Pre-conference Workshop 1：Control and Optimization for Networked Systems
Speaker: Ling Shi
Affiliation: Hong Kong University of Science and Technology, China
Title: Sensor Scheduling for Remote State Estimation with Unknown Communication Channel Statistics: A Learning-Based Approach
Abstract: We consider sensor scheduling with unknown communication channel statistics. We formulate two types of scheduling problems with the communication rate being a soft or hard constraint, respectively. We first present some structural results on the optimal scheduling policy using dynamic programming and assuming the channel statistics is known. We prove that the Q-factor is monotonic and submodular, which leads to the threshold-like structures in both types of problems. Then we develop a stochastic approximation and parameter learning frameworks to deal with the two scheduling problems with unknown channel statistics. We utilize their structures to design specialized learning algorithms. We prove the convergence of these algorithms. Performance improvement compared with the standard Q-learning algorithm is shown through numerical examples.
Speaker: Fumin Zhang
Affiliation: Georgia Institute of Technology, USA
Title: Bio-Inspired Autonomy for Mobile Sensor Networks
Abstract: There is an increasing trend for robots to serve as networked mobile sensing platforms that are able to collect data and interact with humans in various types of environment in unprecedented ways. The need for undisturbed operation posts higher goals for autonomy. This talk reviews recent developments in autonomous collective foraging in a complex environment that explicitly integrates insights from biology with models and provable strategies from control theory and robotics. The methods are rigorously developed and tightly integrated with experimental effort with promising results achieved.
Affiliation: The University of Tennessee, USA
Title: Nonlinear Modal Decoupling: a new paradigm of stability analysis and control for power systems and other multi-oscillator systems
Abstract: Anharmonic oscillations or vibrations are general phenomena of real-life dynamical systems such as power grids, mechanical systems and biological systems. Such systems can be modeled as large multi-oscillator systems and their global stabilities and behaviors away from equilibria are fundamentally difficult to analyze. For nonlinear multi-oscillator systems, we propose a new methodology named Nonlinear Modal Decoupling (NMD) that inversely constructs as many decoupled nonlinear oscillators as the system’s oscillation modes of interests. These decoupled oscillators together provide a fairly accurate representation of the original system’s behaviors within an extended region about the equilibrium under small and large disturbances. Every individual decoupled oscillator has only one degree of freedom and hence can easily be analyzed as a two-body problem, yet inferring dynamics and stability of the original system associate with the corresponding mode. This talk will present the NMD methodology in detail and use multi-generator power systems as examples to demonstrate NMD-based stability analysis and control. This new methodology can also be applied to nonlinear, multi-oscillator systems from other fields.
Speaker: Hua Geng
Affiliation: Tsinghua University, China
Title: Operation of Large-scale Renewable Energy Conversion System: from unit to Cluster
Abstract: In order to cope with the worsening environmental and energy crisis, it has become an inevitable trend to replace fossil-based traditional (thermal) power generation with renewable energy such as wind power and photovoltaic power. Compared with traditional power generation, renewable energy generation has the characteristics of clustering, distribution and power electronic based, which brings many technical challenges in terms of control and optimization. This talk discusses the challenges towards safe and reliable operation of large-scale renewable energy cluster from the perspective of individual synchronization and cluster cooperation.
Pre-conference Workshop 2：Advances on Brain Signal Processing
Affiliation: University of Electronic Science and Technology of China
Title: Probing Brain with EEG Brain Networks
Abstract: As for the high level cognition process, brain involves multiple brain areas functioning as a network to process the information. Moreover, various mental disorders have been proved to be closely related to the abnormality of brain network. EEG with merits of easy operation, mobility and especially high temporal resolution is potential to reveal the dynamic brain networks during cognition process. However, due to the low SNR of EEG and volume conduction, it is very challenging to construct the reliable EEG networks. In this talk, we will introduce the network analysis specifically developed for EEG, which can reliably construct the connectivity, causality and time-varying networks based on scalp EEG. Then, we will talk about how to apply the network analysis to brain computer interface, cognition and clinical studies.
Speaker: Ming Dong
Affiliation: Tianjin University, China
Title: Developments and Challenges of Brain-computer Interface
Abstract: A BCI is a system that measures brain activity and converts it into artificial output that replaces, restores, enhances, supplements, or improves natural CNS output. The system has drawn much attention from scientists across the world, due to its wide applications in military and medication. After decades of development, BCI has been updated quickly in paradigm design, algorithm and hardware innovations, resulting in further improvements in BCI performance. Here we mainly reviewed the development history of BCI, and focused on the latest research achievements and applications in civil use. Finally, we briefly discussed the existing problems in BCI technology, aiming at promoting the development of BCI.
Speaker: Xun Chen
Affiliation: University of Science and Technology of China
Title: Recent Studies on Artifact Removal of EEG Signals
Abstract: EEG signals have been widely used in clinical diagnosis, human-computer interaction, cognitive science research, etc. However, due to the weakness, EEG data are quite susceptible to various noise interferences, including electrooculogram (EOG), electrocardiogram (ECG), electromyogram (EMG), and motion artifacts. In this talk, we will first review the literature for EEG denoising issue. Then, we will present the near future trends and needs for EEG acquisition devices. In particularly, we will focus more on the challenging EMG artifact removal issue, considering its inevitability in the mobile situation. Subsequently, we will introduce an emerging technique, i.e. joint blind source separation (JBSS), based on which, we proposed a series of denoising frameworks for EEG in multichannel, few-channel and single-channel settings. We also explored the relationship among the three distinct denoising strategies and obtained interesting findings. Finally, according to our experience, challenges and recommendations will be given.