Fumin Zhang

Fumin Zhang, Zhenwu Shi, Shayok Mukhopadhyay

This paper establishes a novel analytical approach to quantify robustness of scheduling and battery management for battery supported cyber-physical systems. A dynamic schedulability test is introduced to determine whether tasks are schedulable within a finite time window. The test is used to measure robustness of a real-time scheduling algorithm by evaluating the strength of computing time perturbations that break schedulability at runtime. Robustness of battery management is quantified analytically by an adaptive threshold on the state of charge. The adaptive threshold significantly reduces the false alarm rate for battery management algorithms to decide when a battery needs to be replaced.

Zhenwu Shi, Ningshi Yao, Fumin Zhang

This paper studies the operation and scheduling of electric loads in micro-grid, a highly automated and distributed cyber-physical energy system (CPES). We establish rigorous mathematical expressions for electric loads and battery banks in the micro-grid by considering their characteristics and constraints. Based on these mathematical models, we propose a novel real-time scheduling analysis method for priority-based energy management in micro-grid, named Significant Moments Analysis (SMA). SMA pinpoints all the crucial moments when electrical operations are requested among the micro-grid and establishes a dynamic model to describe the scheduling behavior of electric loads. Using SMA, we can check the scheduling feasibility and predict whether the micro-grid can generate enough power to support the execution of electric loads. In the case where the power is insufficient to supply load demands, SMA can provide accurate information about the amount of insufficient power and the time when the insufficiency happens. Simulated results are presented to show the effectiveness of the proposed analysis method.

Ruochu Yang, Chad Lembke, Fumin Zhang et al.

Underwater gliders have been widely used in oceanography for a range of applications. However, unpredictable events like shark strikes or remora attachments can lead to abnormal glider behavior or even loss of the instrument. This paper employs an anomaly detection algorithm to assess operational conditions of underwater gliders in the real-world ocean environment. Prompt alerts are provided to glider pilots upon detecting any anomaly, so that they can take control of the glider to prevent further harm. The detection algorithm is applied to multiple datasets collected in real glider deployments led by the University of Georgia's Skidaway Institute of Oceanography (SkIO) and the University of South Florida (USF). In order to demonstrate the algorithm generality, the experimental evaluation is applied to four glider deployment datasets, each highlighting various anomalies happening in different scenes. Specifically, we utilize high resolution datasets only available post-recovery to perform detailed analysis of the anomaly and compare it with pilot logs. Additionally, we simulate the online detection based on the real-time subsets of data transmitted from the glider at the surfacing events. While the real-time data may not contain as much rich information as the post-recovery one, the online detection is of great importance as it allows glider pilots to monitor potential abnormal conditions in real time.

Sravya Kondrakunta, Venkatsampath Raja Gogineni, Michael T. Cox et al.

Intelligent physical systems as embodied cognitive systems must perform high-level reasoning while concurrently managing an underlying control architecture. The link between cognition and control must manage the problem of converting continuous values from the real world to symbolic representations (and back). To generate effective behaviors, reasoning must include a capacity to replan, acquire and update new information, detect and respond to anomalies, and perform various operations on system goals. But, these processes are not independent and need further exploration. This paper examines an agent's choices when multiple goal operations co-occur and interact, and it establishes a method of choosing between them. We demonstrate the benefits and discuss the trade offs involved with this and show positive results in a dynamic marine search task.

Nathan Gaby, Fumin Zhang, Xiaojing Ye

We develop a versatile deep neural network architecture, called Lyapunov-Net, to approximate Lyapunov functions of dynamical systems in high dimensions. Lyapunov-Net guarantees positive definiteness, and thus it can be easily trained to satisfy the negative orbital derivative condition, which only renders a single term in the empirical risk function in practice. This significantly reduces the number of hyper-parameters compared to existing methods. We also provide theoretical justifications on the approximation power of Lyapunov-Net and its complexity bounds. We demonstrate the efficiency of the proposed method on nonlinear dynamical systems involving up to 30-dimensional state spaces, and show that the proposed approach significantly outperforms the state-of-the-art methods.

Ye Yuan, Junlin Li, Liang Li et al.

The study presents a general framework for discovering underlying Partial Differential Equations (PDEs) using measured spatiotemporal data. The method, called Sparse Spatiotemporal System Discovery ($\text{S}^3\text{d}$), decides which physical terms are necessary and which can be removed (because they are physically negligible in the sense that they do not affect the dynamics too much) from a pool of candidate functions. The method is built on the recent development of Sparse Bayesian Learning; which enforces the sparsity in the to-be-identified PDEs, and therefore can balance the model complexity and fitting error with theoretical guarantees. Without leveraging prior knowledge or assumptions in the discovery process, we use an automated approach to discover ten types of PDEs, including the famous Navier-Stokes and sine-Gordon equations, from simulation data alone. Moreover, we demonstrate our data-driven discovery process with the Complex Ginzburg-Landau Equation (CGLE) using data measured from a traveling-wave convection experiment. Our machine discovery approach presents solutions that has the potential to inspire, support and assist physicists for the establishment of physical laws from measured spatiotemporal data, especially in notorious fields that are often too complex to allow a straightforward establishment of physical law, such as biophysics, fluid dynamics, neuroscience or nonlinear optics.

Zhenwu Shi, Fumin Zhang

When multiple model predictive controllers are implemented on a shared control area network (CAN), their performance may degrade due to the inhomogeneous timing and delays among messages. The priority based real-time scheduling of messages on the CAN introduces complex timing of events, especially when the types and number of messages change at runtime. This paper introduces a novel hybrid timing model to make runtime predictions on the timing of the messages for a finite time window. Controllers can be designed using the optimization algorithms for model predictive control by considering the timing as optimization constraints. This timing model allows multiple controllers to share a CAN without significant degradation in the controller performance. The timing model also provides a convenient way to check the schedulability of messages on the CAN at runtime. Simulation results demonstrate that the timing model is accurate and computationally efficient to meet the needs of real-time implementation. Simulation results also demonstrate that model predictive controllers designed when considering the timing constraints have superior performance than the controllers designed without considering the timing constraints.