Electrochemical impedance spectroscopy (EIS) is one of the most widely used techniques for battery monitoring and characterization. However, EIS measurement is a time-consuming process, since it must be performed after the battery relaxation time interval. In this article, a method for performing fast broadband EIS during battery relaxation by compensating for the effect of the transient is proposed. The approach is based on the local rational method (LRM), which is a nonparametric frequency-domain system identification technique, and eliminates the need for long waiting time before starting the measurement process. The proposed approach is validated by numerical simulations and experiments, proving its capability of compensating the effect of the transient and outperforming other nonparametric techniques, such as the local polynomial method (LPM). In particular, experimental tests performed on a 18650 lithium-ion battery show that the proposed flexible LRM approach is capable of compensating the transient behavior and providing usable EIS estimates immediately after the battery discharge is finished. This behavior is demonstrated using a broadband multisine excitation signal of 20 s duration, spanning a frequency range from 50 mHz to 100 Hz. 

In this study we define a comprehensive method for analyzing electrochemical impedance spectra of lithium batteries using equivalent circuit models, and for information extraction on state-of-charge and state-of-health from impedance data by means of machine learning methods. Estimation of circuit parameters typically implies a non-linear optimization problem. A detailed method for estimating initial values of the optimization algorithm is described, emphasizing short computation times and efficient convergence to global minimum. Parameters identifiability is investigated through an analysis of the injectivity of the model, Cramer–Rao lower bound, profile likelihood, and sensitivity analysis. An exploratory data analysis is presented to estimate the degree of correlation between impedance spectra (or circuit parameters) and battery state-of-charge or state-of-health, prior to the implementation of any machine learning algorithm. A publicly available dataset of impedance spectra of five lithium-polymer batteries is used to test the whole procedure. Estimation of state-of-charge and state-of-health is performed by implementing Gaussian process regression.

Measuring the position of robots or other devices in challenging environments, with tight accuracy requirements and insufficient coverage from satellite navigation systems, is crucial for improving existing applications and enabling novel solutions. In this article, two of the most commonly used positioning technologies are ultrawide band (UWB) ranging systems (RSs) and magnetic RS. These technologies present limitations that can be overcome by combining them and exploiting their complementary features. Specifically, the UWB RSs provide long-range accuracy and the standard deviation in the line-of-sight (LOS) condition is distance invariant, but the accuracy degrades in short-range and non-LOS (NLOS) conditions. Contrastingly, a magnetic RS exhibits robustness to NLOS scenarios and provides high accuracy over short distances, yet does not perform as well as UWB RSs over long distances. In this article, the compatibility and complementary characteristics of these two RSs are experimentally proven for the first time. The robustness in NLOS conditions of the magnetic RS and the low dispersion with increasing distance of the UWB RS represent the complementary characteristics. This complementarity is leveraged by the proposed fusion method: the adaptive tightly coupled extended Kalman filter (ATCEKF). The proposed fusion method is experimentally characterized in mixed LOS and NLOS conditions, resulting in 6.9 cm of error in an outdoor environment. The positioning error decreased by 55% with respect to the UWB standalone positioning system and 32% with respect to the magnetic standalone positioning system. Therefore, the proposed fusion method improves the robustness and accuracy of position estimation. These results are also confirmed by an additional experiment. In it, the new small-sized and battery-powered mobile node are implemented using a Raspberry Pi 4 Model B.

Estimating the state of charge (SOC) of batteries is fundamental for the proper management and safe operation of numerous systems, including electric vehicles, smart energy grids, and portable electronics. While there is no practical method for direct measurement of SOC, several estimation approaches have been developed, including a growing number of machine-learning-based techniques. Machine learning methods are intrinsically data-driven but can also benefit from a-priori knowledge embedded in a model. In this work, we first demonstrate, through exploratory data analysis, that it is possible to discriminate between different SOC from electrochemical impedance spectroscopy (EIS) measurements. Then we propose a SOC estimation approach based on EIS and an equivalent circuit model to provide a compact way to describe the frequency domain and time-domain behavior of the impedance of a battery. We experimentally validated this approach by applying it to a dataset consisting of EIS measurements performed on four lithium-ion cylindrical cells at different SOC values. The proposed approach allows for very efficient model training and produces a low-dimensional SOC classification model that achieves above 93% accuracy. The resulting low-dimensional classification model is suitable for embedding into battery-powered systems and for online SOC estimation.