Sampling or task jitter affects the performance of digital control systems but realistic simulation of this effect has not been possible to date. Our previous work has developed a novel method to simulate sampling jitter in MATLAB/Simulink simulation software where the jitter is generated randomly. What has been missing is a way to capture sampling jitter from a target platform and then feed this timing information into the simulation. This paper presents a low-cost and novel solution to these problems. The method uses an Arduino board to capture task jitter from two different hardware platforms with multiple stressing conditions. Then the recorded performance data is used to drive realistic simulations of a control system. Measurement shows that the task jitter data does not follow any specific random distribution such as Gaussian or Uniform. Furthermore, very occasional timing patterns, which may not be picked up while testing a real system, can result in extreme controller responses. This novel method allows comparisons of different platforms and reduces the effort required to choose the most appropriate platform for full implementation.

采样或任务抖动会影响数字控制系统的性能，但迄今为止尚无法对此效果进行实际的仿真。我们以前的工作开发了一种新颖的方法，可以在MATLAB / Simulink仿真软件中模拟采样抖动，该抖动是随机产生的。缺少的是一种从目标平台捕获采样抖动，然后将此时序信息馈入仿真的方法。本文提出了针对这些问题的低成本且新颖的解决方案。该方法使用Arduino板来捕获来自两个具有多种压力条件的不同硬件平台的任务抖动。然后，所记录的性能数据将用于驱动控制系统的逼真的仿真。测量表明，任务抖动数据没有遵循任何特定的随机分布，例如高斯或均匀分布。此外，测试实际系统时可能无法拾取的非常偶然的时序模式可能导致极端的控制器响应。这种新颖的方法允许对不同平台进行比较，并减少了选择最合适的平台进行全面实施所需的工作。