Embedded System Applications 

Abstracts of Current Research Subjects

An embedded system is a dedicated computer system to perform specified functions. Nowadays, an embedded system is commonly used in many areas including intelligent robots, advanced vehicles, HCI (human-computer interface), mobile communication, and factory automation. We are interested in performing researches and developments of embedded systems using microprocessors, FPGAs, and real-time OS.

(1) FPGA-based vision/speech processing accelerator 

Real-time processing of vision and speech signals is important to many application fields which require quick response from environmental events. Because the computational burden for processing vision and speech signals is quite heavy, it takes long time to use conventional computers for processing these signals. Our research is to design and implement the dedicated hardware architectures to accelerate the processing of vision/speech signals, which includes stereo vision, object tracking using particle filter and color feature, face detection, line detection, sound source localization, and sound noise reduction. We design these special-purpose systems using hardware description languages such as VHDL, and implement using FPGAs to show their real-time performance. 

(2) Distributed control systems for automotive applications 

Nowadays, a vehicle includes multiple ECUs (electronic control units). These ECUs are distributed and considered as real-time systems. We are interested in designing real-time embedded systems, their communication network, and the analysis of their reliability, fault-tolerance, and timing requirement. Our research includes: 1) to design a reliable hard real-time embedded system using standard software architecture such as AUTOSAR and OSEK for automotive control units, 2) to develop a gateway embedded system for in-vehicle networks such as LIN, CAN, and FlexRay, and 3) to improve a development environment for complex distributed real-time systems by using model-based programming.

(3) HCI using embedded systems 

Our aim is to make HCI more convenient by studying following research topics: “Usefulness”, “Usability”, and “Affect”. Our HCI research includes: 1) Level of Autonomy(LOA) – The interface for robots according to the LOA, 2) Test for user convenience – The measurement of convenience of input devices, 3) Virtual guard – The safeguard system to control remote robot, and 4) Non-contact user interface – The interface using cameras and infrared sensors. 

(4) Software platform and user interface of mobile phones 

In the mobile market, it is important to develop efficient mobile software platform and innovative user interface (UI). We are interested in analyzing mobile software platform, mobile UI, and a remote control system using a mobile phone. Our research includes: 1) to analyze and improve architecture of mobile software platform such as Google Android, LiMo (Linux Mobile), and Windows Mobile, 2) to develop an intuitive mobile phone UI, and 3) to control a remote system using a mobile phone.

(5) Motion controllers for motor and hydraulic systems 

Recently, high performance motion controllers with complex functions are required. We are interested in designing and implementing advanced motion controllers for various industrial applications. High performance CPU, real-time OS, and FPGA have been used for implementing our high performance controllers which can support multi-axis applications with precise position, velocity profile generation, synchronized control, and industrial network-based distributed control.

1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31
  1. Vision/Speech SoC
  2. Electronic Control Unit
  3. Motion Control
  4. Embedded System
  5. Simulation
  6. Neck Gesture Based User Interface