Real-Time Identification of Vehicle Motion-Modes
- Publisher:
- Springer Nature
- Publication Type:
- Conference Proceeding
- Citation:
- Vibration Engineering for a Sustainable Future, 2021, 2, pp. 167-173
- Issue Date:
- 2021-04-22
Closed Access
Filename | Description | Size | |||
---|---|---|---|---|---|
Chen2021_Chapter_Real-TimeIdentificationOfVehic.pdf | Published version | 1.1 MB |
Copyright Clearance Process
- Recently Added
- In Progress
- Closed Access
This item is closed access and not available.
A linear 7-DOF vehicle model includes seven motion-modes. Recognition of these motion-modes can improve chassis dynamic control strategies for active suspensions. The motion-mode energy method (MEM) was proposed to identify dominant vehicle motion-mode at each time instant. This method is based on an assumption that all 14 vehicle states and four road inputs are given. It is hard to achieve in practice. This paper proposes a novel method to classify vehicle dominant motion-modes in real-time based on long short-term memory (LSTM) classification networks. The effectiveness and accuracy of the trained networks are verified on a 7-DOF vehicle model under several driving scenarios. This method helps design of vehicle dynamic control strategies and improves active suspension performance.
Please use this identifier to cite or link to this item: