Search Results

The design of transportation vehicles, whether passenger or commercial, typically involves a lengthy process from concept to prototype and eventual manufacture. To improve competitiveness, original equipment manufacturers are continually exploring ways to shorten the design process. The application of digital tools such as computer-aided-design and computer-aided-engineering, as well as model-based computer simulation enable team members to virtually design and evaluate ideas within realistic operating environments. Recent advances in machine learning (ML)/artificial intelligence (AI) can be integrated into this paradigm to shorten the initial design sequence through the creation of digital agents. A digital agent can intelligently explore the design space to identify promising component features which can be collectively assessed within a virtual vehicle simulation.

Driving style and skill classification are of great significance in human-oriented advanced driver-assistance system (ADAS) development. In this paper, we propose Fully-Connected Deep Neural Networks (FC-DNN) to classify drivers’ styles and skills with naturalistic driving data. Followed by the data collection and pre-processing, FC-DNN with a series of deep learning optimization algorithms are applied. In the experimental part, the proposed model is validated and compared with other commonly used supervised learning methods including the k-nearest neighbors (KNN), support vector machine (SVM), decision tree (DT), random forest (RF), and multilayer perceptron (MLP). The results show that the proposed model has a higher Macro F1 score than other methods. In addition, we discussed the effect of different time window sizes on experimental results. The results show that the driving information of 1s can improve the final evaluation score of the model.

This paper proposes a novel reconfigurable battery balancing topology and reinforcement learning-based intelligent balancing management system. The different degradations cause a significant loss of battery pack available capacity, as the pack power output relies on the weakest cell due to the relevant physical requirements. To handle this capacity drop issue, a reconfigurable battery topology is adopted to improve the usability of the heterogeneous battery. There are some existing battery reconfigurable topologies in the literature. However, these studies rely on the limited options of topology designs, and there is a lack of study on the reconfigurability of these designs and other possible new designs. Also, it is rare to find an optimal management system for the reconfigurable battery topology. To fill these research gaps, this paper explores existing battery reconfigurable topology designs and proposes a new reconfigurable topology for battery balancing.

The integration of sensors, actuators, and real-time control in transportation systems enables intelligent system operation to minimize energy consumption and maximize occupant safety and vehicle reliability. The operating cycle of military ground vehicles can be on- and off-road in harsh weather and adversarial environments, which demands continuous subsystem functionality to fulfill missions. Onboard diagnostic systems can alert the operator of a degraded operation once established fault thresholds are exceeded. An opportunity exists to estimate vehicle maintenance needs using model-based predicted trends and eventually compiled information from fleet operating databases. A digital twin, created to virtually describe the dynamic behavior of a physical system using computer-mathematical models, can estimate the system behavior based on current and future operating scenarios while accounting for past effects.

In this paper, a reinforcement learning-based method is proposed to adapt autonomous vehicle passengers’ expectation of comfort through in-situ human-vehicle interaction. Ride comfort has a significant influence on the user’s experience and thus acceptance of autonomous vehicles. There is plenty of research about the motion planning and control of autonomous vehicles. However, limited studies have explicitly considered the comfort of passengers in autonomous vehicles. This paper studies the comfort of humans in autonomous vehicles longitudinal autonomous driving. The paper models and then improves passengers’ feelings about autonomous driving behaviors. This proposed approach builds a control and adaptation strategy based on reinforcement learning using human’s in-situ feedback on autonomous driving. It also proposes an adaptation of humans to autonomous vehicles to account for improper human driving expectations.

Integrated modeling of vehicle, tire and terrain is a fundamental challenge to be addressed for off-road autonomous navigation. The complexities arise due to lack of tools and techniques to predict the continuously varying terrain and environmental conditions and the resultant non-linearities. The solution to this challenge can now be found in the plethora of data driven modeling and control techniques that have gained traction in the last decade. Data driven modeling and control techniques rely on the system’s repeated interaction with the environment to generate a lot of data and then use a function approximator to fit a model for the physical system with the data. Getting good quality and quantity of data may involve extensive experimentation with the physical system impacting developer’s resource. The process is computationally expensive, and the overhead time required is high.

Modeling uncertainties pose a significant challenge in the development and deployment of model-based vehicle control systems. Most model- based automotive control systems require the use of a well estimated vehicle dynamics prediction model. The ability of first principles-based models to represent vehicle behavior becomes limited under complex scenarios due to underlying rigid physical assumptions. Additionally, the increasing complexity of these models to meet ever-increasing fidelity requirements presents challenges for obtaining analytical solutions as well as control design. Alternatively, deterministic data driven techniques including but not limited to deep neural networks, polynomial regression, Sparse Identification of Nonlinear Dynamics (SINDy) have been deployed for vehicle dynamics system identification and prediction.

Learning only from a small set of examples remains a huge challenge in machine learning. Despite recent breakthroughs in the applications of neural networks, the applicability of these techniques has been limited by the requirement for large amounts of training data. What’s more, the standard supervised machine learning method does not provide a satisfactory solution for learning new concepts from little data. However, the ability to learn enough information from few samples has been demonstrated in humans. This suggests that humans may make use of prior knowledge of a previously learned model when learning new ones on a small amount of training examples. In the area of autonomous driving, the model learns to drive the vehicle with training data from humans, and most machine learning based control algorithms require training on very large datasets. Collecting and constructing training data set takes a huge amount of time and needs specific knowledge to gather relevant information.

Recent trends in autonomy have emphasized end-to-end deep-learning-based methods that have shown a lot of promise in overcoming the requirements and limitations of feature-engineering. However, while promising, the black-box nature of deep-learning frameworks now exacerbates the need for testing with end-to-end deployments. Further, as exemplars of systems-of-systems, autonomous vehicles (AVs) engender numerous interconnected component-, subsystem and system-level interactions. The ensuing complexity creates challenges for verification and validation at the various component, subsystem- and system-levels as well as end-to-end testing. While simulation-based testing is one promising avenue, oftentimes the lack of adequate fidelity of AV and environmental modeling limits the generalizability. In contrast, full-scale AV testing presents the usual limitations of time-, space-, and cost.

Semiconductor devices in electric vehicle (EV) motor drive systems are considered the most fragile components with a high occurrence rate for open circuit fault (OCF). Various signal-based and model-based methods with explicit mathematical models have been previously published for OCF diagnosis. However, this proposed work presents a model-free machine learning (ML) approach for a single-switch OCF detection and localization (DaL) for a two-level, three-phase inverter. Compared to already available ML models with complex feature extraction methods in the literature, a new and simple way to extract OCF feature data with sufficient classification accuracy is proposed. In this regard, the inherent property of active thermal management (ATM) based model predictive control (MPC) to quantify the conduction losses for each semiconductor device in a power converter is integrated with an ML network.

This study investigates the use of machine learning methods for the selection of energy storage devices in military electrified vehicles. Powertrain electrification relies on proper selection of energy storage devices, in terms of chemistry, size, energy density, and power density, etc. Military vehicles largely vary in terms of weight, acceleration requirements, operating road environment, mission, etc. This study aims to assist the energy storage device selection for military vehicles using the data-drive approach. We use Machine Learning models to extract relationships between vehicle characteristics and requirements and the corresponding energy storage devices. After the training, the machine learning models can predict the ideal energy storage devices given the target vehicles design parameters as inputs. The predicted ideal energy storage devices can be treated as the initial design and modifications to that are made based on the validation results.

It’s hard to achieve complete self-driving using hand-crafting generalized decision-making rules, while the end-to-end self-driving system is low in complexity, does not require hand-crafting rules, and can deal with complex situations. Modular-based self-driving systems require multi-task fusion and high-precision maps, resulting in high system complexity and increased costs. In end-to-end self-driving, we usually only use camera to obtain scene status information, so image processing is very important. Numerous deep learning applications benefit from multi-task learning, as the multi-task learning can accelerate model training and improve accuracy with combine all tasks into one model, which reduces the amount of calculation and allows these systems to run in real-time. Therefore, the approach of obtaining rich scene state information based on multi-task learning is very attractive. In this paper, we propose an approach to multi-task learning for semantics, geometry and motion.

In this research, the interior noise of a passenger car was measured, and the sound quality metrics including sound pressure level, loudness, sharpness, and roughness were calculated. An artificial neural network was designed to successfully apply on automotive interior noise as well as numerous different fields of technology which aim to overcome difficulties of experimentations and save cost, time and workforce. Sound pressure level, loudness, sharpness, and roughness were estimated by using the artificial neural network designed by using the experiment values. The predicted values and experiment results are compared. The comparison results show that the realized artificial intelligence model is an appropriate model to estimate the sound quality of the automotive interior noise. The reliability value is calculated as 0.9995 by using statistical analysis.

To reduce the interference and conflict of human-machine cooperative control, lighten the operation workload of drivers, and improve the friendliness and acceptability of intelligent vehicles, a personalized human-machine cooperative lane-change trajectory tracking control method was proposed. First, a lane-changing driving data acquisition test was carried out to collect different driving behaviors of different drivers and form the data pool for the machine learning method. Two typical driving behaviors from an aggressive driver and a moderate driver are selected to be studied. Then, a control structure combined by feedforward and feedback control based on Long Short Term Memory (LSTM) and model-based optimum control was introduced. LSTM is a machine learning method that has the ability of memory. It is used to capture the lane-changing behaviors of each driver to achieve personalization. For each driver, a specific personalized controller is trained using his driving data.

Artificial intelligence systems are highly accepted as a technology to offer an alternative way to tackle complex and non-linear problems. They can learn from data, and they are able to handle noisy and incomplete data. Once trained, they can perform prediction and generalization at high speed. The aim of the present study is to propose a novel approach utilizing the adaptive neuro-fuzzy inference system (ANFIS) and the fuzzy clustering method for automotive ride performance estimation. This study investigated the relationship between the automotive ride performance and relative parameters including speed, spring stiffness, damper coefficients, ratios of sprung and unsprung mass. A Takagi-Sugeno fuzzy inference system associated with artificial neuro network was employed. The C-mean fuzzy clustering method was used for grouping the data and identifying membership functions.

Range anxiety and lack of adequate access to fast charging are proving to be important impediments to electric vehicle (EV) adoption. While many techniques to fast charging EV batteries (model-based & model-free) have been developed, they have focused on a single Lithium-ion cell. Extensions to battery packs are scarce, often considering simplified architectures (e.g., series-connected) for ease of modeling. Computational considerations have also restricted fast-charging simulations to small battery packs, e.g., four cells (for both series and parallel connected cells). Hence, in this paper, we pursue a model-free approach based on reinforcement learning (RL) to fast charge a large battery pack (comprising 444 cells). Each cell is characterized by an equivalent circuit model coupled with a second-order lumped thermal model to simulate the battery behavior. After training the underlying RL, the developed model will be straightforward to implement with low computational complexity.

Roundabouts are one of the most complex traffic environments in urban roads, and a key challenge for intelligent driving decision-making. Deep reinforcement learning, as an emerging solution for intelligent driving decisions, has the advantage of avoiding complex algorithm design and sustainable iteration. For the decision difficulty in roundabout scenarios, this paper proposes an artificial potential field based Soft Actor-Critic (APF-SAC) algorithm. Firstly, based on the Carla simulator and Gym framework, a reinforcement learning simulation system for roundabout driving is built. Secondly, to reduce reinforcement learning exploration difficulty, global path planning and path smoothing algorithms are designed to generate and optimize the path to guide the agent.

Autonomous driving technology, as the product of the fifth stage of the information technology revolution, is of great significance for improving urban traffic and environmentally friendly sustainable development. Autonomous driving can be divided into three main modules. The input of the decision module is the perception information from the perception module and the output of the control strategy to the control module. The deep reinforcement learning method proposes an end-to-end decision-making system design scheme. This paper adopts the Deep Deterministic Policy Gradient Algorithm (DDPG) that incorporates the Priority Experience Playback (PER) method. The framework of the algorithm is based on the actor-critic network structure model. The model takes the continuously acquired perception information as input and the continuous control of the vehicle as output.

Aiming at the problems of long lane change intention recognition, complicated lane change model, and huge amount of processing data in the current research, this paper uses principal component analysis to improve the driver’s lane change intention recognition model using traditional pattern recognition. Firstly collect 7 parameters including driver operation and vehicle running characteristics. After data standardization and PCA (principal component analysis), the top three principal components that can reflect the information content of the original data are nearly 90%. Then, a lane-change intent recognition model based on GMM-HMM was established, three lane change intents cannot be directly observed as the hidden state of the model; and three principal component quantities obtained through linear changes are used as observational measurements.

Under complex and extreme operating conditions, the road adhesion coefficient emerges as a critical state parameter for tire force analysis and vehicle dynamics control. In contrast to model-based estimation methods, intelligent tire technology enables the real-time feedback of tire-road interaction information to the vehicle control system. This paper proposes an approach that integrates intelligent tire systems with machine learning to acquire precise road adhesion coefficients for vehicles. Firstly, taking into account the driving conditions, sensor selection is conducted to develop an intelligent tire hardware acquisition system based on MEMS (Micro-Electro-Mechanical Systems) three-axis acceleration sensors, utilizing a simplified hardware structure and wireless transmission mode. Secondly, through the collection of real vehicle experiment data on different road surfaces, a dataset is gathered for machine learning training.