arXiv subjects

Xiaohan Zhang

Publications and source records attributed to Xiaohan Zhang.

139 records · Page 8Linked to original sources

arXiv2021

Efficient Golf Ball Detection and Tracking Based on Convolutional Neural Networks and Kalman Filter

This paper focuses on the problem of online golf ball detection and tracking from image sequences. An efficient real-time approach is proposed by exploiting convolutional neural networks (CNN) based object detection and a Kalman filter based prediction. Five classical deep learning-based object detection networks are implemented and evaluated for ball detection, including YOLO v3 and its tiny version, YOLO v4, Faster R-CNN, SSD, and RefineDet. The detection is performed on small image patches instead of the entire image to increase the performance of small ball detection. At the tracking stage, a discrete Kalman filter is employed to predict the location of the ball and a small image patch is cropped based on the prediction. Then, the object detector is utilized to refine the location of the ball and update the parameters of Kalman filter. In order to train the detection models and test the tracking algorithm, a collection of golf ball dataset is created and annotated. Extensive comparative experiments are performed to demonstrate the effectiveness and superior tracking performance of the proposed scheme.

cs.CV↗

arXiv2021

Calibration-Free Vector Magnetometry Using Nitrogen-Vacancy Center in Diamond Integrated with Optical Vortex Beam

We report a new method to determine the orientation of individual nitrogen-vacancy (NV) centers in a bulk diamond and use them to realize a calibration-free vector magnetometer with nanoscale resolution. Optical vortex beam is used for optical excitation and scanning the NV center in a [111]-oriented diamond. The scanning fluorescence patterns of NV center with different orientations are completely different. Thus, the orientation information on each NV center in the lattice can be known directly without any calibration process. Further, we use three differently oriented NV centers to form a magnetometer and reconstruct the complete vector information on the magnetic field based on the optically detected magnetic resonance(ODMR) technique. Compared with previous schemes to realize vector magnetometry using an NV center, our method is much more efficient and is easily applied in other NV-based quantum sensing applications.

quant-ph↗

arXiv2021

Episodic memory governs choices: An RNN-based reinforcement learning model for decision-making task

Typical methods to study cognitive function are to record the electrical activities of animal neurons during the training of animals performing behavioral tasks. A key problem is that they fail to record all the relevant neurons in the animal brain. To alleviate this problem, we develop an RNN-based Actor-Critic framework, which is trained through reinforcement learning (RL) to solve two tasks analogous to the monkeys' decision-making tasks. The trained model is capable of reproducing some features of neural activities recorded from animal brain, or some behavior properties exhibited in animal experiments, suggesting that it can serve as a computational platform to explore other cognitive functions. Furthermore, we conduct behavioral experiments on our framework, trying to explore an open question in neuroscience: which episodic memory in the hippocampus should be selected to ultimately govern future decisions. We find that the retrieval of salient events sampled from episodic memories can effectively shorten deliberation time than common events in the decision-making process. The results indicate that salient events stored in the hippocampus could be prioritized to propagate reward information, and thus allow decision-makers to learn a strategy faster.

q-bio.NC↗

arXiv2020

Task-Motion Planning for Safe and Efficient Urban Driving

Autonomous vehicles need to plan at the task level to compute a sequence of symbolic actions, such as merging left and turning right, to fulfill people's service requests, where efficiency is the main concern. At the same time, the vehicles must compute continuous trajectories to perform actions at the motion level, where safety is the most important. Task-motion planning in autonomous driving faces the problem of maximizing task-level efficiency while ensuring motion-level safety. To this end, we develop algorithm Task-Motion Planning for Urban Driving (TMPUD) that, for the first time, enables the task and motion planners to communicate about the safety level of driving behaviors. TMPUD has been evaluated using a realistic urban driving simulation platform. Results suggest that TMPUD performs significantly better than competitive baselines from the literature in efficiency, while ensuring the safety of driving behaviors.

cs.RO↗

arXiv2020

Actor-Critic Algorithm for High-dimensional Partial Differential Equations

We develop a deep learning model to effectively solve high-dimensional nonlinear parabolic partial differential equations (PDE). We follow Feynman-Kac formula to reformulate PDE into the equivalent stochastic control problem governed by a Backward Stochastic Differential Equation (BSDE) system. The Markovian property of the BSDE is utilized in designing our neural network architecture, which is inspired by the Actor-Critic algorithm usually applied for deep Reinforcement Learning. Compared to the State-of-the-Art model, we make several improvements including 1) largely reduced trainable parameters, 2) faster convergence rate and 3) fewer hyperparameters to tune. We demonstrate those improvements by solving a few well-known classes of PDEs such as Hamilton-Jacobian-Bellman equation, Allen-Cahn equation and Black-Scholes equation with dimensions on the order of 100.

cs.LG↗

arXiv2020

Scale Calibrated Training: Improving Generalization of Deep Networks via Scale-Specific Normalization

Standard convolutional neural networks(CNNs) require consistent image resolutions in both training and testing phase. However, in practice, testing with smaller image sizes is necessary for fast inference. We show that trivially evaluating low-resolution images on networks trained with high-resolution images results in a catastrophic accuracy drop in standard CNN architectures. We propose a novel training regime called Scale calibrated Training(SCT) which allows networks to learn from various scales of input simultaneously. By taking advantages of SCT, single network can provide decent accuracy at test time in response to multiple test scales. In our analysis, we surprisingly find that vanilla batch normalization can lead to sub-optimal performance in SCT. Therefore, a novel normalization scheme called Scale-Specific Batch Normalization is equipped to SCT in replacement of batch normalization. Experiment results show that SCT improves accuracy of single Resnet-50 on ImageNet by 1.7% and 11.5% accuracy when testing on image sizes of 224 and 128 respectively.

cs.CV↗

arXiv2020

Plasticity without phenomenology: a first step

A novel, concurrent multiscale approach to meso/macroscale plasticity is demonstrated. It utilizes a carefully designed coupling of a partial differential equation (pde) based theory of dislocation mediated crystal plasticity with time-averaged inputs from microscopic Dislocation Dynamics (DD), adapting a state-of-the-art mathematical coarse-graining scheme. The stress-strain response of mesoscopic samples at realistic, slow, loading rates up to appreciable values of strain is obtained, with significant speed-up in compute time compared to conventional DD. Effects of crystal orientation, loading rate, and the ratio of the initial mobile to sessile dislocation density on the macroscopic response, for both load and displacement controlled simulations are demonstrated. These results are obtained without using any phenomenological constitutive assumption, except for thermal activation which is not a part of microscopic DD. The results also demonstrate the effect of the internal stresses on the collective behavior of dislocations, manifesting, in a set of examples, as a Stage I to Stage II hardening transition.

cs.CE↗

arXiv2020

Finite Element Approximation of Finite Deformation Dislocation Mechanics

We develop and demonstrate the first general computational tool for finite deformation static and dynamic dislocation mechanics. A finite element formulation of finite deformation (Mesoscale) Field Dislocation Mechanics theory is presented. The model is a minimal enhancement of classical crystal/$J_2$ plasticity that fundamentally accounts for polar/excess dislocations at the mesoscale. It has the ability to compute the static and dynamic finite deformation stress fields of arbitrary (evolving) dislocation distributions in finite bodies of arbitrary shape and elastic anisotropy under general boundary conditions. This capability is used to present a comparison of the static stress fields, at finite and small deformations, for screw and edge dislocations, revealing heretofore unexpected differences. The computational framework is verified against the sharply contrasting predictions of geometrically linear and nonlinear theories for the stress field of a spatially homogeneous dislocation distribution in the body, as well as against other exact results of the theory. Verification tests of the time-dependent numerics are also presented. Size effects in crystal and isotropic versions of the theory are shown to be a natural consequence of the model and are validated against available experimental data. With inertial effects incorporated, the development of an (asymmetric) propagating Mach cone is demonstrated in the finite deformation theory when a dislocation moves at speeds greater than the linear elastic shear wave speed of the material.

cond-mat.mtrl-sci↗

arXiv2020

Deep Reinforcement Learning for Fog Computing-based Vehicular System with Multi-operator Support

This paper studies the potential performance improvement that can be achieved by enabling multi-operator wireless connectivity for cloud/fog computing-connected vehicular systems. Mobile network operator (MNO) selection and switching problem is formulated by jointly considering switching cost, quality-of-service (QoS) variations between MNOs, and the different prices that can be charged by different MNOs as well as cloud and fog servers. A double deep Q network (DQN) based switching policy is proposed and proved to be able to minimize the long-term average cost of each vehicle with guaranteed latency and reliability performance. The performance of the proposed approach is evaluated using the dataset collected in a commercially available city-wide LTE network. Simulation results show that our proposed policy can significantly reduce the cost paid by each fog/cloud-connected vehicle with guaranteed latency services.

cs.DC↗

arXiv2019

Spherical harmonics method for computing the image stress due to a spherical void

We develop an efficient numerical method for calculating the image stress field induced by spherical voids in materials. The method is applied to dislocation-void interactions in dislocation dynamics simulations. We obtain a complete set of basis functions for the solution of the image stress problem, as well as their corresponding displacement, stress, and traction fields in terms of the linear combination of spherical harmonics. Using the fast transformation between the real and spherical-harmonics spaces provided by the SHTOOLS package, the method is more efficient than other image stress solvers such as the finite-element method. This method can readily be extended for solving elasticity problems involving inclusions and inhomogeneities, contact between spheres, as well as other differential equations with spherical boundaries beyond elasticity.

physics.comp-ph↗

arXiv2018

Critical Strain for Surface Nucleation of Dislocations in Silicon

A long-standing discrepancy exists between experiments and atomistic models concerning the critical strain needed for surface nucleation of dislocations in silicon-germanium systems. While dislocation nucleation is readily observed in hetero-epitaxial thin films with misfit strains less than 4%, existing atomistic models predict that a critical strain over 7.8% is needed to overcome the kinetic barrier for dislocation nucleation. Using zero-temperature energy barrier calculations and finite-temperature Molecular Dynamics simulations, we show that 3-dimensional surface features such as a sharply bent step can lower the predicted critical nucleation strain of a shuffle-set dislocation to 6.4%, and that of a shuffle-glide dislocation complex to 5.3%. Consistent findings are obtained using both the Stillinger-Weber (SW) and modified embedded-atom method (MEAM) potentials, providing support to the physical relevance of the shuffle-glide dislocation complex, which was previously considered as an artifact of the SW potential.

physics.app-ph↗

arXiv2016

A non-traditional view on the modeling of nematic disclination dynamics

Nonsingular disclination dynamics in a uniaxial nematic liquid crystal is modeled within a mathematical framework where the kinematics is a direct extension of the classical way of identifying these line defects with singularities of a unit vector field representing the nematic director. It is well known that the universally accepted Oseen-Frank energy is infinite for configurations that contain disclination line defects. We devise a natural augmentation of the Oseen-Frank energy to account for physical situations where, under certain conditions, infinite director gradients have zero associated energy cost, as would be necessary for modeling half-integer strength disclinations within the framework of the director theory. Equilibria and dynamics (in the absence of flow) of line defects are studied within the proposed model. Using appropriate initial/boundary data, the gradient-flow dynamics of this energy leads to non-singular, line defect equilibrium solutions, including those of half-integer strength. However, we demonstrate that the gradient flow dynamics for this energy is not able to adequately describe defect evolution. Motivated by similarity with dislocation dynamics in solids, a novel 2D-model of disclination dynamics in nematics is proposed. The model is based on the extended Oseen-Frank energy and takes into account thermodynamics and the kinematics of conservation of defect topological charge. We validate this model through computations of disclination equilibria, annihilation, repulsion, and splitting. We show that the energy function we devise, suitably interpreted, can serve as well for the modeling of equilibria and dynamics of dislocation line defects in solids making the conclusions of this paper relevant to mechanics of both solids and liquid crystals.

cond-mat.soft↗

arXiv2015

NFCU: A New Friendship-based Routing with Buffer Management in Opportunistic Networks

Routing in opportunistic networks is a very important and challenging problem because opportunistic network utilizes the contact opportunities of mobile nodes to achieve data communication.Social-based routing uses behavior of human beings which can form a community with the same interests to deliver the message.In this paper,we analyze the drawbacks of the original friendship-based algorithm, which defined social pressure metric to determine nodes'friendship community, but the metric couldn't distinguish the distribution characterization of the connection length which has an important impact on the selection of links with better quality. Further, the existing friendship-based routing doesn't consider the buffer management, which is vital for routing design in opportunistic networks. We propose a New Friendship-based routing with buffer management based on Copy Utility, named NFCU. NFCU algorithm, which not only considers the contact periods in constructing social pressure metric to solve the drawbacks of the original friendship-based routing scheme efficiently, but also considers the buffer management, that can efficiently determine which copy of the message should be deleted timely according to the copy utility function. Our proposed strategy can reduce the network overhead significantly, and increase the message delivery ratio. The extensive simulation results demonstrate that NFCU performs better than the original friendship-based routing. Moreover, we compare NFCU with other four classicalrouting schemes in opportunistic networks in terms of message delivery ratio, average delay, and comprehensive metric- message delivery ratio*(1/average delay). The simulation results show that our scheme NFCU can achieve better performance.

cs.NI↗