| 引用本文: | 刘瀚蔚,王旭东,黄剑龙.基于模糊神经网络算法的智能汽车轨迹自适应跟踪控制研究(J/M/D/N,J:杂志，M：书，D：论文，N：报纸).期刊名称,2025，42（6）：63-71 |
| CHEN X. Adap tive slidingmode contr ol for discrete2ti me multi2inputmulti2 out put systems[ J ]. Aut omatica, 2006, 42(6): 4272-435 |
|
| 摘要: |
| 目的 针对传统单一控制方法在智能汽车轨迹跟踪控制领域应用工况受限,实时性较低,精确度不高的问
题,提出一种基于模糊神经网络算法的横纵向协同控制策略(ANFIS-LQR/ PID),旨在保证智能汽车在不同道路条
件下轨迹跟踪的精确性和驾乘人员的舒适性。 方法 对汽车的运动情况进行分析并构建模型,将其投影到 Frenet 坐
标系下,并以期望坐标和当前坐标的偏差值作为状态变量,构建一个跟踪误差模型。 将设计的自适应模糊神经网
络调节策略与车辆横纵向协同控制器进行融合,实现了横向线性二次控制(LQR)与纵向比例积分微分(PID)权重
系数的实时调节,有效解决了不同车速下的跟踪控制问题。 结果 通过 PreScan-CarSim / Simulink 软件搭建联合仿
真平台,设计了连续弯道跟车、双车道变速超车、大曲率弯道加速跟车等 3 种涉及不同横纵向控制问题的工况进行
仿真验证。 结论 仿真结果表明:本文所设计的基于模糊神经网络算法的横纵向协同控制器在各种工况下均能将车
辆的轨迹跟踪误差稳定地控制在限制范围内,同时在最优控制下保证了驾乘人员的舒适性。 |
| 关键词: 智能汽车 跟踪控制 LQR PID ANFIS |
| DOI： |
| 分类号: |
| 基金项目: |
|
| Research on Adaptive Tracking Control of Intelligent Vehicle Trajectory Based on Fuzzy Neural NetworkAlgorithm |
|
LIU Hanwei WANG Xudong HUANG Jianlong
|
|
School of Mechanical Engineering Chongqing Technology and Business University Chongqing 400067 China
|
| Abstract: |
| Objective Aiming at the problems of limited application scenarios poor real-time performance and low
accuracy of traditional single-control methods in the field of intelligent vehicle trajectory tracking control a laterallongitudinal collaborative control strategy based on a fuzzy neural network algorithm ANFIS - LQR/ PID is proposed. The
aim is to ensure the accuracy of trajectory tracking of intelligent vehicles under different road conditions and the comfort of
drivers and passengers. Methods The motion of the vehicle is analyzed and a model is constructed. This model is
projected onto the Frenet coordinate system and a tracking error model is built with the deviation values between the
desired and current coordinates as state variables. The designed adaptive fuzzy neural network adjustment strategy is
integrated with the vehicle?? s lateral-longitudinal collaborative controller to achieve real-time adjustment of the weight
coefficients of lateral linear quadratic regulation LQR and longitudinal proportional-integral-derivative PID so as to
effectively solve the problem of tracking control under different vehicle speeds. Results The PreScan-CarSim / Simulink
software is used to build a joint simulation platform. The platform is used to simulate three working conditions continuous
curve following two-lane variable-speed overtaking and large-curvature curve acceleration following involving different
lateral-longitudinal control problems for verification. Conclusion The simulation results show that the lateral-longitudinalcollaborative controller based on the fuzzy neural network algorithm designed in this paper can stably control the vehicle?? s
trajectory tracking error within the specified range under various working conditions. Additionally it ensures the comfort of
drivers and passengers under optimal control. |
| Key words: intelligent vehicle tracking control LQR PID ANFIS |
