Human-Robot-AI Symbiotic Mobile Mapping Solution for Fast and Regular Examination of Civil and Environmental Engineering Infrastructures
Project Description
There are more than 30,000 old buildings and various types of civil and environmental engineering infrastructure in Hong Kong that require regular checking and maintenance, and some even require urgent repairs. The current practice of manual visual inspection is time-consuming, labor-intensive, and potentially inaccurate. Further aggravated by manpower shortages, these constraints could cause ineffective maintenance and lead to public safety issues, hindering the city’s daily operation. In addition, urban forestry in Hong Kong also faces a similar challenge. To address these pressing needs and transform Hong Kong into a truly smart and resilient city, this project aims to develop an integrated solution empowered by human-machine synergy: the Human-Robot-AI Symbiotic Mobile Mapping Solution for Fast and Regular Examination of Rapidly Evolving Civil Engineering Infrastructures of Hong Kong on a city scale.
Supervisor
WANG Yu-Hsing
Quota
5
Course type
UROP1000
UROP1100
UROP2100
UROP3100
UROP4100
Applicant's Roles
Participants need to be good at programming because they need to program for (1) AI model training, validation, and testing; (2) robot controls; (3) data fusion (for mobile mapping).
Participants need to be interested in new sensing technologies.
Applicant's Learning Objectives
1. Participants will learn how to apply AI (deep learning) for defect detection on infrastructures.
2. Participants will learn how to operate the robot and the mobile mapping system to automate the infrastructure inspection.
3. Participants will learn AI quality control.
4. Participants will learn how to implement acting and continuous learning to enable efficient training for the AI models.
Complexity of the project
Challenging