不可靠车载传输环境下的智能汽车轨迹跟踪控制方法

Trajectory Tracking Control of Intelligent Vehicle in Unreliable Onboard Data Transmission Environment

针对外界扰动、参数摄动、数据传输时滞、转向输出滞后等因素给车辆运动控制带来的严峻挑战,基于鲁棒保性能控制理论提出了一种面向不可靠车载传输环境的智能汽车轨迹跟踪控制策略。通过系统扩维的方式引入转向系统动力学,建立增广无滞后不确定轨迹跟踪动力学模型,以描述执行器动态特性影响下的车路耦合动力学响应。基于Lyapunov-Krasovskii(LK)泛函构建时滞相关稳定性判据,考虑模型参数摄动所引起的系统失配问题,采用不等式放大法对不确定交叉项进行放大,引入H∞指标对广义外界扰动进行抑制,并通过保性能指标配置控制性能偏好,设计具备线性参数时变(Linear-parameter-varying,LPV)特征的纵横向鲁棒协同控制器。最后通过多个典型工况对鲁棒保性能控制策略的有效性与优越性进行了验证。研究结果表明:在信号传输存在时滞、转向输出存在滞后的状态下,所提出的控制策略能够产生光滑平顺的控制输出,可保证车辆的行驶稳定性;尤其是在低附着、高速转向等恶劣行驶工况下,所提出的鲁棒保性能控制策略能够有效补偿模型失配所带来的不利影响,实现轨迹精确跟踪与横向稳定控制的有效兼顾。因此,该算法可在统一架构下综合考虑多种不确定性因素对车辆运动控制的影响,进一步提升车辆在复杂行驶工况下的轨迹跟踪控制性能,也可为数据共享环境下的车辆运动控制提供有意义的参考与技术支撑。

External disturbances,parameter perturbances,data transmission delays,and steering lags pose severe challenges in trajectory tracking control.To overcome these challenges,a novel control strategy is proposed to provide reliable commands to intelligent vehicles in an unreliable data transmission environment based on robust guaranteed cost control theory.The dynamics of the steering subsystem were considered using integrated modeling.Consequently,an augmented uncertain lag-free model was established that can more precisely describe the response transition of the vehicle-road coupling system under the influence of the steering actuator dynamic characteristics.Subsequently,a delay-dependent stability criterion was constructed based on the Lyapunov-Krasovskii function.Given the model mismatches caused by the perturbation of the model parameters,the uncertain cross terms were transformed into deterministic equivalent terms by leveraging the inequality amplification approach.Moreover,the H∞and guaranteed cost indexes were introduced to suppress the generalized external disturbances and configure control preferences.The final robust feedback gains that consider the linear time-varying parameters were designed based on a linear-parameter-varying technique.Several driving maneuvers were performed to verify the effectiveness and superiority of the proposed strategy.The results demonstrate that the proposed controller can provide smooth and effective outputs to guarantee a stable transportation of the ego-vehicle in the simultaneous presence of data transmission delays and control lags.Moreover,the proposed control approach can compensate for the adverse effects caused by model mismatches and achieve a trade-off between trajectory tracking accuracy and dynamic stability under harsh driving conditions,such as low adhesion surfaces and cornering at high speeds.Therefore,our control strategy can successfully address the aforementioned problems in a unified framework and further improve the trajectory tracking performance of vehicles under complex driving conditions.Simultaneously,it provides meaningful reference and technical support for the motion control of a vehicle in an information-sharing environment.