Page 91 - Undergraduate Research Opportunities Program Proceedings 2017-18

School of Engineering
Department of Electronic & Computer Engineering

A Robot Tour Guide

Supervisor: SHI Bertram Emil / ECE Co-Supervisor: LIU Ming / ECE
Student: NARAYANA SWAMY Thomas Rohit / IIM Course: UROP1100, Summer

This paper is a continuation and further development on the previous Robot tour guide project for the Electronic and
Computer Engineering Department at the Hong Kong University of Science and Technology. The focus of this paper is on
the integration work and new implementation techniques to further improve the reliability of the system and also get
everything working together as one system. The following paper compares a wide range of ROS SLAM packages such as
Gmapping and Google Cartographer. The other focus of this paper will be on the integration technique of the various
components of this robot tour guide project such as the user localization based on the Wi-Fi localization method
implemented and posting the users location to the robot directly. Furthermore developing the on bot table app that
allows the user to post desired destinations directly to the robot system, which returns a real time map navigation for
the user to look at.

Fleet Formation Control

Supervisor: SHI Ling / ECE
Student: CHEN Yifei / ELEC Course: UROP1100, Summer
CHEN Zhuo / CPEG UROP1100, Summer
DU Donghong / ELEC UROP1100, Summer
LYU Shengrui / ELEC UROP1100, Summer

Our goal is to realize the fleet formation control in our FYP, which entails the technique of designing a rational path for
individual boat, analyzing the path and realizing the relevant control on the boats and finally letting the group of ships
‘interact’ in effective way and adjust their path to make the whole transportation system work properly. The whole
project will be divided into 3 milestones: individual path planning, eschewing obstacle during the path and finally the
fleet formation controlling. For our UROP project, and main endeavor will be placed in the first part.

Robot Maze

Supervisor: SHI Ling / ECE
Student: XU Xinyuan / SENG Course: UROP1100, Summer

The project is to generate a random maze each time, using searching and AI algorithm to solve it and finally integrating
the algorithm to the real-world robot so that it can solve a real-world maze. The robot only knows the location where it
started and where the exit is. It needs to explore the maze, remember it and finally find a way from the origin to the
destination. Equipped with some basic AI algorithm, it can find the optimal way to solve the maze at the second time
after exploring at the first time. The integration of algorithm and robot is still in progress.

Indoor Autonomous Driving Robots
Supervisor: SONG Shenghui / ECE
Student: CHENG Tim Ta-ying / CPEG Course: UROP1100, Spring

This report aims to identify the process of using machine learning models to explore patterns and compare with
traditional models to investigate the effectiveness. The exploration can be separated into two aspects: Applying linear
regression model on pressure sensor values to determine a person’s position on the mattress and using neural network

81