使用惯性传感器对半活性悬挂进行实时振动能量预测:以物理为导向的深度学习方法
Jian Cheng1, Fanhua Qin1, Leyao Wang1
1School of Mechanical and Electrical Engineering, North China Institute of Aerospace Engineering, Langfang 065000, China.
这项研究介绍了一种半活性悬浮体的物理信息化门卷积神经网络 (PI-GCNN). PI-GCNN预测了未来的道路冲击能量,使控制响应更快,车辆稳定性更好.
科学领域:
- 控制系统工程 控制系统工程
- 汽车应用中的人工智能
- 信号处理 信号处理
背景情况:
- 响应延迟和传感器噪声是闭环控制系统的关键挑战,特别是影响半活性悬挂.
- 物理启动延迟和信号过阶段延迟导致控制响应滞后于道路冲击激发.
研究的目的:
- 开发半活性悬浮的预测控制框架,以克服延迟问题.
- 通过预测未来的多模式能源演变来实现前控制.
主要方法:
- 提出了一种基于物理知识的门式卷积神经网络 (PI-GCNN),利用连续波纹变换 (CWT) 来进行时间频率分析.
- 实施了以非对称的稀疏物理损失为噪声抑制和冲击灵敏度而训练的物理引导门机制.
- 使用重型卡车模拟和PVS 9真实世界数据集验证了模型.
主要成果:
- 在实时基线上,PI-GCNN实现了100-200毫秒的预测阶段领先.
- 证明了具有0.10M参数和0.25ms单推理延迟的异常计算效率.
- 成功创建了悬挂阻尼器的有价值的启动窗口.
结论:
- 通过预测控制,PI-GCNN有效地解决了半活性悬浮中的延迟和噪声挑战.
- 该模型的效率使其适用于资源有限的汽车边缘计算平台.
- 这种方法在提高车辆动态和驾驶舒适度方面取得了重大进展.
更多相关视频
08:27Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
Published on: January 5, 2024
06:45Design and Application of a Fault Detection Method Based on Adaptive Filters and Rotational Speed Estimation for an Electro-Hydrostatic Actuator
Published on: October 28, 2022
相关概念视频
Linear Approximation in Time Domain
For a simple pendulum with a mass evenly distributed along its length and the center of mass located at half the pendulum's length,...
PD Controller: Design
Designing a continuous-data controller requires selecting and linking components like adders and integrators, which are fundamental in Proportional,...
Mechanical Systems
Problem Solving: Energy in Simple Harmonic Motion
Consider the spring in a shock absorber of a car. The spring attached to the wheel executes simple harmonic motion while the car is moving on a bumpy road. The force on the...
Relative Motion Analysis using Rotating Axes - Acceleration
Time differentiation is...
Relative Motion Analysis - Acceleration
