exhibit better estimation performances than regression-based ones.

The energy required by a two wheels electric vehicle (TWEV) to complete a trip is lower than common electric cars or internal combustion vehicles. However, there are considerable losses along the electric driving chain. Those losses added to a limited energy storage cause an impact over the TWEV autonomy. This appears to be the main factor, which limits the large-scale market penetration of TWEV. This paper aims to analyze the multiphysic behavior of the complete power-chain in order to study its effect on its energetic losses. Even when many dynamics model oriented to hardware design approach can represent the come multiphysic behavior of one or two elements of the power train, the approach proposed in this paper presents a balanced representation of all power chain able to be used in real-time optimization. This study will help to improve the capabilities of an onboard TWEV efficiency estimator system which uses a longitudinal force model. As a conclusion, the error of autonomy estimation is compared with thermal considerations and without them according to different operating points.

The expectations regarding autonomous vehicles are very high to transform the future mobility and ensure more road safety. Autonomous driving system should be able in the short term to detect dangerous situations and respond appropriately and thus increase driving safety. Understanding the intentions of drivers has recently received growing interest. A long-term prediction method based on gated unit-recurrent neural network model is proposed for the problem of trajectory prediction of surrounding vehicles. A deep neural network with Long-short term memory (LSTM) and Gated Recurrent Units (GRU) structure is used to analyze the spatial-temporal features of the past trajectory. Through sequences learning, the system generates the future trajectory of other traffic participants for different horizons of prediction. We evaluate all models with standard metric (Root mean square error RMSE), loss function convergence and processing time. After comparing the different models, our experiments revealed that the proposed GRU based models is indeed better than LSTM based models in term of accuracy and processing speed.

Battery short circuit (SC), including both internal short circuit (ISC) and external short circuit (ESC), is an important stage before thermal runaway (TR). Therefore, on-line incipient battery SC detection is of vital importance to guarantee a safe and reliable operation of Lithium-Ion Batteries (LIBs). In this paper, based on a slightly modified battery equivalent circuit model (ECM), the purpose of battery incipient SC detection is achieved from the perspective of fault estimation. The proposed soft SC diagnosis method is independent of battery intrinsic properties as much as possible. A robust fault estimator, under the form of proportional-integral observer, is designed by solving two linear-matrix-inequality (LMI) constraints. Simulation studies based on an A123-M1 cell have verified the effectiveness of the proposed SC detection method.

This paper deals with a comparison study between two hybrid systems composed by batteries with fuel cell and super capacitors with fuel cell. A sizing algorithm is used to define the optimal sizes of the hybrid source. The comparison between the two systems is based on the weight, volume and cost. The batteries prove best performances in case of sizing according to the energy. However, the super capacitors provide the optimal sizes of the hybrid source in case of sizing according to the total recovered power. It is noted that the hybridization of the HP batteries with the fuel cell is an interesting solution to make up the Energy Storage System for automotive applications with high drive range.

A fault-tolerant control design strategy based upon sliding mode control and a descriptor observer for Upschitz system is presented. The aim is to mitigate the vehicle speed faults and provide accurate measurements for the control. Thus, the additive faults affecting the vehicle speed sensor are estimated, and accurate speed measurments are used to control the vehicle speed instead of the faulty ones. Sufficient conditions and obsrever gain are designed by use of Lyapunov theory, satisfying L2-gain norm and H∞enterions. These conditions are derived under the well known Linear Matrix Inequality. The optimal designed gains ensure robustness against disturbances and additive sensor faults. A nonlinear longitudinal vehicle dynamic is considered to demonstrate the performance of the proposed design to achieve the spacing control task. Computer simulations are addressed to validate the proposed controller in autonomous vehicle scenario.

This paper focalizes on sizing of hybrids sources composed with Fuel cells FCs and battery pack. Also an experimental validation of energy management of FC/Battery electric vehicle is tested. The system is composed of a fuel cell system as the main source and batteries as an assisting one. This last one is connected to a bidirectional DC/DC converter, and a DC/DC boost converter is associated to the fuel cell stack. To increase the power efficiency and to achieve the best performances of the hybrid source, an online EM strategy is used to share the power between the main and the auxiliary source by determining the power profile of each one. This strategy is based on frequency separation; it takes into account the slow dynamics of FC, fuel consumption and the batteries limits. Also, it is needed to make these results as a reference to be compared with other strategies which are currently under development in our laboratory. In the objective to verify the efficiency of the proposed approach, both simulation and experimental results leads to confirm its efficiency, the robustness and stability regrading dynamic performances during power demand, and regenerative braking, fuel consumption.

The paper proposes and evaluates an experimental passive sensor fault tolerant control for a gyroscope system. The sensor fault occurrence reduces the performance and may even cause the instability. This work focuses on developing fault tolerant control when these drawbacks are occurred. A passive fault tolerant control (PFTC) scheme is developed to counteract a sensor failure and parameters uncertainties. A second sliding mode FTC strategy based on super twisting algorithm ensures the stability robustness of the gyroscope system in the presence of the additive faults. For this purpose, the passive fault tolerant control (PFTC) approach is designed to preserve the stability and to maintain an acceptable performance when the sensor failure appears. The effectiveness of the proposal fault tolerant control strategy is validated by simulation and experiment results in presence of the sensor fault.

In this paper, we present a comparative study of four voting algorithms for two Induction Motor drive fault tolerant control to speed sensor failure schemes. In both Output and Input Fault Tolerant controller, the voting algorithm chooses the most appropriate output signal to ensure the best behavior in degraded mode. The performances are evaluated through simulations of a 7.5kW Induction Motor drive with robustness testing against parametric variations, as well as under load testing. The results show that Euler, Newton-Raphson and Maximum Likelihood voting algorithms are more efficient than the Weighted Average in both Fault Tolerant Control schemes.

This paper presents a vehicle transmission simulator coupled to a longitudinal vehicle dynamic model. The transmission simulator uses electric actuators to reproduce the mechanical characteristics of a real vehicle engine and its transmission chain. The developed approach allows to validate transmission and vehicle dynamic studies (control of automatic or robotized gearboxes, test of heat engine, dynamic behavior and passenger comfort...) without need to the real transmission system and the real environment of the vehicle. The proposed system is used to carry out a vehicle automated driving in the case of a car- following operation.

This paper presents a new architecture of control for microteleoperation system using a fault tolerant control (FTC) strategy to compensate for the degradation effects of microgipping system. The proposed strategy uses passivity approach for the bilateral controller and robust fault tolerant control (FTC) for the two-fingered microgripper system. First, the bilateral controller architecture uses the passivity approach for the teleoperation system in the presence of time delay and scaling factor variations. Second, the FTC controller is designed in such a way that the performance and robustness may be done separately which has the potential to overcome the conflict between performance and robustness in the traditional feedback framework. The slave controller architecture is controlled by the PI controller for a nominal model and the Hinfincontroller for faulty operation. The simulation and experimental results show clearly the effectiveness of the proposed approach.

In this paper, a bilateral controller for a microteleoperation system is presented using passivity approaches. We showed that the application of wave variable formalism allows the passivity of the system in spite of the communication delays between the master and the slave and the varying scaling factors. Conditions of the passivity and of the transparency are given for the micro teleoperation system. Finally, experimental results are presented showing the stability-transparency performances of the resulting system with constant and variable time-delay

This paper presents a force-reflecting macro-micro teleoperator operating in a remote microenvironment with transmission time delays. In order to manipulate micro-objects, it is inevitable to consider the scaling effect problem between worlds with different physical characteristics, with minimal loss of physical information and robustness against variation of scaling gains. Accordingly, the controllers are designed based on the framework H/sub /spl infin// loop shaping procedure approach. This approach allows a convenient means to trade-off the robustness for a pre-specified time-delay margin, variation of force scaling factors and uncertainties in master and slave modeling. The validity of the proposed method is demonstrated by simulations.

Developments on micro/nano manipulation and Internet technologies allow potential impacts on networked-micro-automation of small industrial products. However, reliable and efficient tele-micromanipulation systems with haptic feedback over the Internet face strong problems due to the nonlinear nature of the microenvironment and time-varying delays in communication lines. Towards this end, this paper presents a robust bilateral controller design using H/sub /spl infin//-optimal control and /spl mu/-synthesis frameworks. This approach allows a convenient means to tradeoff the optimization of various performance criteria (micro scale force/position) and robustness for a prespecified time-delay margin and force scaling factors. The validity of the proposed method is demonstrated by simulations and experiments for a pick-and-place micromanipulation task.

This study aims to develop a force-reflecting macromicro teleoperator with different scaled worlds. A bilateral control system for scaled teleoperation provides the human operator with a feel of the task at the micro-scale. However, reliable and efficient tele-micromanipulation systems with haptic feedback over the Internet face to strong problems due to the nonlinear nature of microenvironment and time-varying delays in communication lines. A robust bilateral controller design framework using H/sub 2/-optimal control approach is proposed. A comparative study for assessing the robustness against time-varying delays is performed through two different designs, i.e., a /spl mu/-synthesis framework and a Pade approximation. The proposed approaches allow a convenient means to tradeoff the robustness for a pre-specified time-delay margin. The validity of the proposed method is demonstrated by simulations.