arXiv subjects

Xinjun Liu

Publications and source records attributed to Xinjun Liu.

9 recordsLinked to original sources

arXiv2025

EquiMus: Energy-Equivalent Dynamic Modeling and Simulation of Musculoskeletal Robots Driven by Linear Elastic Actuators

Dynamic modeling and control are critical for unleashing soft robots' potential, yet remain challenging due to their complex constitutive behaviors and real-world operating conditions. Bio-inspired musculoskeletal robots, which integrate rigid skeletons with soft actuators, combine high load-bearing capacity with inherent flexibility. Although actuation dynamics have been studied through experimental methods and surrogate models, accurate and effective modeling and simulation remain a significant challenge, especially for large-scale hybrid rigid--soft robots with continuously distributed mass, kinematic loops, and diverse motion modes. To address these challenges, we propose EquiMus, an energy-equivalent dynamic modeling framework and MuJoCo-based simulation for musculoskeletal rigid--soft hybrid robots with linear elastic actuators. The equivalence and effectiveness of the proposed approach are validated and examined through both simulations and real-world experiments on a bionic robotic leg. EquiMus further demonstrates its utility for downstream tasks, including controller design and learning-based control strategies.

cs.RO↗

arXiv2021

SuperMeshing: A Novel Method for Boosting the Mesh Density in Numerical Computation within 2D Domain

Due to the limit of mesh density, the improvement of the spatial precision of numerical computation always leads to a decrease in computing efficiency. Aiming at this inability of numerical computation, we propose a novel method for boosting the mesh density in numerical computation within the 2D domain. Based on the low mesh-density stress field in the 2D plane strain problem computed by the finite element method, this method utilizes a deep neural network named SuperMeshingNet to learn the non-linear mapping from low mesh-density to high mesh-density stress field, and realizes the improvement of numerical computation accuracy and efficiency simultaneously. We adopt residual dense blocks to our mesh-density boost model called SuperMeshingNet for extracting abundant local features and enhancing the prediction capacity of the model. Experimental results show that the SuperMeshingNet proposed in this work can effectively boost the spatial resolution of the stress field under the multiple scaling factors: 2X, 4X, 8X. Compared to the results of the finite element method, the predicted stress field error of SuperMeshingNet is only 0.54%, which is within the acceptable range of stress field estimation, and the SuperMeshingNet predicts the maximum stress value also without significant accuracy loss. We publicly share our work with full detail of implementation at https://github.com/zhenguonie/2021_SuperMeshing_2D_Plane_Strain.

math.NA↗

arXiv2021

SuperMeshing: A New Deep Learning Architecture for Increasing the Mesh Density of Metal Forming Stress Field with Attention Mechanism and Perceptual Features

In stress field analysis, the finite element analysis is a crucial approach, in which the mesh-density has a significant impact on the results. High mesh density usually contributes authentic to simulation results but costs more computing resources, leading to curtailing efficiency during the design process. To eliminate this drawback, we propose a new data-driven mesh-density boost model named SuperMeshingNet that strengthens the advantages of finite element analysis (FEA) with low mesh-density as inputs to the deep learning model, which consisting of Res-UNet architecture, to acquire high-density stress field instantaneously, shortening computing time and cost automatically. Moreover, the attention mechanism and the perceptual features are utilized, enhancing the performance of SuperMeshingNet. Compared to the baseline that applied the linear interpolation method, SuperMeshingNet achieves a prominent reduction in the mean squared error (MSE) and mean absolute error (MAE) on test data, which contains prior unseen cases. Based on the data set of metal forming, the comparable experiments are proceeded to demonstrate the high quality and superior precision of the reconstructed results generated by our model. The well-trained model can successfully show more excellent performance than the baseline and other methods on the multiple scaled mesh-density, including $2\times$, $4\times$, and $8\times$. With the refined result owning broaden scaling of mesh density and high precision, the FEA process can be accelerated with seldom cost on computation resources. We publicly share our work with full detail of implementation at https://github.com/zhenguonie/2021_SuperMeshing_2D_Metal_Forming

cs.CE↗

arXiv2021

Artificial Intelligence Enhanced Rapid and Efficient Diagnosis of Mycoplasma Pneumoniae Pneumonia in Children Patients

Artificial intelligence methods have been increasingly turning into a potentially powerful tool in the diagnosis and management of diseases. In this study, we utilized logistic regression (LR), decision tree (DT), gradient boosted decision tree (GBDT), support vector machine (SVM), and multilayer perceptron (MLP) as machine learning models to rapidly diagnose the mycoplasma pneumoniae pneumonia (MPP) in children patients. The classification task was carried out after applying the preprocessing procedure to the MPP dataset. The most efficient results are obtained by GBDT. It provides the best performance with an accuracy of 93.7%. In contrast to standard raw feature weighting, the feature importance takes the underlying correlation structure of the features into account. The most crucial feature of GBDT is the "pulmonary infiltrates range" with a score of 0.5925, followed by "cough" (0.0953) and "pleural effusion" (0.0492). We publicly share our full implementation with the dataset and trained models at https://github.com/zhenguonie/2021_AI4MPP.

cs.LG↗

arXiv2020

NbO2-based memristive neurons for burst-based perceptron

Neuromorphic computing using spike-based learning has broad prospects in reducing computing power. Memristive neurons composed with two locally active memristors have been used to mimic the dynamical behaviors of biological neurons. In this work, the dynamic operating conditions of NbO2-based memristive neurons and their transformation boundaries between the spiking and the bursting are comprehensively investigated. Furthermore, the underlying mechanism of bursting is analyzed and the controllability of the number of spikes during each burst period is demonstrated. Finally, pattern classification and information transmitting in a perceptron neural network by using the number of spikes per bursting period to encode information is proposed. The results show a promising approach for the practical implementation of neuristor in spiking neural networks.

cs.ET↗

arXiv2019

Schottky barrier induced asymmetry in the negative differential resistance response of Nb/NbOx/Pt cross-point devices

The negative differential resistance (NDR) response of Nb/NbOx/Pt cross-point devices is shown to have a polarity dependence due to the effect of the metal/oxide Schottky barriers on the contact resistance. Three distinct responses are observed under opposite polarity testing: bipolar S-type NDR, bipolar snap-back NDR, and combined S-type and snap-back NDR, depending on the stoichiometry of the oxide film and device area. In-situ thermoreflectance imaging is used to show that these NDR responses are associated with strong current localisation, thereby justifying the use of a previously developed two-zone, core shell thermal model of the device. The observed polarity dependent NDR responses, and their dependence on stoichiometry and area are then explained by extending this model to include the effect of the polarity dependent contact resistance. This study provides an improved understanding of the NDR response of metal/oxide/metal structures and informs the engineering of devices for neuromorphic computing and non-volatile memory applications.

physics.app-ph↗

arXiv2019

Current localisation and redistribution as the basis of discontinuous current controlled negative differential resistance in NbOx

In-situ thermo-reflectance imaging is used to show that the discontinuous, snap-back mode of current-controlled negative differential resistance (CC-NDR) in NbOx-based devices is a direct consequence of current localization and redistribution. Current localisation is shown to result from the creation of a conductive filament either during electroforming or from current bifurcation due to the super-linear temperature dependence of the film conductivity. The snap-back response then arises from current redistribution between regions of low and high current-density due to the rapid increase in conductivity created within the high current density region. This redistribution is further shown to depend on the relative resistance of the low current-density region with the characteristics of NbOx cross-point devices transitioning between continuous and discontinuous snap-back modes at critical values of film conductivity, area, thickness and temperature, as predicted. These results clearly demonstrate that snap-back is a generic response that arises from current localization and redistribution within the oxide film rather than a material-specific phase transition, thus resolving a long-standing controversy.

physics.app-ph↗

arXiv2019

Origin of current-controlled negative differential resistance modes and the emergence of composite characteristics with high complexity

Current-controlled negative differential resistance has significant potential as a fundamental building block in brain-inspired neuromorphic computing. However, achieving desired negative differential resistance characteristics, which is crucial for practical implementation, remains challenging due to little consensus on the underlying mechanism and unclear design criteria. Here, we report a material-independent model of current-controlled negative differential resistance to explain a broad range of characteristics, including the origin of the discontinuous snap-back response observed in many transition metal oxides. This is achieved by explicitly accounting for a non-uniform current distribution in the oxide film and its impact on the effective circuit of the device, rather than a material-specific phase transition. The predictions of the model are then compared with experimental observations to show that the continuous S-type and discontinuous snap-back characteristics serve as fundamental building blocks for composite behaviour with higher complexity. Finally, we demonstrate the potential of our approach for predicting and engineering unconventional compound behaviour with novel functionality for emerging electronic and neuromorphic computing applications.

physics.app-ph↗

arXiv2018

DS-SLAM: A Semantic Visual SLAM towards Dynamic Environments

Simultaneous Localization and Mapping (SLAM) is considered to be a fundamental capability for intelligent mobile robots. Over the past decades, many impressed SLAM systems have been developed and achieved good performance under certain circumstances. However, some problems are still not well solved, for example, how to tackle the moving objects in the dynamic environments, how to make the robots truly understand the surroundings and accomplish advanced tasks. In this paper, a robust semantic visual SLAM towards dynamic environments named DS-SLAM is proposed. Five threads run in parallel in DS-SLAM: tracking, semantic segmentation, local mapping, loop closing, and dense semantic map creation. DS-SLAM combines semantic segmentation network with moving consistency check method to reduce the impact of dynamic objects, and thus the localization accuracy is highly improved in dynamic environments. Meanwhile, a dense semantic octo-tree map is produced, which could be employed for high-level tasks. We conduct experiments both on TUM RGB-D dataset and in the real-world environment. The results demonstrate the absolute trajectory accuracy in DS-SLAM can be improved by one order of magnitude compared with ORB-SLAM2. It is one of the state-of-the-art SLAM systems in high-dynamic environments. Now the code is available at our github: https://github.com/ivipsourcecode/DS-SLAM

cs.RO↗