Changxuan Zhou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Changxuan Zhou has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Kaiyue Zhao
School of Information Engineering, Chang'an University, China
1 shared publication - Qiuhong Tong
School of Automotive Engineering, Chang'an University, China
1 shared publication - Wei Wu
School of Automobile, Chang'an University, China
1 shared publication - Wu Zhang
Shaanxi Intelligent Connected Vehicle Research Institute Co., Ltd, China
1 shared publication - Yimeng Wang
School of Automobile, Chang'an University, China
1 shared publication
Latest Publications
- Open Access
- 17 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-7
A Novel Transverse Control System for Electromagnetic Tracking Car Based on BP Neural Network
In recent years, the phenomenon of predictive control on neural networks (NNs) has captivated substantial attention in the intelligent vehicle field. Though existing intelligent vehicle control methods demonstrate good control effect in sufficient sensing range, they perform poorly in insufficient sensing range. To address the issue of poor predictability in insufficient sensing range, a novel transverse control system is proposed based on neural network from the perspective of nonlinear mapping. Initially, we design a framework for the BP neural network. Subsequently, the module is trained on the data collected from the short look-ahead when the long look-ahead was used as the PID sensor. Finally, we deployed the model to the controller of the electromagnetic car. According to the comparison experiments with traditional PID control methods, the experimental results show that the method proposed in this paper improve the prediction time by 13.1%. At the same time, it can also reduce the system shaking compared with the traditional PID control for the same length of short look-ahead condition. Furthermore, the experimental results confirming its superiority when intelligent vehicle's sensing range is deficiency.
- 17 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessA Novel Transverse Control System for Electromagnetic Tracking Car Based on BP Neural Network
In recent years, the phenomenon of predictive control on neural networks (NNs) has captivated substantial attention in the intelligent vehicle field. Though existing intelligent vehicle control methods demonstrate good control effect in sufficient sensing range, they perform poorly in insufficient sensing range. To address the issue of poor predictability in insufficient sensing range, a novel transverse control system is proposed based on neural network from the perspective of nonlinear mapping. Initially, we design a framework for the BP neural network. Subsequently, the module is trained on the data collected from the short look-ahead when the long look-ahead was used as the PID sensor. Finally, we deployed the model to the controller of the electromagnetic car. According to the comparison experiments with traditional PID control methods, the experimental results show that the method proposed in this paper improve the prediction time by 13.1%. At the same time, it can also reduce the system shaking compared with the traditional PID control for the same length of short look-ahead condition. Furthermore, the experimental results confirming its superiority when intelligent vehicle's sensing range is deficiency.