Robotics & Sensing Technology

This project proposes a learning from demonstration (LfD) system that allows robots to be not only taught by human via demonstration but also adjusted by themselves. Moreover, a multimodal perception approach is provided so that the robotic arm can perform flexible automation without complicated coding by professionals. In addition, an edge AI chip is implemented to increase the overall performances and flexibility, expanding the potential of the system to be incorporated with other applications. In the future, we will not only focus on vision-based tasks but also consider other scenarios where voice is required to be learned in the LfD system.

In this project, techniques including action recognition, object recognition and grasping strategies are developed using visual sensors based on deep learning approaches. Integrated with the hardware implemented visual simultaneous localization and mapping, a mobile learning from demonstration (LfD) robot system is designed and constructed, capable of reproducing behavior demonstrated by humans to achieve low volume automation.

This project goes on developing embedded AI deep learning technology focus on the ADAS/Self-driving applications. We develop the technology from five aspects, including automatic object labeling tool and labeled datasets, deep learning software and hardware technology development, various ADAS/Self-driving object detection and behavior predition technology, self-driving control technology as well as virtual simulation environment establishment for ADAS/Self-driving applications.

Based on deep learning technology and applying image analysis as the core, a variety of simple and interesting contactless human-computer interaction interfaces are constructed, including gestures, virtual keyboards, and handwriting text input in the air. These technologies provide users with a brand-new contactless interactive experience in many fixed-camera and mobile scenarios. We also implement value-added services and recommendation services regarding location information, so that users can effectively obtain their desired content, like well-classified photos of one specific site or the following journey recommendation.

We focus on the development of AI and multi-sensory data fusion technology. Based on the developed approaches, we can improve machine (i.e., robot) perception in a smart environment and increase the intelligence of robots. The core research topic here is how to utilize artificial intelligence to carry out a wide range of sensory data analysis, and then extract high-level information to enhance the efficiency of work as well as provide advanced services for human.

Using LINE Bot technology to develop an intelligent customer service system. The system support functions include:
1. Natural Language Processing transfer to SQL technology
2. Multi-round dialogue
3. User profiling
4. Intelligence search engine
5. Consumer behaviour pattern mining
6. Context-aware recommendation
Commercial IoT system information can also be integrated, so that businesses can collect more complete customer information, conduct deeper analysis (such as creating user profiling), and perform more flexible.