International Journal of Modelling, Identification and Control (3 papers in press)

Regular Issues

• Open Access

  Constant pressure PID control of non invasive intracranial pressure monitor based on wavelet neural network ( Free Full-text Access )

  by Gaigai Zhang, Qingyun Yang 
  Abstract: To enhance the accuracy of constant pressure control in non-invasive intracranial pressure monitors, a study on constant pressure PID control of non-invasive intracranial pressure monitors based on wavelet neural networks is proposed. Collect intracranial pressure signals from non-invasive intracranial pressure monitors through sensors and preprocess them using wavelet filtering; Construct a wavelet neural network to learn the intrinsic patterns and features in signal data, and estimate the true intracranial pressure values; Based on the estimation results from the wavelet neural network, the parameters of the PID controller are dynamically adjusted and input as feedback signals. Adjust the non-invasive intracranial pressure monitor according to the magnitude and trend of the deviation value. The results show that the proposed method can achieve a steady-state error of less than 0.1mmHg, with a maximum overshoot of 6.2% and a maximum sensitivity of only 0.20, demonstrating high control accuracy and stability.
  Keywords: non-invasive intracranial pressure monitor; wavelet neural network; PID controller; constant pressure control.
  DOI: 10.1504/IJMIC.2025.10073014
   
  
• Outlier detection algorithm based on deviation characteristic  
  by Yong Wang, Hongbin Wang, Pengcheng Sun, Xinliang Yin 
  Abstract: Outlier mining focuses on researching rare events through detection and analysis to dig out the valuable knowledge from them. In the static data set environment, the traditional LOF algorithm calculates the local outlier factor through the whole data set and requires a lot of computing time. To solve this problem, the algorithm divides the data space into grids, and calculates the local outlier factor based on the centroids of the grids. Since the grid number is less than data point number, the time complexity is obviously reduced under acceptable error. When the new data points are added, it can rapidly detect outliers. The contrast experiment results show that the new algorithm can reduce the computation time and improve the efficiency, while achieving comparable accuracy.
  Keywords: outlier detection; local outlier factor; deviation characteristic; fast LOF detection algorithm.
  
  
• A lightweight attention mechanism and self-supervised denoising approach for robust vehicle detection in adverse weather conditions  
  by Lina Sun 
  Abstract: Vehicle detection under adverse weather conditions remains a major challenge due to severe image degradation and complex noise. Traditional denoising methods often fail to retain essential features, while lightweight models typically trade accuracy for efficiency, limiting real-time application. This paper presents a self-supervised denoising framework tailored for foggy and rainy scenarios. It includes three modules: a global perception mask mapper for identifying noise regions, a denoising network that separates clean and noisy components, and a regularised re-visibility loss to enhance performance in blind spots. To ensure deployment feasibility, a lightweight attention module based on the general meta-mobile block is introduced, balancing speed and accuracy. Experiments on benchmark datasets demonstrate notable gains in image clarity and vehicle detection accuracy. The framework offers a robust, efficient solution for real-world systems like autonomous vehicles, and lays the groundwork for future research on adaptive denoising in low-visibility environments.
  Keywords: self-supervised learning; image denoising; vehicle detection; attention mechanism; model lightweighting.
  DOI: 10.1504/IJMIC.2025.10072537