arXiv subjects

Qingfeng Li

Publications and source records attributed to Qingfeng Li.

At least 37 records · Page 2Linked to original sources

arXiv2024

NID-SLAM: Neural Implicit Representation-based RGB-D SLAM in dynamic environments

Neural implicit representations have been explored to enhance visual SLAM algorithms, especially in providing high-fidelity dense map. Existing methods operate robustly in static scenes but struggle with the disruption caused by moving objects. In this paper we present NID-SLAM, which significantly improves the performance of neural SLAM in dynamic environments. We propose a new approach to enhance inaccurate regions in semantic masks, particularly in marginal areas. Utilizing the geometric information present in depth images, this method enables accurate removal of dynamic objects, thereby reducing the probability of camera drift. Additionally, we introduce a keyframe selection strategy for dynamic scenes, which enhances camera tracking robustness against large-scale objects and improves the efficiency of mapping. Experiments on publicly available RGB-D datasets demonstrate that our method outperforms competitive neural SLAM approaches in tracking accuracy and mapping quality in dynamic environments.

cs.RO↗

arXiv2023

Importance of physical information on the prediction of heavy-ion fusion cross section with machine learning

In this work, the Light Gradient Boosting Machine (LightGBM), which is a modern decision tree based machine-learning algorithm, is used to study the fusion cross section (CS) of heavy-ion reaction. Several basic quantities (e.g., mass number and proton number of projectile and target) and the CS obtained from phenomenological formula are fed into the LightGBM algorithm to predict the CS. It is found that, on the validation set, the mean absolute error (MAE) which measures the average magnitude of the absolute difference between $log_{10}$ of the predicted CS and experimental CS is 0.129 by only using the basic quantities as the input, this value is smaller than 0.154 obtained from the empirical coupled channel model. MAE can be further reduced to 0.08 by including an physical-informed input feature. The MAE on the test set (it consists of 280 data points from 18 reaction systems that not included in the training set) is about 0.19 and 0.53 by including and excluding the physical-informed feature, respectively. We further verify the LightGBM predictions by comparing the CS of $^{ 40,48}{\rm Ca }$+$^{78}{\rm Ni}$ obtained from the density-constrained time-dependent Hartree-Fock approach. Our study demonstrates the importance of physical information in predicting fusion cross section of heavy-ion reaction with machine learning.

nucl-th↗

arXiv2023

Effective field theory with resonant P-wave interaction

A new effective field theory has been developed to describe shallow $P$-wave resonances using nonlocal, momentum-dependent two-body potentials. This approach is expected to facilitate many-body calculations and has been demonstrated to converge and to be renormalizable in perturbative calculations at subleading orders. The theory has been applied to the neutron-alpha system, with good agreement found between its predictions and a phase-shift analysis of neutron-alpha elastic scattering. In the three-body system consisting of two neutrons and an alpha particle, the nonlocal potential in this framework has been found to recover the same qualitative features as previously shown with energy-dependent formulations.

nucl-th↗

arXiv2023

Studies of nuclear equation of state with the HIRFL-CSR external-target experiment

The HIRFL-CSR external-target experiment (CEE) under construction is expected to provide novel opportunities to the studies of the thermodynamic properties, namely the equation of state of nuclear matter (nEOS) with heavy ion collisions at a few hundreds MeV/u beam energies. Based on Geant 4 packages, the fast simulations of the detector responses to the collision events generated using transport model are conducted. The overall performance of CEE, including spatial resolution of hits, momentum resolution of tracks and particle identification ability has been investigated. Various observables proposed to probe the nEOS, such as the production of light clusters, $\rm t/^3He$ yield ratio, the radial flow, $π^{-}/π^{+}$ yield ratio and the neutral kaon yields, have been reconstructed. The feasibility of studying nEOS beyond the saturation density via the aforementioned observables to be measured with CEE has been demonstrated.

nucl-ex↗

arXiv2023

Machine learning transforms the inference of the nuclear equation of state

Our knowledge of the properties of dense nuclear matter is usually obtained indirectly via nuclear experiments, astrophysical observations, and nuclear theory calculations. Advancing our understanding of the nuclear equation of state (EOS, which is one of the most important properties and of central interest in nuclear physics) has relied on various data produced from experiments and calculations. We review how machine learning is revolutionizing the way we extract EOS from these data, and summarize the challenges and opportunities that come with the use of machine learning.

nucl-th↗

arXiv2023

Machine learning in nuclear physics at low and intermediate energies

Machine learning is becoming a new paradigm for scientific research in various research fields due to its exciting and powerful capability of modeling tools used for big-data processing task. In this mini-review, we first briefly introduce different methodologies of the machine learning algorithms and techniques. As a snapshot of many applications by machine learning, some selected applications are presented especially for low and intermediate energy nuclear physics, which include topics on theoretical applications in nuclear structure, nuclear reactions, properties of nuclear matter as well as experimental applications in event identification/reconstruction, complex system control and firmware performance. Finally, we also give a brief summary and outlook on the possible directions of using machine learning in low-intermediate energy nuclear physics and possible improvements in ML algorithms.

nucl-th↗

arXiv2023

Impacts of momentum dependent interaction, symmetry energy and near-threshold $NN\to NΔ$ cross sections on isospin sensitive flow and pion observables

Based on the ultra-relativistic quantum molecular dynamics (UrQMD) model, the impacts of momentum dependent interaction, symmetry energy and near-threshold $NN\to NΔ$ cross sections on isospin sensitive collective flow and pion observables are investigated. Our results confirm that the elliptic flow of neutrons and charged particles, i.e. $v_2^n$ and $v_2^{ch}$, are sensitive to the strength of momentum dependence interaction and the elliptic flow ratio, i.e., $v_2^n/v_2^{ch}$, is sensitive to the stiffness of symmetry energy. For describing the pion multiplicity near the threshold energy, accurate $NN\to NΔ$ cross sections are crucial. With the updated momentum dependent interaction and $NN\to NΔ$ cross sections in UrQMD model, seven observables, such as directed flow and elliptic flow of neutrons and charged particles, the elliptic flow ratio of neutrons to charged particles, charged pion multiplicity and its ratio $π^-/π^+$, can be well described by the parameter sets with the slope of symmetry energy from 5 MeV to 70 MeV. To describe the constraints of symmetry energy at the densities probed by the collective flow and pion observables, the named characteristic density is investigated and used. Our analysis found that the flow characteristic density is around 1.2$ρ_0$ and pion characteristic density is around 1.5$ρ_0$, and we got the constrains of symmetry energy at characteristic densities are $S(1.2ρ_0)=34\pm 4$ MeV and $S(1.5ρ_0)=36\pm 8$ MeV. These results are consistent with previous analysis by using pion and flow observable with different transport models, and demonstrate a reasonable description of symmetry energy constraint should be presented at the characteristic density of isospin sensitive observables.

nucl-th↗

arXiv2022

How Do Constraints of Nuclear Symmetry Energy Reconcile with Different Models?

By simultaneously describing the data of isospin sensitive nucleonic flow and pion observables, such as $v_2^n/v_2^{ch}$ and $π^-/π^+$, with ultra-relativistic quantum molecular dynamics (UrQMD) model, we got the symmetry energy at flow and pion characteristic densities which are $S(1.2ρ_0)=34\pm 4$ MeV and $S(1.5ρ_0)=36\pm 8$ MeV. Within the uncertainties, the constraints of symmetry energy at characteristic densities are consistent with the previous constraints by using other transport models. The consistency suggests that the reliable constraints on symmetry energy should be presented at the characteristic density of isospin sensitive observables. By using the constraints of symmetry energy at two different characteristic densities, the extrapolated value of $L$ is provided. Within $2σ$ uncertainty, the extrapolated value of $L$ is in $5-70$ MeV which is consistent with the recent combination analysis from PREX-II and astrophyiscs data. Further, the calculations with the constrained parameter sets can describe the data of charged pion multiplicities from S$π$RIT collaboration.

nucl-th↗

arXiv2022

Ultrafast laser welding of silicon

While ultrafast laser welding is an appealing technique for bonding transparent workpieces, it is not applicable for joining silicon samples due to nonlinear propagation effects which dramatically diminishes the possible energy deposition at the interface. We demonstrate that these limitations can be circumvented by local absorption enhancement at the interface thanks to metallic nanolayer deposition. By combining the resulting exalted absorption with filament relocation during ultrafast laser irradiation, silicon samples can be efficiently joined. Shear joining strengths >4 MPa are obtained for 21-nm gold nanolayers without laser-induced alteration of the transmittance. Such remarkable strength values hold promises for applications in microelectronics, optics, and astronomy.

physics.app-ph↗

arXiv2022

Elliptic flow in heavy-ion collisions at intermediate energy: the role of impact parameter, mean field potential, and collision term

Within the ultrarelativistic quantum molecular dynamics (UrQMD) model, by reverse tracing nucleons that are finally emitted at mid-rapidity (|$y_0$| < 0.1) in the entire reaction process, the time evolution of elliptic flow ($v_2$) of these traced nucleons produced in Au+Au collisions at beam energy of 0.4 GeV$/$nucleon with different impact parameters ($b$) is studied. The initial value of $v_2$ is positive and increases with $b$, then it decreases as time passes and tends to saturate at a negative value. It is found that nucleon-nucleon collisions always depress the value of $v_2$ (enhance the out-of-plane emission), while the nuclear mean field potential may slightly raise the value of $v_2$ during the expansion stage in peripheral reactions. The related density mostly probed by $v_2$ of nucleons at mid-rapidity is found to be $\sim$ 60% of the maximum density reached during the collisions.

nucl-th↗

arXiv2022

Effects of Initial Density Fluctuations on Cumulants in Au + Au Collisions at $\sqrt{s_{NN}}$ = 7.7 GeV

Within the ultrarelativistic quantum molecular dynamics (UrQMD) model, the effect of initial density fluctuations on cumulants of the net-proton multiplicity distribution in Au + Au Collisions at $\sqrt{s_{NN}}$ = 7.7 GeV was investigated by varying the minimum distance $d_{\rm min}$ between two nucleons in the initialization. It was found that the initial density fluctuations increased with the decrease of $d_{\rm min}$ from 1.6 fm to 1.0 fm, and the influence of $d_{\rm min}$ on the magnitude of the net-proton number fluctuation in a narrow pseudorapidity window ($Δη\leq$ 4) was negligible even if it indeed affected the density evolution during the collision. At a broad pseudorapidity window ($Δη\geq$ 4), the cumulant ratios were enlarged when the initial density fluctuations were increased with the smaller value of $d_{\rm min}$, and this enhancement was comparable to that observed in the presence of the nuclear mean-field potential. Moreover, the enhanced cumulants were more evident in collisions with a larger impact parameter. The present work demonstrates that the fingerprint of the initial density fluctuations on the cumulants in a broad pseudorapidity window is clearly visible, while it is not obvious as the pseudorapidity window becomes narrow.

nucl-th↗

arXiv2022

Effects of a phase transition on two-pion interferometry in heavy-ion collisions at $\sqrt{s_\mathrm{NN}}=2.4-7.7$ GeV

Hanbury-Brown-Twiss (HBT) correlations for charged pions in central Au+Au collisions at $\sqrt{s_\mathrm{NN}}=2.4 - 7.7~\text{GeV}$ (corresponding to beam kinetic energies in the fixed target frame from $E_{\rm{lab}}=1.23~\text{to}~30~\text{GeV/nucleon}$) are calculated using the UrQMD model with different equations of state. The effects of a phase transition at high baryon densities are clearly observed in the HBT parameters that are explored. It is found that the available data on the HBT radii, $R_{O}/R_{S}$ and $R^{2}_{O}-R^{2}_{S}$, in the investigated energy region favors a relatively stiff equation of state at low beam energies which then turns into a soft equation of state at high collision energies consistent with astrophysical constraints on the high density equation of state of QCD. The specific effects of two different phase transition scenarios on the $R_{O}/R_{S}$ and $R^{2}_{O}-R^{2}_{S}$ are investigated. It is found that a phase transition with a significant softening of the equation of state below 4 times nuclear saturation density can be excluded using HBT data. Our results highlight that the pion's $R_{O}/R_{S}$ and $R^{2}_{O}-R^{2}_{S}$ are sensitive to the stiffness of the equation of state, and can be used to constrain and understand the QCD equation of state in the high baryon density region.

nucl-th↗

arXiv2022

Decoding the nuclear symmetry energy event-by-event in heavy-ion collisions with machine learning

Inferences of the nuclear symmetry energy from heavy-ion collisions are currently based on the comparison of measured observables and transport model simulations. Only the expectation values of observables over all considered events are used in these approaches, however, observables can be obtained event-by-event both in experiments and transport model simulations. By using the light gradient boosting machine (LightGBM), a modern machine-learning algorithm, we present a framework for inferring the density-dependent nuclear symmetry energy from observables in heavy-ion collisions on the event-by-event analysis. The ultrarelativistic quantum molecular dynamics (UrQMD) model simulations are used as training data. The symmetry energy slope parameter extracted with LightGBM event-by-event from test data also by UrQMD has an average spread of approximately 30~MeV from the truth, and is found to be robust against variations in model parameters. In addition, LightGBM can identify features that have the greatest effect on the physics of interest, thereby offering valuable insights. Our study suggests that the present framework can be a powerful tool and may offer a new paradigm to study the underlying physics in heavy-ion collisions.

nucl-th↗

arXiv2022

Accessing the in-medium effects on nucleon-nucleon elastic cross section with collective flows and nuclear stopping

A systematic study of the in-medium correction factor ($F$) on nucleon-nucleon elastic cross section is performed within the Ultra-relativistic Quantum Molecular Dynamics (UrQMD) model. The effects of the beam energy dependence of $F$ on the directed, elliptic flow and nuclear stopping in $^{197}$Au+$^{197}$Au collisions with energy ranging from $0.09$ to $0.8A$ GeV are explored. It is found that the directed, elliptic flow and nuclear stopping at relatively low energies are very sensitive to $F$, and the sensitivity gradually weakens with increasing beam energy. The beam energy dependent in-medium correction factor $F$ is deduced from the comparison of the excitation functions of the directed, elliptic flow and nuclear stopping between the calculated results and the FOPI experimental data.

nucl-th↗

arXiv2022

An Experiment to Test the Mechanical Losses of Different Bonding Techniques in Fused Silica

High-purity glasses are used for their low optical and mechanical loss, which makes them an excellent material for oscillators in optical systems, such as inertial sensors. Complex geometries often require the assembly of multiple pieces of glass and their permanent bonding. One common method is hydroxide catalysis bonding, which leaves an enclosed medium layer. This layer has different mechanical properties to the bulk glass around it. The higher mechanical loss of this layer makes it more susceptible to displacement noise originating from the conversion of energy from oscillation to heat and vice versa. Therefore, other methods are needed to bond together glass assemblies. To investigate this, we have set up an experiment to measure the mechanical losses of several different types of bond commonly used in fused silica manufacturing, namely; plasma activated direct bonding, hydroxide catalysis bonding, laser welding, and adhesive bonding. In this paper we present the experimental design and show initial results of the first test sample.

physics.ins-det↗

arXiv2021

Applying machine learning to determine impact parameter in nuclear physics experiments

Machine Learning (ML) algorithms have been demonstrated to be capable of predicting impact parameter in heavy-ion collisions from transport model simulation events with perfect detector response. We extend the scope of ML application to experimental data by incorporating realistic detector response of the S$π$RIT Time Projection Chamber into the heavy-ion simulation events generated from the UrQMD model to resemble experimental data. At 3 fm, the predicted impact parameter is 2.8 fm if simulation events with perfect detector is used for training and testing; 2.4 fm if detector response is included in the training and testing, and 5.8 fm if ML algorithms trained with perfect detector is applied to testing data that has included detector response. The last result is not acceptable illustrating the importance of including the detector response in developing the ML training algorithm. We also test the model dependence by applying the algorithms trained on UrQMD model to simulated events from four different transport models as well as using different input parameters on UrQMD model. Using data from Sn+Sn collisions at E/A=270 MeV, the ML determined impact parameters agree well with the experimentally determined impact parameter using multiplicities, except in the very central and very peripheral regions. ML selects central collision events better and allows impact parameters determination beyond the sharp cutoff limit imposed by experimental methods.

physics.ins-det↗

arXiv2021

Finding signatures of the nuclear symmetry energy in heavy-ion collisions with deep learning

A deep convolutional neural network (CNN) is developed to study symmetry energy $E_{\rm sym}(ρ)$ effects by learning the mapping between the symmetry energy and the two-dimensional (transverse momentum and rapidity) distributions of protons and neutrons in heavy-ion collisions. Supervised training is performed with labelled data-set from the ultrarelativistic quantum molecular dynamics (UrQMD) model simulation. It is found that, by using proton spectra on event-by-event basis as input, the accuracy for classifying the soft and stiff $E_{\rm sym}(ρ)$ is about 60% due to large event-by-event fluctuations, while by setting event-summed proton spectra as input, the classification accuracy increases to 98%. The accuracy for 5-label (5 different $E_{\rm sym}(ρ)$) classification task are about 58% and 72% by using proton and neutron spectra, respectively. For the regression task, the mean absolute error (MAE) which measures the average magnitude of the absolute differences between the predicted and actual $L$ (the slope parameter of $E_{\rm sym}(ρ)$) are about 20.4 and 14.8 MeV by using proton and neutron spectra, respectively. Fingerprints of the density-dependent nuclear symmetry energy on the transverse momentum and rapidity distributions of protons and neutrons can be identified by convolutional neural network algorithm.

nucl-th↗

arXiv2021

Application of machine learning in the determination of impact parameter in the $^{132}$Sn+$^{124}$Sn system

Background: $^{132}$Sn+$^{124}$Sn collisions at the beam energy of 270 MeV$/$nucleon have been performed at the Radioactive Isotope Beam Factory (RIBF) in RIKEN to investigate the nuclear equation of state. Reconstructing impact parameter is one of the important tasks in the experiment as it relates to many observables. Purpose: In this work, we employ three commonly used algorithms in machine learning, the artificial neural network (ANN), the convolutional neural network (CNN) and the light gradient boosting machine (LightGBM), to determine impact parameter by analyzing either the charged particles spectra or several features simulated with events from the ultra-relativistic quantum molecular dynamics (UrQMD) model. Method: To closely imitate experimental data and investigate the generalizability of the trained machine learning algorithms, incompressibility of nuclear equation of state and the in-medium nucleon-nucleon cross sections are varied in the UrQMD model to generate the training data. Results: The mean absolute error $Δb$ between the true and the predicted impact parameter is smaller than 0.45 fm if training and testing sets are sampled from the UrQMD model with the same parameter set. However, if training and testing sets are sampled with different parameter sets, $Δb$ would increase to 0.8 fm. Conclusion: The generalizability of the trained machine learning algorithms suggests that these machine learning algorithms can be used reliably to reconstruct impact parameter in experiment.

nucl-th↗