Application of deep learning in perception and navigation of mobile robotics

Mobile robotics are playing an increasing important role in many areas such as delivery, house cleaning, homecare, etc. How to increase the capability of robotics in perception and navigation is an on-going problem for decades. Due to the dynamic and uncertain chrematistics of the environment, it is still challenging for real application. In recent years, deep-learning has significantly changed the world in many sectors. It greatly enhanced the development of visual perception, object detection, speech recognition, natural language processing, etc. It provides an end to end learning structure that captures representations from big training data. In this project, deep learning techniques will be investigated and used to improve the current approaches for visual perception, path and motion planning of robotics to achieve satisfactory performance for real application.

Faculty Supervisor:

Clarence de Silva

Student:

Min Xia

Partner:

Vestorch Technology Ltd

Discipline:

Engineering - mechanical

Sector:

Information and communications technologies

University:

Program: