Feng Sun

Feng Sun

Auto-generated profile, based on publicly available information

Feng Sun

Harbin Engineering University, Harbin, China

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 September 2019 in IEEE Access
Verified
Journal Article
Open access
FSFeng Sun
Feng Sun
HWHanrui Wu
Hanrui Wu
ZLZhihang Luo
Zhihang Luo
WGWenwen Gu
Wenwen Gu
YYYuguang Yan
Yuguang Yan
QDQing Du
Qing Du
Domain adaptation aims at extracting knowledge from an auxiliary source domain to assist the learning task in a target domain. When the data distribution of the target domain is different from that of the source domain, the direct use of source data for building a classifier for the target learning task cannot achieve promising performance. In this work, we propose a novel unsupervised domain adaptation method called Feature Selection for Domain Adaptation (FSDA), in which we aim to select a set of informative features. The benefits are two-fold. The first is to reduce the mismatch between the data distributions in the source and target domains by selecting a set of informative features in which they share similar properties. The second is to remove noisy features in the source domain such that the learning performance can be enhanced. We formulate a new sparse learning model for structured multiple outputs, including a vector to select informative features that can be used to jointly minimize the domain discrepancy and eliminate noisy features, and a classifier to perform prediction on the selected features. We develop a cutting-plane algorithm to solve the resulting optimization problem. Extensive experiments on real-world data sets are tested to demonstrate the effectiveness of the proposed method compared with the other existing methods.

1 August 2013
Verified
Conference Paper
Closed access
QLQiurong Li
Qiurong Li
FSFeng Sun
Feng Sun
Based on particle filter (PF) and cubature Kalman filter (CKF), with the maximum posterior principle (MAP), a new filter algorithm - the adaptive cubature particle filter (ACPF) is derivated. From the theory it can be seen, that ACPF algorithm not only has the strict mathematical derivation, but also can improve filtering accuracy in the system under the condition of high dimension. ACPF has the advantages of high reliability, low sensitivity, strong robustness, strong stability and convergence. The ACPF and several filter algorithms such as PF, UKF and CKF which are often used in recent years, are applied to the simulation of GPS/INS integrated navigation system, experiments show that ACPF is better than the others. The simulation results has proved the correctness of the theoretical derivation of the conclusion.

1 December 2013
Verified
Conference Paper
Closed access
WZWang Zongyuan
Wang Zongyuan
SFSun Feng
Sun Feng
The method of decentralized estimation is one that is fulfilled through decomposition of state space. Target tracking model with turning angular unknown has both nonlinear state function and measure function. Regular particle filter and UKF are both divergent through simulation. Then decentralized particle filter method was used by dividing state and state function into two parts accordingly, through decoupling correlated noise of state function, designing importance function, the two parts were estimated by particle filter separately. The outcome illustrated indicates it not only free of degeneracy, but also having a high accuracy of filtering, shortening time of filtering. At last, a decentralized estimation based on mixed nonlinear filter is addressed, which the weight of local level is not estimated.

7 October 2019
Verified
Conference Paper
Closed access
PDF file
FSFeng Sun
Feng Sun
HXHe Xu
He Xu
LJLiye Jiao
Liye Jiao
ZZZitong Zhao
Zitong Zhao
MCMin Cheng
Min Cheng
Researches on erosion resistance of valves from the perspective of variable stiffness and energy mitigating are rarely to be seen so far. The ballistic protection for body armor usually requires both high strength and high energy mitigation. Based on the above characteristics of the body armor, ballistic protection for body armor is analogous to the valve plug erosion resistance. In this work, we introduce a new energy mitigating body armor inspired valve plug (EMAVP) to improve its anti-erosion performance. The EMAVP has a gradient hierarchical structure in hardness, variable stiffness materials and structures. The authors analyze the variety of stiffness and energy mitigating of the EMAVP. The fluid-structure interaction (FSI) method is used to simulate the erosion performance, the EMAVP and normal valve plug specimens were tested in the comprehensive experimental system of water hydraulic valve and jet erosion facility. Results show that the EMAVP is superior to normal valve plug, with higher energy mitigating and better anti-erosion performance.