Zhonghe He

Zhonghe He

Auto-generated profile, based on publicly available information

Zhonghe He

North China University of Technology,School of Electrical and Control Engineering,Beijing,China

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

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

Representative Publications

View All Publications
1 September 2012
Verified
Conference Paper
Closed access
XHXingguang Han
Xingguang Han
YCYangzhou Chen
Yangzhou Chen
JSjianjun shi
jianjun shi
ZHZhonghe He
Zhonghe He
This paper proposes a macroscopic traffic flow model for urban traffic road network. First, dynamic digraph is used to model the urban traffic road network's topological structure. Then, traffic flow data which is based on an extended cell transmission model and the traffic shock wave theory are endowed to the dynamic digraph to construct a weighted dynamic digraph model. Finally, the modeling method is applied to a local road network to illustrate the availability of the modeling method when compared with real data.

17 July 2013 in Mathematical Problems in Engineering
Verified
Journal Article
Open access
PDF file
YGYanrong Ge
Yanrong Ge
YCYangzhou Chen
Yangzhou Chen
YZYaxiao Zhang
Yaxiao Zhang
ZHZhonghe He
Zhonghe He
The paper deals with the state consensus problem of high-order discrete-time linear multiagent systems (DLMASs) with fixed information topologies. We consider three aspects of the consensus analysis and design problem: (1) the convergence criteria of global state consensus, (2) the calculation of the state consensus function, and (3) the determination of the weighted matrix and the feedback gain matrix in the consensus protocol. We solve the consensus problem by proposing a linear transformation to translate it into a partial stability problem. Based on the approach, we obtain necessary and sufficient criteria in terms of Schur stability of matrices and present an analytical expression of the state consensus function. We also propose a design process to determine the feedback gain matrix in the consensus protocol. Finally, we extend the state consensus to the formation control. The results are explained by several numerical examples.

1 January 2016
Verified
Conference Paper
Open access
PDF file
ZZZhongcheng Zhou
Zhongcheng Zhou
XZXiang Zhu
Xiang Zhu
ZHZhonghe He
Zhonghe He
YQYinchuan Qu
Yinchuan Qu

1 July 2015
Verified
Conference Paper
Closed access
HZHe Zhonghe
He Zhonghe
CZChi Zhang
Chi Zhang
Li WangLi Wang
Li Wang
Typical reason of traffic congestion within urban road traffic networks (URTNs), is that the distribution of traffic flows is unbalanced, i.e., occupancy of vehicles in some links is high, however there still exists much space being not utilized sufficiently, in other links. Motivated by this problem, the paper proposes a real-time signal control strategy, called the UTBC (Urban Traffic Balance Control, for short) strategy, under which traffic states within links can reach consensus, i.e., the network can realize the balanced distribution of traffic flows. First, a signal control model of URTNs based on basic conservation principle of vehicle flows is proposed, which is a discrete-time linear time-invariant control system, with control input being phase green-time durations. Furthermore, the state-feedback signal control law (i.e., the UTBC strategy), is proposed, such that the closed-loop signal control system can realize consensus. Moreover, ALINEA ramp metering algorithm is considered. At last, we illustrate joint applications of the UTBC and ALINEA strategies by conducting simulation tests for a real-world network in Wangjing District in Beijing. The simulation results show that the UTBC strategy is feasible and effective.

15 July 2022
Verified
Conference Paper
Closed access
XZXiaoping Zhang
Xiaoping Zhang
TYTianhang Yang
Tianhang Yang
LWLi Wang
Li Wang
ZHZhonghe He
Zhonghe He
SLshida Liu
shida Liu
Deep learning is an important branch of neural networks, which has high accuracy in classification and regression problems, and has been widely used. However, its performance is greatly affected by the parameters. In this paper, an improved particle swarm algorithm named as PSO-C is proposed to automatically train the parameters of the feedforward neural networks. In the proposed algorithm, the curiosity factor is introduced to divide the particles into two categories with different curiosity characteristics so as to improve the exploration ability and information mining ability of the particle swarms. At the same time, a chaotic factor is also introduced to avoid the local optimum problem during the neural network’s training. The simulation results show that the PSO-C has better optimization effect on the whole.

7 January 2022
Verified
Conference Paper
Open access
CLChang Li
Chang Li
YWYiding Wang
Yiding Wang
ZHZhonghe He
Zhonghe He
Fully-convolutional Siamese network vehicle tracking algorithm (SiamFC) has become a research hotspot in the field of vehicle tracking in recent years. However, this kind of tracker algorithm lacks deep feature extraction ability and overall perception of the target vehicle, is easy to lose in the case of complex background, scale changing and changing illumination. This paper proposes to use the Inception network as the backbone network, and according to the characteristics of tracking model, optimized it from two aspects: cutting the feature map and adjusting the total network step size, making inception network can extract the high-level semantic information of the vehicle while maintaining the sensitivity to the change of vehicle position. And introducing the Non-local network (NLNet) into the branch networks to enhance the overall perception of the tracking model for the target vehicle. Experiments show that the tracking speed and robustness of the proposed algorithm are significantly improved in the case of scale changing, illumination variation and complex background. Better tracking results can be obtained in dataset OTB2013 and OTB2015. Compared with the classical trackers such as SiamFC, the tracking accuracy is improved by 4.2% in OTB2015, and the tracking success rate is improved by 2.6% in OTB2015, and the tracking speed is up to 90FPS.