Zhijun Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zhijun Wang has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Guohui Wu
School of Clinical Medicine, Guangdong Pharmaceutical University, Guangzhou 510000, China
2 shared publications - Huiyuan Xiong
Guangdong Provincial Key Laboratory of Intelligent Transportation System, School of Intelligent Systems Engineering, Shenzhen Campus of Sun Yat-sen University, Shenzhen, China
2 shared publications - Yuelong Pan
China Nuclear Power Design Co.,Ltd
2 shared publications - Zhineng Long
Sun Yat-sen University
1 shared publication - Zichao Yang
Guangdong Provincial Key Laboratory of Intelligent Transportation System School of Intelligent Systems Engineering, Sun Yat‐sen University Shenzhen People's Republic of China
1 shared publication
Latest Publications
- 17 August 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 71 (00189456) , 1-13
Steering Actuator Fault Diagnosis for Autonomous Vehicle With an Adaptive Denoising Residual Network
A steering actuator is the core component of an autonomous vehicle. Because of noise interference, the extraction of useful fault features for an accurate fault diagnosis of steering actuator becomes an arduous task. This article proposes an adaptive denoising residual network (AD-ResNet) for fault diagnosis of steering actuator with reduced noise interference and improved accuracy. First, a high-complexity model based on long short-term memory (LSTM) is built to estimate redundant signals, and then, fault features are generated by comparing redundant signals with the measured sensor signals. Second, an attention unit is designed to determine the unique set of thresholds for each fault feature. Finally, hard thresholding is embedded in a deep network as a nonlinear transform layer to reduce noise. Simulation experiments were performed at different signal-to-noise ratios (SNRs), and the results confirmed that the proposed method could adaptively denoise the high-noise signals. The proposed method was then validated against a fault dataset that was built using road experiments with an autonomous vehicle, and it achieved an accuracy, a sensitivity, a specificity, and an F1-score of 91.66%, 91.66%, 97.22%, and 0.9159, respectively, averaged over all fault states. The proposed method outperformed the comparison methods in the experiments.
- 17 August 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 71 (00189456) , 1-13
Steering Actuator Fault Diagnosis for Autonomous Vehicle With an Adaptive Denoising Residual Network
A steering actuator is the core component of an autonomous vehicle. Because of noise interference, the extraction of useful fault features for an accurate fault diagnosis of steering actuator becomes an arduous task. This article proposes an adaptive denoising residual network (AD-ResNet) for fault diagnosis of steering actuator with reduced noise interference and improved accuracy. First, a high-complexity model based on long short-term memory (LSTM) is built to estimate redundant signals, and then, fault features are generated by comparing redundant signals with the measured sensor signals. Second, an attention unit is designed to determine the unique set of thresholds for each fault feature. Finally, hard thresholding is embedded in a deep network as a nonlinear transform layer to reduce noise. Simulation experiments were performed at different signal-to-noise ratios (SNRs), and the results confirmed that the proposed method could adaptively denoise the high-noise signals. The proposed method was then validated against a fault dataset that was built using road experiments with an autonomous vehicle, and it achieved an accuracy, a sensitivity, a specificity, and an F1-score of 91.66%, 91.66%, 97.22%, and 0.9159, respectively, averaged over all fault states. The proposed method outperformed the comparison methods in the experiments.
- Open Access
- 25 May 2022
- Published by Wiley in Journal of Sensors
- Vol. 2022, 1-12
Design and Implementation of Digital Twin-Assisted Simulation Method for Autonomous Vehicle in Car-Following Scenario
The automated system replaces the driver, which makes autonomous vehicle to improve safety and convenience, so the market of autonomous vehicle is huge. However, the real-world application of autonomous vehicles faces many challenges due to the immaturity of automated systems. As a consequence, simulation verification plays an irreplaceable role in the application of autonomous vehicle (AV). Car-following is the most common driving scenario in mixed traffic flows, so it is essential to develop an appropriate and effective simulation method for AV. Combined with the existing AV simulation methods and digital twin (DT) technology, this paper proposes a DT-assisted method for AV simulation in a car-following scenario. The method makes the physical vehicle interact with the DT vehicle, and the DT vehicle can dynamically regulate the physical entities through real-time simulation data; the simulation verification can be displayed in the DT scenario to ensure the security of the simulation. Meanwhile, a DT-assisted simulation framework of AV is proposed, the framework includes physical entity components, DT components, and data processing and evaluation components. Besides, a DT-assisted simulation platform is developed base on Unity engine. Finally, the DT-assisted simulation of AV in the car-following scenario is implemented in field experiment. The experimental results show that the proposed method can be effectively conducted AV simulation in car-following, and the average of communication latency is 52.3 ms, which is smaller than the update frequency 15 Hz (66.6 ms) between DT-assisted platform and AV. The DT-assisted simulation method of AV proposed in this paper is applied in the car-following scenario, which effectively solves the challenges of car-following scenario simulation through virtual-real interaction.
- 25 May 2022
- Published by Wiley in Journal of Sensors
- Vol. 2022, 1-12
Open AccessDesign and Implementation of Digital Twin-Assisted Simulation Method for Autonomous Vehicle in Car-Following Scenario
The automated system replaces the driver, which makes autonomous vehicle to improve safety and convenience, so the market of autonomous vehicle is huge. However, the real-world application of autonomous vehicles faces many challenges due to the immaturity of automated systems. As a consequence, simulation verification plays an irreplaceable role in the application of autonomous vehicle (AV). Car-following is the most common driving scenario in mixed traffic flows, so it is essential to develop an appropriate and effective simulation method for AV. Combined with the existing AV simulation methods and digital twin (DT) technology, this paper proposes a DT-assisted method for AV simulation in a car-following scenario. The method makes the physical vehicle interact with the DT vehicle, and the DT vehicle can dynamically regulate the physical entities through real-time simulation data; the simulation verification can be displayed in the DT scenario to ensure the security of the simulation. Meanwhile, a DT-assisted simulation framework of AV is proposed, the framework includes physical entity components, DT components, and data processing and evaluation components. Besides, a DT-assisted simulation platform is developed base on Unity engine. Finally, the DT-assisted simulation of AV in the car-following scenario is implemented in field experiment. The experimental results show that the proposed method can be effectively conducted AV simulation in car-following, and the average of communication latency is 52.3 ms, which is smaller than the update frequency 15 Hz (66.6 ms) between DT-assisted platform and AV. The DT-assisted simulation method of AV proposed in this paper is applied in the car-following scenario, which effectively solves the challenges of car-following scenario simulation through virtual-real interaction.