Advancing Hyperspectral and Multispectral Image Fusion: An Information-Aware Transformer-Based Unfolding Network.

In hyperspectral image (HSI) processing, the fusion of the high-resolution multispectral image (HR-MSI) and the low-resolution HSI (LR-HSI) on the same scene, known as MSI-HSI fusion, is a crucial step in obtaining the desired high-resolution HSI (HR-HSI). With the powerful representation ability, convolutional neural network (CNN)-based deep unfolding methods have demonstrated promising performances. However, limited receptive fields of CNN often lead to inaccurate long-range spatial features, and inherent input and output images for each stage in unfolding networks restrict the feature transmission, thus limiting the overall performance. To this end, we propose a novel and efficient information-aware transformer-based unfolding network (ITU-Net) to model the long-range dependencies and transfer more information across the stages. Specifically, we employ a customized transformer block to learn representations from both the spatial and frequency domains as well as avoid the quadratic complexity with respect to the input length. For spatial feature extractions, we develop an information transfer guided linearized attention (ITLA), which transmits high-throughput information between adjacent stages and extracts contextual features along the spatial dimension in linear complexity. Moreover, we introduce frequency domain learning in the feedforward network (FFN) to capture token variations of the image and narrow the frequency gap. Via integrating our proposed transformer blocks with the unfolding framework, our ITU-Net achieves state-of-the-art (SOTA) performance on both synthetic and real hyperspectral datasets.

Predictive Neural Network Modeling for Almond Harvest Dust Control.

This study introduces a neural network-based approach to predict dust emissions, specifically PM2.5 particles, during almond harvesting in California. Using a feedforward neural network (FNN), this research predicted PM2.5 emissions by analyzing key operational parameters of an advanced almond harvester. Preprocessing steps like outlier removal and normalization were employed to refine the dataset for training. The network's architecture was designed with two hidden layers and optimized using tanh activation and MSE loss functions through the Adam algorithm, striking a balance between model complexity and predictive accuracy. The model was trained on extensive field data from an almond pickup system, including variables like brush speed, angular velocity, and harvester forward speed. The results demonstrate a notable predictive accuracy of the FNN model, with a mean squared error (MSE) of 0.02 and a mean absolute error (MAE) of 0.01, indicating high precision in forecasting PM2.5 levels. By integrating machine learning with agricultural practices, this research provides a significant tool for environmental management in almond production, offering a method to reduce harmful emissions while maintaining operational efficiency. This model presents a solution for the almond industry and sets a precedent for applying predictive analytics in sustainable agriculture.

Design of an Integrated Controller for a Sweeping Mechanism of a Low-Dust Almond Pickup Machine.

California is the world's biggest producer and exporter of almonds. Currently, the sweeping of almonds during the harvest creates a significant amount of dust, causing air pollution in the neighboring urban areas. A low-dust sweeping system was designed to reduce the dust during the sweeping of almonds in the orchard. The system includes a feedback control system to control the sweeper brushes' height and their angular velocity by adjusting the forward velocity of the harvester and the brushes' rotational speeds to avoid any extra overlapping sweeping, which increases dust generation. The governing kinematic equations for sweepers' angular velocity and vehicle forward speed were derived. The feedback controllers for synchronizing these speeds were designed to optimize brush/dust contact to minimize dust generation. The sweepers' height controller was also designed to stabilize the gap between the brushes and the orchard floor and track the road trajectory. Controllers were simulated and tuned for a fast response for agricultural applications with less than a second response delay. Results showed that the designed system has acceptable performance and generates low amounts of dust within the acceptable range of California ambient air quality standards.

A reinforcement learning-based near-optimal hierarchical approach for motion control: Design and experiment.

As a data-driven design method, model-free optimal control based on reinforcement learning provides an effective way to find optimal control strategies. The design of model-free optimal control is sensitive to system data because it relies on data rather than detailed dynamic models. A prerequisite for generating applicable data is that the system must be open-loop stable (with a stable equilibrium point), which restricts the data-based control design methods in actual control problems and leads to rare experimental studies or verification in the literature. To improve this situation and enrich its applications, we propose a pre-stabilized mechanism and apply it to the motion control of a mechanical system together with a reinforcement learning-based model-free optimal control method, which constitutes a so-called hierarchical control structure. We design two real-time control experiments on an underactuated system to verify its effectiveness. The control results show that the proposed hierarchical control is quite promising in controlling this mechanical system, even though it is open-loop unstable with unknown dynamics.

Optimal full-state feedback observer integrated backstepping control of chemical processes with unknown internal dynamics.

This paper studies the estimation and control problems of chemical processes with unknown internal dynamics. An observer with optimal full-state feedback characteristics for estimating the state variables and unknown dynamics is presented. Unlike other observers that need to know the frequency characteristics of the system, the pole of the proposed observer is determined automatically in a LQR formulation and the observer stability is also inherently ensured. In order to suppress the unknown internal dynamics, the proposed observer is then applied to the control design leading to an observer integrated backstepping control method. The proposed method does not depend on the detailed mathematical model of the system while the stability of the closed-loop system is guaranteed. The stability of the closed-loop system is proven in the Lyapunov sense. Extensive numerical simulations are presented to validate the proposed method.

Preparation and evaluation of osteogenic nano-MgO/PMMA bone cement for bone healing in a rat critical size calvarial defect.

The clinical outcomes of polymethylmethacrylate (PMMA) bone cement used to fill gaps or marrow cavities of bones and bone defects are limited due to poor handling properties, mismatched mechanical properties with natural bone and lack of osteogenesis for bone healing. In this study, a series of PMMA bone cements containing active nano-MgO particles (nano-MgO/PMMA) were prepared. The handling and mechanical properties were systemically evaluated according to an International Standardization Organization standard (ISO 5833:2002). The biocompatibility and osteogenic activity of nano-MgO/PMMA were also analysed in vitro. The osteogenic effects of nano-MgO/PMMA were assessed in a rat calvarial critical bone defect model. The addition of less than 15 wt% nano-MgO to PMMA improved the handling properties of PMMA. Compared with PMMA, the compression modulus and strength of 20MP (20 wt% nano-MgO to PMMA) decreased to 0.725 ± 0.023 GPa and 25.38 ± 2.82 MPa, respectively. In vitro studies with MC3T3-E1 showed that nano-MgO/PMMA had better biocompatibility than the PMMA group after 7 days of culture. The nano-MgO/PMMA groups showed more calcium nodules and higher osteogenic gene expression levels than PMMA after 12 days of osteogenic induction of the rat BMSCs. The in vivo studies analysed by micro-CT and histomorphology results proved that nano-MgO/PMMA could significantly enhance new bone formation. The mean new bone mineral density in the nano-MgO/PMMA group was 50% greater than that in the PMMA group. In addition, biomechanical tests showed that nano-MgO/PMMA was superior to PMMA in bone-bonding strength after 12 weeks implantation. Therefore, the nano-MgO/PMMA bone cement has good potential in joint fixation and bone defect filling applications.

New periodic-chaotic attractors in slow-fast Duffing system with periodic parametric excitation.

A new type of responses called as periodic-chaotic motion is found by numerical simulations in a Duffing oscillator with a slowly periodically parametric excitation. The periodic-chaotic motion is an attractor, and simultaneously possesses the feature of periodic and chaotic oscillations, which is a new addition to the rich nonlinear motions of the Duffing system including equlibria, periodic responses, quasi-periodic oscillations and chaos. In the current slow-fast Duffing system, we find three new attractors in the form of periodic-chaotic motions. These are called the fixed-point chaotic attractor, the fixed-point strange nonchaotic attractor, and the critical behavior with the maximum Lyapunov exponent fluctuating around zero. The system periodically switches between one attractor with a fixed single-well potential and the other with time-varying two-well potentials in every period of excitation. This behavior is apparently the mechanism to generate the periodic-chaotic motion.

A neural network-evolutionary computational framework for remaining useful life estimation of mechanical systems.

This paper presents a framework for estimating the remaining useful life (RUL) of mechanical systems. The framework consists of a multi-layer perceptron and an evolutionary algorithm for optimizing the data-related parameters. The framework makes use of a strided time window along with a piecewise linear model to estimate the RUL for each mechanical component. Tuning the data-related parameters in the optimization framework allows for the use of simple models, e.g. neural networks with few hidden layers and few neurons at each layer, which may be deployed in environments with limited resources such as embedded systems. The proposed method is evaluated on the publicly available C-MAPSS dataset. The accuracy of the proposed method is compared against other state-of-the art methods in the literature. The proposed method is shown to perform better than the compared methods while making use of a compact model.

Dynamics of delay-coupled FitzHugh-Nagumo neural rings.

This paper studies the dynamical behaviors of a pair of FitzHugh-Nagumo neural networks with bidirectional delayed couplings. It presents a detailed analysis of delay-independent and delay-dependent stabilities and the existence of bifurcated oscillations. Illustrative examples are performed to validate the analytical results and to discover interesting phenomena. It is shown that the network exhibits a variety of complicated activities, such as multiple stability switches, the coexistence of periodic and quasi-periodic oscillations, the coexistence of periodic and chaotic orbits, and the coexisting chaotic attractors.