Dr. Shen's research activities were mainly involved in computer science, sensors, mechanical engineering and data science. His current research areas include:
Transformation from physical objects to digital models is marked as a revolutionary step in human history toward the information age. Different types of sensors, contact or non-contact, can be used to facilitate such a transformation. The main problems of contact sensors, including touch trigger probes on Coordinate Measuring Machines (CMMs) or robotic arms, are their extremely low rate of data acquisition and failure in measuring small details. Although analogue probes can increase the measurement speed significantly, they still cause depression of deformable materials (rubber or foam) or objects (metal sheet or biological tissue). With the proliferation of non-contact optical sensors such as laser scanners and LIDARs in recent years, fast reconstruction of physical objects has been practiced in industry, medicine, military, and many other fields. We have conducted a series of studies on processing 3D data from Laser sensors, and will continue our efforts in extending the existing methods for processing LIDAR data in autonomous driving.
Feature extraction and machine learning are an effective way in analyzing 2D image data for autonomous vehicles. In the meantime, we also developed computed tomography method for materials science and energy sector. Evolutionary computing is a way for us to solve engineering problems with no gradient information.
We focus on the application of brain-computer interface in autonomous driving. It is one of effective ways to obtain the different states of drivers under various driving conditions.
We focus on the agent-based technology in the virtual testing of autonomous driving. Digital control and collision detection are implemented in our parallel computation framework.