ISCTIS 2021 Keynote Speakers
Prof. Jingbo Xia, Xiamen University Tan Kah Kee College, China
Speech Title: Research on UWB/BLE-based Fusion Indoor Positioning Algorithm and System Application
Abstract: The accelerating advancement of 5G networks, Internet of Things, and data processing methods in the 21st century raised people's demand for location-based services, especially the indoor locating services. Therefore, indoor locating services have gradually replaced outdoor locating services to become a new research focus and challenge. Given that the traditional indoor locating systems, which are largely based on positioning technologies such as WiFi and radar, performed unsatisfactorily in terms of cost and, the most vital, accuracy.The Wireless Sensor Network (WSN) has become the core of the Internet of Things and has been widely utilized to tackle the difficulties in the indoor location during the past ten years. The present study analyzed and compared the working principles, research status, and the pros and cons of several commonly used wireless sensors, and drew a conclusion that single indoor locating technology inevitably comes with corresponding limitations, which can be solved through the combination of multiple complementary locating technology. We employed UWB and BLE as the subsystems of the indoor combination locating system, analyzed and optimized the respective locating algorithm model of each subsystem, and proposes an optimization model for combination locating algorithm based on EKF-PF, to enhance the accuracy. Eventually, we built a novel set of low-cost, convenient and flexible indoor combination locating system based on Bluetooth-UWB technology that supports both two-dimensional and three-dimensional spatial location.
Prof. Xizheng Zhang, Hunan Institute of Engineering, China
Speech Title: Research on Adaptive robust control strategy for hybrid energy storage system of electric vehicles
Abstract: Air pollution, global warming and oil depletion are the most concerning problems in the world. Gradually strict emission and fuel efficiency standards have prompted people to actively develop safer, cleaner and more efficient electric vehicles(EVs). Hybrid Energy Storage System (HESS) refers to a combination of two or more different energy storage types, which is composed of many individual energy storage elements or cells. HESS plays a significant role in the energy provision and economic benefit of EVs owing to its inherent adaptability to the fluctuating current flow of the load-required power.
Aiming at the problems such as power coordination control and operation interval optimization of HESS under complex working conditions, our research adopts an adaptive mechanism to estimate system parameters, and a L2 gain adaptive robust control strategy is proposed to control the battery/super capacitor current and DC bus voltage. The simulation results show that the control performance is significantly improved compared with the traditional PID, sliding mode and passive control.
Prof. Yongtang Zhang, Neusoft Institute Guangdong, China
Speech Title: A C-RAN Dynamic Resource Allocation Method for DRL
Abstract: Mobile edge computing (MEC) technology has emerged as a promising example of Cloud radio access networks (C-RAN) providing near distance services, thereby reducing service latency and reducing energy consumption. A multi-user MEC system is considered to solve the problem of computing unload policy and resource allocation policy. We took deferred total cost and energy consumption as optimization objectives to obtain an optimal strategy in a dynamic environment. An optimization framework based on Deep reinforcement learning (DRL) is proposed to solve the problem of resource allocation. Deep neural network (DNN) is used to estimate the value function of critics, and the information is directly extracted from the current state without the need to obtain accurate channel state. Thus, the state space complexity of the optimization target is reduced. The participants used another DNN to represent the parametric random strategy and improved the strategy with the help of the critics. The simulation results show that compared with other schemes, this scheme significantly reduces the total power consumption.
Prof. Xiaofang Yuan, Hunan University, China
Speech Title: Multi-objective Path Planning for Vehicle on 3D Map
Abstract: In the traditional vehicle area, energy-based motion control technology and battery technology usually use to solve the energy-saving problem. Our research proposes a new solution from the perspective of path planning. For vehicles traveling on the complex 3D terrains, the energy consumption of up-slope is far greater than that of the flat road and down-slope. To realize this, a path with a good trade-off between the energy consumption and distance would be the expected route for electric vehicle and mining transportation. A novel multi-objective path planning method is investigated to solve this problem for EV and mining truck. The simulation experiments prove that the proposed method can generate an optimal path which saves much energy in comparison with the path provided by the distance-based method.