• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Kybernetes: The International Journal of Systems & Cybernetics >T-S fuzzy model based generalized predictive control of vehicle yaw stability
【24h】

T-S fuzzy model based generalized predictive control of vehicle yaw stability

机译：基于T-S模糊模型的车辆横摆稳定性广义预测控制

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Purpose - The purpose of this paper is to design a steering control for vehicles to protect the vehicle from spin and to realize improved cornering performance. Design/methodology/approach - The improved cornering performance is realized based on Takagi-Sugeno fuzzy model and generalized predictive control (GPC). A new approach to establish model of the vehicle is presented on the basis of fuzzy neural network. The network which inputs and outputs are composed of five layers of forward structure is utilized to build the structure and parameters of T-S fuzzy model through learning from training data. In this way, the vehicle dynamic system is divided into many linear sub-systems, and the system output is the weighted-sum of these sub-systems' outputs. A CARIMA model can be derived from the presented fuzzy model, and GPC is applied to deal with the control problem of vehicle stability. Findings - Vehicle model can be divided into local linear models, corresponding controller can be developed. Simulation results show that fuzzy model based on GPC can be applied to improve stability of the vehicle effectively. Research limitations/implications - As an exploration of a new approach, the training data are from simulation, and the result of the paper will be applied in actual vehicle trials. Practical implications - The paper presents useful advice for developing a vehicle stability controller. Originality/value - The paper presents a new approach to establish a model of the vehicle on the basis of fuzzy neural network, which is valuable for establishing a new controller for vehicle stability.
机译：目的-本文的目的是设计一种用于车辆的转向控制装置，以保护车辆免受打滑并提高转弯性能。设计/方法/方法-基于Takagi-Sugeno模糊模型和广义预测控制（GPC），可提高转弯性能。在模糊神经网络的基础上，提出了一种新的车辆模型建立方法。输入和输出由五层前向结构组成的网络通过从训练数据中学习来构建T-S模糊模型的结构和参数。以此方式，车辆动态系统被分为许多线性子系统，并且系统输出是这些子系统的输出的加权和。可以从提出的模糊模型中导出CARIMA模型，并应用GPC来解决车辆稳定性的控制问题。研究结果-车辆模型可以分为局部线性模型，可以开发相应的控制器。仿真结果表明，基于GPC的模糊模型可以有效地提高车辆的稳定性。研究的局限性/意义-作为一种新方法的探索，训练数据来自模拟，论文的结果将应用于实际的车辆试验中。实际意义-本文为开发车辆稳定性控制器提供了有用的建议。原创性/价值-本文提出了一种基于模糊神经网络建立车辆模型的新方法，这对于建立新的车辆稳定性控制器很有用。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号