Many real world systems, including smart mechatronics systems, can be better characterized by dynamic systems of non-integer order. Using non-integer order or fractional order calculus, fractional order systems can be modeled more authentically. Due to the nature of the infinite dimensional model, proper approximations to fractional order differentiator (s ^alpha , alpha isin Ropf) are fundamentally important. This paper contributed a new approximation scheme which is an extension of the well-established Oustaloup's approximation method. The benefits from using the proposed scheme are illustrated by numerical examples in both time and frequency domains

This paper presents a new practical tuning method for fractional order proportional and integral (FO-PI) controller. The plant to be controlled is mainly first order plus delay time (FOPDT). The tuning is optimum in the sense that the load disturbance rejection is optimized yet with a constraint on the maximum or peak sensitivity. We generalized $Ms$ constrained integral (MIGO) based controller tuning method to handle the FO-PI case, called F-MIGO, given the fractional order $α$⁠. The F-MIGO method is then used to develop tuning rules for the FOPDT class of dynamic systems. The final developed tuning rules only apply the relative dead time $τ$ of the FOPDT model to determine the best fractional order $α$ and at the same time to determine the best FO-PI gains. Extensive simulation results are included to illustrate the simple yet practical nature of the developed new tuning rules. The tuning rule development procedure for FO-PI is not only valid for FOPDT but also applicable for other general class of plants.

In recent years, benefiting from deep convolutional neural networks (DCNNs), face parsing has developed rapidly. However, it still has the following problems: (1) Existing state-of-the-art frameworks usually do not satisfy real-time while pursuing performance; (2) similar appearances cause incorrect pixel label assignments, especially in the boundary; (3) to promote multi-scale prediction, deep features and shallow features are used for fusion without considering the semantic gap between them. To overcome these drawbacks, we propose an effective and efficient hierarchical aggregation network called EHANet for fast and accurate face parsing. More specifically, we first propose a stage contextual attention mechanism (SCAM), which uses higher-level contextual information to re-encode the channel according to its importance. Secondly, a semantic gap compensation block (SGCB) is presented to ensure the effective aggregation of hierarchical information. Thirdly, the advantages of weighted boundary-aware loss effectively make up for the ambiguity of boundary semantics. Without any bells and whistles, combined with a lightweight backbone, we achieve outstanding results on both CelebAMask-HQ (78.19% mIoU) and Helen datasets (90.7% F1-score). Furthermore, our model can achieve 55 FPS on a single GTX 1080Ti card with 640 × 640 input and further reach over 300 FPS with a resolution of 256 × 256, which is suitable for real-world applications.

A fuzzy fractional-order PID control algorithm for a general type industrial temperature control system is proposed in this paper. In order to improve the production quality and controlled model accuracy, a fractional-order elementary system is used to describe the temperature control process. The gain coefficients of the proposed fractional-order PID controller is updated online based on a set of fractional-order fuzzy rules which are defined by Mittag–Leffler functions and follow fat-tailed distributions. Therefore, the proposed controller parameters could be auto-tuned according to model uncertainties, noise disturbance, random delay, and etc. Examples of the studied temperature control systems are shown to verify the effectiveness of the proposed controller. The superiority of fractional calculus is fully explored in the presented control methodology. The controlled temperature profile with the proposed algorithm could realize more satisfactory dynamic performance, better robustness respect to environment changes caused by internal and external disturbance.

Diabetic retinopathy (DR) is a common fundus disease that leads to irreversible blindness, which plagues the working-age population. Automatic medical imaging diagnosis provides a non-invasive method to assist ophthalmologists in timely screening of suspected DR cases, which prevents its further deterioration. However, the state-of-the-art deep-learning-based methods generally have a large amount of model parameters, which makes large-scale clinical deployment a time-consuming task. Moreover, the severity of DR is associated with lesions, and it is difficult for the model to focus on these regions. In this paper, we propose a novel deep-learning technique for grading DR with only image-level supervision. Specifically, we first customize the model with the help of self-knowledge distillation to achieve a trade-off between model performance and time complexity. Secondly, CAM-Attention is used to allow the network to focus on discriminative zone, e.g., microaneurysms, soft/hard exudates, etc.. Considering that directly attaching a classifier after the Side branch will disrupt the hierarchical nature of convolutional neural networks, a Mimicking Module is employed that allows the Side branch to actively mimic the main branch structure. Extensive experiments are conducted on two benchmark datasets, with an AUC of 0.965 and an accuracy of 92.9% for the Messidor dataset and 67.96% accuracy achieved for the challenging IDRID dataset, which demonstrates the superior performance of our proposed method.