1. Classification of causes of distribution power system outages using LSTM+FC. Classify the cause of outages given the data including current waveform, weather, faulted phase, etc. (Python, TensorFlow/Keras)

2. XCPonEthernet communication between CANape and ZedBoard Zynq-7000 ARM/FPGA SoC Development Board for online measurement and calibration, cooperated with Daimler. (C/C++, XCP, UDP)

3. FreeRTOS application on Xilinx ZedBoard for multi-tasking in motor control and diagnostic, cooperated with Daimler. (C/C++)

4. CAN/CAN FD communication between CANalyzer and Xilinx ZedBoard, cooperated with Daimler. (C/C++)

1. Vehicle position and context detection & prediction. Built geometric model plus neural network for detection and prediction (within 1 sec) of vehicle’s relative position on self-collected V2V dataset. (MATLAB)

2. Detection of driver stress in real-world driving environment using physiological signals (BR, HR, HRV, GSR) with convolutional neural network (CNN), cooperated with Toyota Research Institute. (Python)

Wireless charging and automatic parking: involving in wireless charging system design, validation and test; working with ADAS group for automatic parking.