arXiv subjects

Marcello Canova

Publications and source records attributed to Marcello Canova.

At least 19 recordsLinked to original sources

arXiv2026

Conformalized Transfer Learning for Li-ion Battery State of Health Forecasting under Manufacturing and Usage Variability

Accurate forecasting of state-of-health (SOH) is essential for ensuring safe and reliable operation of lithium-ion cells. However, existing models calibrated on laboratory tests at specific conditions often fail to generalize to new cells that differ due to small manufacturing variations or operate under different conditions. To address this challenge, an uncertainty-aware transfer learning framework is proposed, combining a Long Short-Term Memory (LSTM) model with domain adaptation via Maximum Mean Discrepancy (MMD) and uncertainty quantification through Conformal Prediction (CP). The LSTM model is trained on a virtual battery dataset designed to capture real-world variability in electrode manufacturing and operating conditions. MMD aligns latent feature distributions between simulated and target domains to mitigate domain shift, while CP provides calibrated, distribution-free prediction intervals. This framework improves both the generalization and trustworthiness of SOH forecasts across heterogeneous cells.

cs.LG↗

arXiv2026

Quadratic Surrogate Attractor for Particle Swarm Optimization

This paper presents a particle swarm optimization algorithm that leverages surrogate modeling to replace the conventional global best solution with the minimum of an n-dimensional quadratic form, providing a better-conditioned dynamic attractor for the swarm. This refined convergence target, informed by the local landscape, enhances global convergence behavior and increases robustness against premature convergence and noise, while incurring only minimal computational overhead. The surrogate-augmented approach is evaluated against the standard algorithm through a numerical study on a set of benchmark optimization functions that exhibit diverse landscapes. To ensure statistical significance, 400 independent runs are conducted for each function and algorithm, and the results are analyzed based on their statistical characteristics and corresponding distributions. The quadratic surrogate attractor consistently outperforms the conventional algorithm across all tested functions. The improvement is particularly pronounced for quasi-convex functions, where the surrogate model can exploit the underlying convex-like structure of the landscape.

cs.NE↗

arXiv2025

Traffic-Aware Eco-Driving Control in CAVs via Learning-based Terminal Cost Model

Connected and Automated Vehicles (CAVs) offer significant potential for improving energy efficiency and lowering vehicle emissions through eco-driving technologies. Control algorithms in CAVs leverage look-ahead route information and Vehicle-to-Everything (V2X) communication to optimize vehicle performance. However, existing eco-driving strategies often neglect macroscopic traffic effects, such as upstream traffic jams, that occur outside the optimization horizon but significantly impact vehicle energy efficiency. This work presents a novel Neural Network (NN)-based methodology to approximate the terminal cost within a model predictive control (MPC) problem framework, explicitly incorporating upstream traffic dynamics. By incorporating traffic jams into the optimization process, the proposed traffic-aware approach yields more energy-efficient speed trajectories compared to traffic-agnostic methods, with minimal impact on travel time. The framework is scalable for real-time implementation while effectively addressing uncertainties from dynamic traffic conditions and macroscopic traffic events.

eess.SY↗

arXiv2025

Augmented Physics-Based Li-ion Battery Model via Adaptive Ensemble Sparse Learning and Conformal Prediction

Accurate electrochemical models are essential for the safe and efficient operation of lithium-ion batteries in real-world applications such as electrified vehicles and grid storage. Reduced-order models (ROM) offer a balance between fidelity and computational efficiency but often struggle to capture complex and nonlinear behaviors, such as the dynamics in the cell voltage response under high C-rate conditions. To address these limitations, this study proposes an Adaptive Ensemble Sparse Identification (AESI) framework that enhances the accuracy of reduced-order li-ion battery models by compensating for unpredictable dynamics. The approach integrates an Extended Single Particle Model (ESPM) with an evolutionary ensemble sparse learning strategy to construct a robust hybrid model. In addition, the AESI framework incorporates a conformal prediction method to provide theoretically guaranteed uncertainty quantification for voltage error dynamics, thereby improving the reliability of the model's predictions. Evaluation across diverse operating conditions shows that the hybrid model (ESPM + AESI) improves the voltage prediction accuracy, achieving mean squared error reductions of up to 46% on unseen data. Prediction reliability is further supported by conformal prediction, yielding statistically valid prediction intervals with coverage ratios of 96.85% and 97.41% for the ensemble models based on bagging and stability selection, respectively.

eess.SY↗

arXiv2025

Optimal Design of Experiment for Electrochemical Parameter Identification of Li-ion Battery via Deep Reinforcement Learning

Accurate parameter estimation in electrochemical battery models is essential for monitoring and assessing the performance of lithium-ion batteries (LiBs). This paper presents a novel approach that combines deep reinforcement learning (DRL) with an optimal experimental design (OED) framework to identify key electrochemical parameters of LiB cell models. The proposed method utilizes the twin delayed deep deterministic policy gradient (TD3) algorithm to optimize input excitation, thereby increasing the sensitivity of the system response to electrochemical parameters. The performance of this DRL-based approach is evaluated against a nonlinear model predictive control (NMPC) method and conventional tests. Results indicate that the DRL-based method provides superior information content, reflected in higher Fisher information (FI) values and lower parameter estimation errors compared to the NMPC design and conventional test practices. Additionally, the DRL approach offers a substantial reduction in experimental time and computational resources.

eess.SY↗

arXiv2025

Real-Time Optimal Design of Experiment for Parameter Identification of Li-Ion Cell Electrochemical Model

Accurately identifying the parameters of electrochemical models of li-ion battery (LiB) cells is a critical task for enhancing the fidelity and predictive ability. Traditional parameter identification methods often require extensive data collection experiments and lack adaptability in dynamic environments. This paper describes a Reinforcement Learning (RL) based approach that dynamically tailors the current profile applied to a LiB cell to optimize the parameters identifiability of the electrochemical model. The proposed framework is implemented in real-time using a Hardware-in-the-Loop (HIL) setup, which serves as a reliable testbed for evaluating the RL-based design strategy. The HIL validation confirms that the RL-based experimental design outperforms conventional test protocols used for parameter identification in terms of both reducing the modeling errors on a verification test and minimizing the duration of the experiment used for parameter identification.

eess.SY↗

arXiv2025

Modeling Driver Behavior in Speed Advisory Systems: Koopman-based Approach with Online Update

Accurate driver behavior modeling is essential for improving the interaction and cooperation of the human driver with the driver assistance system. This paper presents a novel approach for modeling the response of human drivers to visual cues provided by a speed advisory system using a Koopman-based method with online updates. The proposed method utilizes the Koopman operator to transform the nonlinear dynamics of driver-speed advisory system interactions into a linear framework, allowing for efficient real-time prediction. An online update mechanism based on Recursive Least Squares (RLS) is integrated into the Koopman-based model to ensure continuous adaptation to changes in driver behavior over time. The model is validated using data collected from a human-in-the-loop driving simulator, capturing diverse driver-specific trajectories. The results demonstrate that the offline learned Koopman-based model can closely predict driver behavior and its accuracy is further enhanced through an online update mechanism with the RLS method.

eess.SY↗

arXiv2024

Improving Low-Fidelity Models of Li-ion Batteries via Hybrid Sparse Identification of Nonlinear Dynamics

Accurate modeling of lithium ion (li-ion) batteries is essential for enhancing the safety, and efficiency of electric vehicles and renewable energy systems. This paper presents a data-inspired approach for improving the fidelity of reduced-order li-ion battery models. The proposed method combines a Genetic Algorithm with Sequentially Thresholded Ridge Regression (GA-STRidge) to identify and compensate for discrepancies between a low-fidelity model (LFM) and data generated either from testing or a high-fidelity model (HFM). The hybrid model, combining physics-based and data-driven methods, is tested across different driving cycles to demonstrate the ability to significantly reduce the voltage prediction error compared to the baseline LFM, while preserving computational efficiency. The model robustness is also evaluated under various operating conditions, showing low prediction errors and high Pearson correlation coefficients for terminal voltage in unseen environments.

eess.SY↗

arXiv2024

Data-Driven Personalized Energy Consumption Range Estimation for Plug-in Hybrid Electric Vehicles in Urban Traffic

In urban traffic environments, driver behaviors exhibit considerable diversity in vehicle operation, encompassing a range of acceleration and braking maneuvers as well as adherence to traffic regulations, such as speed limits. It is well-established that these intrinsic driving behaviors significantly influence vehicle energy consumption. Therefore, establishing a quantitative relationship between driver behavior and energy usage is essential for identifying energy-efficient driving practices and optimizing routes within urban traffic. This study introduces a data-driven approach to predict the equivalent fuel consumption of a plug-in hybrid electric vehicle (PHEV) based on an integrated model of driver behavior and vehicle energy consumption. Unlike traditional models that provide point predictions of fuel consumption, this approach uses Conformalized Quantile Regression (CQR) to offer prediction intervals that capture the variability and uncertainty in fuel consumption. These intervals reflect changes in fuel consumption, as well as variations in driver behavior, and vehicle and route conditions. To develop this model, driver-specific data were collected through a driver-in-the-loop simulator, which tested different human drivers responses. The CQR model was then trained and validated using the experimental data from the driver-in-the-loop simulator, augmented by the synthetic data generated from Monte Carlo simulations conducted using a calibrated microscopic driver behavior and vehicle energy model. The CQR model was evaluated by comparing its predictions of equivalent fuel consumption intervals with those of baseline prediction interval methods that rely solely on conformal prediction.

eess.SY↗

arXiv2024

Parameter Identification for Electrochemical Models of Lithium-Ion Batteries Using Bayesian Optimization

Efficient parameter identification of electrochemical models is crucial for accurate monitoring and control of lithium-ion cells. This process becomes challenging when applied to complex models that rely on a considerable number of interdependent parameters that affect the output response. Gradient-based and metaheuristic optimization techniques, although previously employed for this task, are limited by their lack of robustness, high computational costs, and susceptibility to local minima. In this study, Bayesian Optimization is used for tuning the dynamic parameters of an electrochemical equivalent circuit battery model (E-ECM) for a nickel-manganese-cobalt (NMC)-graphite cell. The performance of the Bayesian Optimization is compared with baseline methods based on gradient-based and metaheuristic approaches. The robustness of the parameter optimization method is tested by performing verification using an experimental drive cycle. The results indicate that Bayesian Optimization outperforms Gradient Descent and PSO optimization techniques, achieving reductions on average testing loss by 28.8% and 5.8%, respectively. Moreover, Bayesian optimization significantly reduces the variance in testing loss by 95.8% and 72.7%, respectively.

eess.SY↗

arXiv2024

Diagnosing and Decoupling the Degradation Mechanisms in Lithium Ion Cells: An Estimation Approach

Understanding battery degradation in electric vehicles (EVs) under real-world conditions remains a critical yet under-explored area of research. Central to this investigation is the challenge of estimating the specific degradation modes in aged cells with no available information on usage history, bypassing the conventional yet invasive method of tear-down tests. Using an electrochemical model, this study pioneers a methodology to decouple and isolate the aging mechanisms in batteries sourced from EVs with varying mileages. A robust correlation is established between the model parameters and distinct degradation processes, enabling the diagnosis and estimation of each mechanism's impact on the battery's parameters. This paper sheds light on battery degradation in real-world scenarios and demonstrates the feasibility of their identification, isolation, and approximate quantification of their effects.

math.OC↗

arXiv2023

Eco-Driving Control of Connected and Automated Vehicles using Neural Network based Rollout

Connected and autonomous vehicles have the potential to minimize energy consumption by optimizing the vehicle velocity and powertrain dynamics with Vehicle-to-Everything info en route. Existing deterministic and stochastic methods created to solve the eco-driving problem generally suffer from high computational and memory requirements, which makes online implementation challenging. This work proposes a hierarchical multi-horizon optimization framework implemented via a neural network. The neural network learns a full-route value function to account for the variability in route information and is then used to approximate the terminal cost in a receding horizon optimization. Simulations over real-world routes demonstrate that the proposed approach achieves comparable performance to a stochastic optimization solution obtained via reinforcement learning, while requiring no sophisticated training paradigm and negligible on-board memory.

cs.LG↗

arXiv2023

A Deep Reinforcement Learning Framework for Eco-driving in Connected and Automated Hybrid Electric Vehicles

Connected and Automated Vehicles (CAVs), in particular those with multiple power sources, have the potential to significantly reduce fuel consumption and travel time in real-world driving conditions. In particular, the Eco-driving problem seeks to design optimal speed and power usage profiles based upon look-ahead information from connectivity and advanced mapping features, to minimize the fuel consumption over a given itinerary. In this work, the Eco-driving problem is formulated as a Partially Observable Markov Decision Process (POMDP), which is then solved with a state-of-art Deep Reinforcement Learning (DRL) Actor Critic algorithm, Proximal Policy Optimization. An Eco-driving simulation environment is developed for training and evaluation purposes. To benchmark the performance of the DRL controller, a baseline controller representing the human driver, a trajectory optimization algorithm and the wait-and-see deterministic optimal solution are presented. With a minimal onboard computational requirement and a comparable travel time, the DRL controller reduces the fuel consumption by more than 17% compared against the baseline controller by modulating the vehicle velocity over the route and performing energy-efficient approach and departure at signalized intersections, over-performing the more computationally demanding trajectory optimization method

eess.SY↗

arXiv2022

Data-driven Driver Model for Speed Advisory Systems in Partially Automated Vehicles

Vehicle control algorithms exploiting connectivity and automation, such as Connected and Automated Vehicles (CAVs) or Advanced Driver Assistance Systems (ADAS), have the opportunity to improve energy savings. However, lower levels of automation involve a human-machine interaction stage, where the presence of a human driver affects the performance of the control algorithm in closed loop. This occurs for instance in the case of Eco-Driving control algorithms implemented as a velocity advisory system, where the driver is displayed an optimal speed trajectory to follow to reduce energy consumption. Achieving the control objectives relies on the human driver perfectly following the recommended speed. If the driver is unable to follow the recommended speed, a decline in energy savings and poor vehicle performance may occur. This warrants the creation of methods to model and forecast the response of a human driver when operating in the loop with a speed advisory system. This work focuses on developing a sequence to sequence long-short term memory (LSTM)-based driver behavior model that models the interaction of the human driver to a suggested desired vehicle speed trajectory in real-world conditions. A driving simulator is used for data collection and training the driver model, which is then compared against the driving data and a deterministic model. Results show close proximity of the LSTM-based model with the driving data, demonstrating that the model can be adopted as a tool to design human-centered speed advisory systems.

cs.RO↗

arXiv2022

Safe Model-based Off-policy Reinforcement Learning for Eco-Driving in Connected and Automated Hybrid Electric Vehicles

Connected and Automated Hybrid Electric Vehicles have the potential to reduce fuel consumption and travel time in real-world driving conditions. The eco-driving problem seeks to design optimal speed and power usage profiles based upon look-ahead information from connectivity and advanced mapping features. Recently, Deep Reinforcement Learning (DRL) has been applied to the eco-driving problem. While the previous studies synthesize simulators and model-free DRL to reduce online computation, this work proposes a Safe Off-policy Model-Based Reinforcement Learning algorithm for the eco-driving problem. The advantages over the existing literature are three-fold. First, the combination of off-policy learning and the use of a physics-based model improves the sample efficiency. Second, the training does not require any extrinsic rewarding mechanism for constraint satisfaction. Third, the feasibility of trajectory is guaranteed by using a safe set approximated by deep generative models. The performance of the proposed method is benchmarked against a baseline controller representing human drivers, a previously designed model-free DRL strategy, and the wait-and-see optimal solution. In simulation, the proposed algorithm leads to a policy with a higher average speed and a better fuel economy compared to the model-free agent. Compared to the baseline controller, the learned strategy reduces the fuel consumption by more than 21\% while keeping the average speed comparable.

cs.LG↗

arXiv2021

Model Predictive Control for Automotive Climate Control Systems via Value Function Approximation

Among the auxiliary loads in light-duty vehicles, the air conditioning system is the single largest energy consumer. For electrified vehicles, the impact of heating and cooling loads becomes even more significant, as they compete with the powertrain for battery energy use and can significantly reduce the range or performance. While considerable work has been made in the field of optimal energy management for electrified vehicles and optimization of vehicle velocity for eco-driving, few contributions have addressed the application of energy-optimal control for heating and cooling loads. This paper proposes an energy management strategy for the thermal management system of an electrified powertrain, based on Model Predictive Control. Starting from a nonlinear model of the vapor compression refrigeration system that captures the dynamics of the refrigerant in the heat exchangers and the power consumption of the system, a constrained multi-objective optimal control problem is formulated to reduce energy consumption while tracking a desired thermal set point. An efficient implementation of MPC is proposed for real-time applications by introducing a terminal cost obtained from the approximation of the global optimal solution.

eess.SY↗

arXiv2021

Real-time Eco-Driving Control in Electrified Connected and Autonomous Vehicles using Approximate Dynamic Programming

Connected and Automated Vehicles (CAVs), particularly those with a hybrid electric powertrain, have the potential to significantly improve vehicle energy savings in real-world driving conditions. In particular, the Eco-Driving problem seeks to design optimal speed and power usage profiles based on available information from connectivity and advanced mapping features to minimize the fuel consumption over an itinerary. This paper presents a hierarchical multi-layer Model Predictive Control (MPC) approach for improving the fuel economy of a 48V mild-hybrid powertrain in a connected vehicle environment. Approximate Dynamic Programming (ADP) is used to solve the Receding Horizon Optimal Control Problem (RHOCP), where the terminal cost for the RHOCP is approximated as the base-policy obtained from the long-term optimization. The controller was extensively tested virtually (using both deterministic and Monte Carlo simulations) across multiple real-world routes where energy savings of more than 20\% have been demonstrated. Further, the developed controller was deployed and tested at a proving ground in real-time on a test vehicle equipped with a rapid prototyping embedded controller. Real-time in-vehicle testing confirmed the energy savings observed in simulation and demonstrated the ability of the developed controller to be effective in real-time applications.

eess.SY↗

arXiv2021

Eco-Driving of Connected and Autonomous Vehicles with Sequence-to-Sequence Prediction of Target Vehicle Velocity

The Eco-Driving control problem seeks to perform fuel efficient speed planning for a Connected and Autonomous Vehicle (CAV) that can exploit information available from advanced mapping, and from Vehicle-to-Everything (V2X) communication. The ability of an Eco-Driving strategy to adapt in real time to variable traffic scenarios where surrounding vehicles can be either connected or unconnected is critical for further development and deployment of this technology in the transportation sector. In this work, the Eco-Driving strategy, formulated as a receding-horizon optimal control problem, is integrated with a target vehicle speed prediction model and solved via Dynamic Programming (DP) to determine the optimal speed trajectory in the presence of a human-driven target vehicle. An encoder-decoder architecture analyzes the patterns in the target vehicle velocity recorded over a historic window using a Gated-Recurrent-Unit (GRU) based encoder and generates an estimate of the future velocity trajectory using the GRU based decoder. A sensitivity study is done to analyze the effect of the historical and prediction windows on the accuracy of the velocity predictor. The proposed Eco-Driving controller is evaluated through microscopic simulations using a traffic simulator.

eess.SY↗