After working 10 years in the Center of Geographic Information Systems (GIS) at Georgia Tech, Dr. Tsai joined the tenure-track faculty in 2007 as an Associate Professor; I was promoted to full professor in 2014 in the School of Civil and Environmental Engineering and, also, was appointed as an adjunct professor in the School of Electrical and Computer Engineering at Georgia Tech. Dr. Tsai is a strong advocate of and leader in establishing interdisciplinary research teams and programs (including recruitment of researchers and graduate/undergraduate students with different backgrounds), advising of ECE and CS PhD students with ECE and CS faculty, and mentoring of graduate students from different departments. His research teams have produced innovative research in the development of novel algorithms and spatial and sensing technologies.
Dr. Tsai’s research focuses on the development of Spatial Information and Sensing Optimization (SISO) methodologies, concentrating on applications to roadway/infrastructure health and safety condition evaluation, prediction, and management with a special emphasis on pavements and signs assets.
Dr. Tsai’s unique strengths include 1) emerging sensing technologies and their optimization (2D imaging, 3D laser, Lidar, smart phones, UAV, 3D printing, telematics and CAV, and GPS/GIS technologies); 2) data science and advanced computation methods, including artificial intelligence and machine learning (AI/ML), computer vision, spatial data analysis and visualization, etc., for automatic infrastructure health and safety condition assessment; and 3) intensive experience in leading interdisciplinary research teams to perform innovative research. The following are some highlights of Dr. Tsai’s academic activities:
- A recognized worldwide leader in automatic detection and classification of pavement distresses using emerging 3D laser technology and ML. This is evidenced by
- Pioneering the competitively selected $3.5 million research projects, “Remote Sensing and GIS-enabled Asset Management (RS-GAMS) Phases I and II,” (2010-2014), sponsored by the US Department of Transportation Research and Innovative Technology Administration (USDOT RITA) program; these projects especially recognized for their first-time use of 3D laser technology in the US. Dr. Tsai’s research over the past ten years has been largely responsible for 3D laser technology emerging as a mainstream technology. More than 80% of state departments of transportation in the US have moved from manual visual inspection to automated pavement condition evaluation.
- Being competitively selected for the 2017 AASHTO High Value Research (HVR) Award (a national award) because of my research team’s successful implementation of 3D laser technologies and AI on a large-scale system.
- Deloping and publishing the national technical standard (.PSI) in 2021 on using 3D laser technology for automated pavement condition evaluation based on my team’s research outcomes. Dr. Tsai’s on-going research project (NCHRP 01-60) will develop and publish six national technical standards, including using 3D printing for the verification of 3D technology, by the end of 2022.
- Addressing the nation’s top priority needs on infrastructure investment by implementing 3D laser technologies and ML for automated pavement condition evaluation nationwide, starting with four research projects sponsored by GDOT, FDOT, TxDOT, and Caltrans.
- Being invited to give keynote speech in the international conference in 2021 on “Automated 3D Pavement Condition Evaluation Using Machine Learning for Optimized Asset Management,” and to give multiple presentations in the Transportation Research Board (TRB, 2022) on AI applications to automated pavement condition evaluation.
- Developed innovative crack detection algorithms and their performance measurements.
- The research outcomes have been published in diverse, prestigious journals with high impact factors and at professional conferences, including Computer‐Aided Civil and Infrastructure Engineering, ASCE Journal of Computing in Civil Engineering, IEEE Transactions on Pattern Analysis and Machine Intelligence, IEEE International Intelligent Transportation Systems Conference (ITSC), etc.
- A buffered Hausdorff distance method was developed to quantify the performance of different pavement distress segmentation algorithms; this has established a significant foundation for advancing innovative crack detection algorithm development.
- The developed theories and algorithms are fundamental; thus, they can be applied to broad crack detection of man-made infrastructures, including pavements, bridges, tunnels, dams, buildings, oil pipelines, ocean infrastructure, etc.
- Recent innovative research work that developed a novel crack detection algorithm by combining the strengths of both geometrical modeling with 3D minimal path algorithm and data-driven ML using 3D pavement surface data will make a major technical contribution and breakthrough to overcome the drawbacks of ML.
- Developed an innovative crack fundamental element (CFE) model with a topological representation of cracks to establish a mathematical foundation for modeling large-scale, in-field infrastructure crack characteristics to study real-world crack propagation behavior at multiple scales.
- Established a 10-year “Safer Road” plan (2014 – 2024) by developing and implementing algorithms, methodologies, technologies, and tools to make the nation’s roadways safer. The innovative curve safety assessment method, which uses low-cost smart phones, AI, and crowdsourcing, has been developed through our NCHRP IDEA-214. The overarching goal is to save thousands of lives per year.
- Secured more than $14M in research projects as a Principal Investigator, including four highly prestigious National Academy of Sciences National Cooperative Highway Research Program Innovations Deserving Exploratory Analysis (NCHRP-IDEA) grants and more than $3.5M from the intensively competitive USDOT RITA Remote Sensing and Spatial Information Technology Program.
- Elected Chinese Changjiang Scholar in 2009 (one out of 244 worldwide scholars in all areas, and the only one in transportation), one of the most prestigious scholar’s honor awarded by the Chinese government in recognition of Dr. Tsai’s scholarship and leadership in the field of sensor-based and spatially-enabled infrastructure management.
- Worked with the largest logistics company (SF express) on dynamic roadway mapping using low-cost smart phone and AI for SF’s effective truck routing and with automobile industry (Volvo group) to predict and optimize vehicle energy consumption up to 6% using ML and real-time vehicle data.
- Served as a PI to bring innovation and health to our community by building a world class interdisciplinary team with faculty and researchers having different backgrounds (transportation, logistics, planning, policy, social science, public health, psychology, robotics, safety design, assistive technology, asset management, etc.) to address the growing challenge on safety and mobility of our aging population (competitively selected for funding by GT Seed Grant – Building Teams and Moving Teams Forward).
- Advised 9 postdoctoral fellows; 20 PhD students (15 graduated) with 2 NSF Graduate Research Fellowships and 3 International Road Federation Fellows; 22 MS students with thesis (19 graduated), 4 with Eisenhower Transportation Fellowships; 101 Non-thesis MS students (99 graduated); and 51 undergraduates with 6 Georgia Tech President’s Undergraduate Research Salary Awards and 1 Best Undergraduate Research Award at Georgia Tech (the only student selected at Ga Tech).
- Taught 7 new courses, including AI for Smart Cities, Smart City Infrastructure, to prepare and excite our CEE students with emerging sensing technologies and data science, including AI and computer vision.
- Advised an interdisciplinary student team (SECURE) that just won the $5000 2022 CEE Entrepreneurial Impact Competition - Innovation Award (January 28, 2022).
In moving forward, Dr. Tsai would like to establish two major centers of research and technology: 1) a national center on “Smart Infrastructure Health Condition Assessment” using automatic pavement/infrastructure condition evaluation and optimized asset management with data science (ML, computer vision, and data analytics) and emerging sensing technologies (including but not limited to 2D imaging, 3D laser, Lidar technologies, UAV, smart phones, etc.); this will save great amounts of infrastructure investment money; 2) a national center on “Safer Roads”, using AI and low-cost smart phones to deploy innovative technologies to state and local level (counties/cities) transportation agencies; this will save lives. Dr. Tsai hope to generate the impact in the US first and then broaden its impact worldwide.
Dr. Tsai had also led more than $3.5M of competitively selected research projects on “Remote Sensing and GIS-enabled Asset Management (RS-GAMS)” and “RS-GAMS Phase 2”, sponsored by the USDOT Office of the Assistant Secretary for Research and Technology (USDOT/OST-R), from 2010 to 2014. These projects were designed to intelligently assess roadway asset health conditions by using emerging sensor technologies installed in the intelligent Georgia Tech Sensing Vehicle (GTSV), along with artificial intelligence and machine learning that Dr. Tsai and his research team has integrated and developed (http://www.news.gatech.edu/2013/09/05/road-warriors-gt-researchers-redefine-infrastructure-maintenance). Dr. Tsai’s GDOT research project of “Implementation of Automatic Sign Inventory and Pavement Condition Evaluation on Georgia’s Interstate Highway” has been competitively selected as the 2017 AASHTO High Value Research (HVR) Award because of its innovation and broad impact (https://ce.gatech.edu/news/national-group-honors-research-using-lasers-a...).
His research outcomes were featured in the National Academy of Sciences’ Ignition Magazine (http://www.trb.org/Publications/Blurbs/163652.aspx) because of their significant positive impact on asset management and roadway safety. One of the most noteworthy accomplishment is his development of a methodology to automatically detect sign condition changes using roadway images that are widely available. Although the developed methodology is applied for routine traffic sign asset inventory, it has promising applications for expedited infrastructure condition evaluation along with unmanned aerial vehicle (UAV) and smart phones, following natural disasters because of its automatic and non-contact nature.
He is also the PI of License Plate Recognition, sponsored by the State Road and Tollway Authority (STRA). He also works on logistics routing and scheduling for the United Parcel Service (UPS) and maximizing Port and Transportation System and logistics Productivity by Exploring Alternative Port Operation Strategies. Dr. Tsai is also working on a large-scale project on “Crowdsourcing Transportation Asset Management Using Low-Cost Mobile Devices” to create a synergy between logistics companies and transportation agencies to optimize safety, energy consumption, and logistics efficiency.
He has appointed by the National Academy of Science, Engineering and Medicine to serve in the technical committee, the United States National Cooperative Highway Research Program “NCHRP 20-102 (28) Prepared for Transportation Agencies in Workzones for Autonomous and Connected Vehicles” and “NCHRP 20-102 (06) Road Markings for Machine Vision”, two of 34 connected and automated vehicle research projects sponsored by the USDOT. Dr. Tsai served on the Expert Task Group (ETG) of the US National Strategic Highway Research Program II (SHRP II) for the Naturalistic Driving Study (NDS) to provide guidance on research focuses, including the use of computer vision or processing and analyzing big NDS data, from 2008 to 2015. He is also on the technical committee of the AFD 10 Pavement Management Systems of the Transportation Research Board in the National Academies. Since 2010, he has served as the Associate Editor of ASCE Journal of Computing in Civil Engineering. Dr. Tsai holds a professional engineering (PE) license.
Dr. Tsai has also initiated a new undergraduate course at Georgia Tech (ECE 2811, 3811/3812, 4811/4812) on Smart City Infrastructure Vertically Integrated Project (VIP) course, which is a novel education system since 2016 (for 12 continuous semesters). Since joining the tenure track faculty position in 2007, Dr. Tsai has developed seven new courses: 1) CEE 4803 AI for Smart Cities (Fall 2021), 2) CEE 8813 AI for Smart City Infrastructure (Fall 2019), 3) CEE 6652: Infrastructure Management: IT Applications; 4) CEE 4050: Infrastructure System Management (Spring 2021), including Dr. Tsai’s recent work on Complete Street Asset Management; 5) ECE 2811, 3811/3812, 4811/4812: Smart City Infrastructure -- interdisciplinary undergraduate courses for CEE, CS, ECE, ME, ISYE students (12 continuous semesters from Fall 2016 – Spring 2022); 6) CEE 8813: Pavement Technology (Spring 2019); 7) CEE 8813 Advanced GIS for Smart Cities (Spring 2020). In addition, Dr. Tsai has also continuously taught CEE 6621, GIS in Transportation. Dr. Tsai also plans to offer an undergraduate course on Data Analytics on Transportation Safety in Fall 2022 built upon his intensive research work at national and state levels on transportation safety using emerging sensor technologies, including low-cost smart phones, 3D laser technology, and Lidar technologies. These courses highlight Dr. Tsai’s unique strengths in 1) sensing technologies and their optimized use, 2) data science and data analytics, including AI/ML, computer vision, etc. for automatic infrastructure health and safety condition evaluation, and 3) cross-interdisciplinary teaching to attract the talents from different departments.
Dr. Tsai has been invited to give more than 120 presentations, including the keynote speeches of 1) “Emerging 3D Methodologies for Intelligent Roadway Asset Management” at the 3rd International Conference on Transportation Information and Safety (ICTIS 2015) in Wuhan, China, in 2015, and 2) “Smart Cities: Mapping and Predicting Roadway Condition Using 3D and AI Technologies,” at the 11th International Conference of Mobile Mapping Technology (MMT 2019) in Shenzhen, China in August, 2019, and 3) “Automated 3D Pavement Condition Evaluation Using Machine Learning for Optimized Asset Management”, at the 12th International Conference on Road and Airfield Pavement Technology (ICPT), July 14 – July 15, 2021, Sri Lanka.
To work closely with industry, especially on transportation safety, logistics, energy, Dr. Tsai was also invited to talk on "Dynamic Mapping of Infrastructure and Road Conditions" in the Connected Fleets USA 2017 on September 26, 2017, and on “Dynamic Mapping of Roadway Health and Safety Conditions Using Artificial Intelligence” at the Connected Fleets USA 2019 on November 13, 2019. Dr. Tsai has worked with logistics companies, including UPS and SF Express on their effective truck logistics and value-added using low-cost smart phones and AI on dynamic roadway mapping. Dr. Tsai is currently working with automobile company (Volvo) to predict and optimize vehicle safety and energy consumption using AI and their real-time vehicle data (buses/trucks).
Dr. Tsai’s Interdisciplinary Research Team at Georgia Tech includes researchers and students with the diverse backgrounds in CEE, CS, ECE, and IE.