针对目前解决遥操作系统网络变时延问题的控制方案存在稳定性与操作性间的矛盾、过于依赖被控对象模型精度和抗干扰性差等不足,提出了一种HSIC-Smith控制算法。首先,根据巡检小车典型的动态响应误差曲线,确定能恰当描述系统动态特性的特征模型;其次,在Smith预估原理的基础上,应用仿人智能控制算法对巡检小车不同的运动状态,采用开、闭环相结合的控制模式来减小Smith预估误差带来的影响,从而改善网络时延变化对巡检小车遥操作系统动态性能;最后,将该方案与常规的PID-Smith预估算法进行对比实验。实验结果表明:在存在网络变时延的条件下,该方案能对巡检小车系统迅速、准确地遥操作控制,且对网络时延变化具有较强的适应性。 A control scheme based on HSIC-Smith is proposed to solve the problem that the control scheme of variable-delay system in the teleoperation system has the problem of stability and operability, over-reliance on model accuracy and poor anti-interference. Firstly, according to the typical dynamic response error curve of the inspection car, the characteristic model that can describe the dynamic characteristics of the system properly is determined. Secondly, based on the Smith pre-estimation principle, the artificial intelligence control algorithm is applied to inspect the different movement states of the car, The closed-loop and open-loop control mode is adopted to reduce the impact of Smith’s prediction error, so as to improve the dynamic performance of the remote car operating system. Finally, this scheme is compared with the conventional PID-Smith prediction Algorithm for comparative experiments. The experimental results show that under the condition of network delay, this scheme can quickly and accurately control the inspection trolley system remotely, and has strong adaptability to the network delay variation.