skip to content

Digital Roads of the Future


Title: A graph-based approach for designing road digital twins

Fellow: Dr Junxiang Zhu

Theme: Digital Twins

Abstract: Digital twinning is open to different solutions, and despite the difference in technologies used, these solutions should enable (i) productization, (ii) interoperability, (iii) data security, (iv) futureproof, and (v) static and dynamic data curation. My project for the FUTUREROADS program aims to address the second challenge of the Digital Twin theme: DT2 How can we design a Road Digital Twin. This challenge is caused by questions related to the definition of Digital Twin, i.e., its components, structure, and stakeholders, such as what exactly is a road Digital Twin? What is it made of, and how is it structured? To answer these questions, the following elements are required: Element 1 (E1) - highways stakeholders’ user and information requirements for DTs, Element 2 (E2) - a road Foundation Data Model (FDM), Element 3 (E3) - a road Reference Data Library (RDL), and Element 4 (E4) - a road Integration Cloud Architecture (IA). To deliver such elements, four objectives are identified in this project, including Objective 1 – Identifying the specific information requirements for road digital twin (to deliver E1 and E2), Objective 2 - Developing a graph-based reference data library for road assets (to deliver E3), Objective 3 – Developing an extensible, future proof and scalable road integration cloud architecture (to deliver E4), and Objective 4 – Validating the proposed approach for road digital twinning.

Title: Advanced Planning and Building Permits through Road Digital Twins

Fellow: Dr Judith Fauth

Theme: Digital Twins

Abstract: In this era of resource scarcity, optimising existing resources is crucial. This holds true for sectors beyond traditional perspectives, including planning and building permit (PBP) processes. However, fundamental research in this domain is lacking, and the potential of digital twinning in PBP remains largely unexplored. To address this research gap, the proposed research aims to establish connections between road digital twins (RDTs) and the PBP sector to enhance PBP efficiency. The project consists of four work packages (WPs). WP1 will involve collecting suitable data and defining information packages through literature reviews, legal text analysis, industry collaboration, and empirical studies. In WP2, knowledge management techniques will be employed to organise the acquired information. An ontology will be developed to describe the knowledge and a rule set formulated to automate processing and inferencing queries. To address uncertainties in PBP processes, WP3 will focus on uncertainty analysis. A decision model based on fuzzy agency theory will be developed to enable objective decision-making and overcome uncertainties in regulations and interpretations. Lastly, WP4 will evaluate the business case for leveraging digital twin technologies in PBP. A comprehensive business case will be developed to assess the benefits, costs, and risks of using RDTs, highlighting the value of the project’s knowledge and solutions. In conclusion, the potential of RDTs to revolutionise PBP processes will be demonstrated. By leveraging knowledge management, uncertainty analysis, and comprehensive business case evaluation, this research will pave the way for efficient and informed decision-making, accelerating the digital transformation of the construction industry.

Title: Maintenance of  Road Digital Twin Using Multi-modal data

Fellow: Dr Shirin Malihi

Theme: Digital Twins

Abstract: A rich digital twin needs cost and effort which counteracts the value of digital twin and is not performed for most roads.  We lack the digital twining tools that generates digital twins of existing road assets and allow us to affordably maintain them. We need ways to update the digital twin.

In this fellowship program, I will follow up maintenance of a road digital twin considering time efficiency and customer needs. My proposed solution will enhance this virtual twin using multi-modal data including LiDAR, camera images, thermal images and GPR. This data will be processed using machine learning and big data analyses to enhance the Road DT (RDT) by augmenting objects of interest. First Objective of this project entails conducting an in-depth investigation into the suitability of different modalities in detection process and potential targets and defect that can be detected in an interactive decision-making approach. This analysis aims to identify the strengths and limitations of existing approaches and challenges of pavement surveys. Second objective of this project proceeds to detect the intended targets using the best modality upon  findings of objective 1. Learning-based methods, considering transferring knowledge from other modalities will be adapted considering time. Third objective focusses on combination of the detection process to the existing RDT. A rule-augmented element-based Simultaneous Pose and Correspondence method is customized according to the target domain. Forth Objective evaluates the results and augments other modalities spatial knowledge into RDT. Drawing from the insights gathered in former objectives, development of this system will deliver an enhanced RDT by a multi-phase augmentation of the intended targets in order to fully employ capabilities of RDT technology for pavement maintenance. The designed system will act as an intelligent and adaptable framework for customer needs in next phases.

Title: An integrated road asset monitoring system supported by probabilistic models and artificial intelligence

Fellow: Dr Fengqiao Zhang

Theme: Smart Materials

Abstract: Effective utilization of information is critical for road asset management. National Highways (NH) highlights the importance of data-driven proactive asset management. Despite the advancement of technologies like embedded sensors, smart cars, and Artificial Intelligence (AI) for data interpretation, NH has not fully capitalized on these innovations. Data from diverse sources often remain underutilized, indicating a lack of efficient integration tools for comprehensive understanding and decision-making in road asset management, both reactively and proactively.

This research aims to examine emerging technologies, integrating their data to improve our understanding of road asset performance, now and in the future. A shift from deterministic to probabilistic analysis is important. Given that every technology encompasses uncertainties that cannot be fully eliminated, it's essential to quantify these uncertainties and incorporate the data alongside its uncertainties. This integration can be accomplished via various probabilistic models such as Bayesian updating, as well as AI methods such as ensemble learning.

Moreover, as technologies continuously evolve, there will be a persistent necessity to identify, comprehend, and incorporate new advancements. Accordingly, this research will also establish a structured and future-proof system to integrate technologies dynamically for road monitoring. The system remains relevant, adaptable, and updated in the face of future developments. The proposed system will undergo validation through both laboratory and on-site experiments. Furthermore, we will provide a comprehensive user manual to facilitate NH in effectively implementing the proposed changes, both in the present and in the future.

Title: Enhancing equity, diversity, and inclusion in active mobility: a study on underrepresented groups' perspectives in road infrastructure planning in Cambridge

Fellow: Dr Khashayar Kazemzadeh

Theme: Sustainability

Abstract: Underrepresented groups, including women, older adults, and underrepresented males, face challenges that hinder their participation and impact equity, diversity, and inclusion EDI in urban road infrastructure. This research project aims to address these barriers, focusing on road infrastructure planning and design to improve the role of underrepresented groups in active mobility and promote greater EDI in ridership. The project consists of four work packages (WPs). WP1 involves conducting a comprehensive analysis of the state-of-the-art literature using bibliometric and systematic literature review techniques to identify key research themes. WP2 utilises stated preference experiments to investigate how specific infrastructure factors, such as different types of pavement distress (e.g., potholes and cracks), contribute to user discomfort. WP3 assesses factors within each infrastructure, such as the types of road users and speed differences, that contribute to concerns related to EDI in active mobility. Building on the findings from the previous work packages, WP4 employs a dedicated stated preference experiment to develop a letter-based index that captures EDI concerns among transport users in active mobility. This research project aims to inform road infrastructure design and promote EDI in active mobility. The insights gained will be invaluable for planners, policymakers, and partners, enabling them to prioritise inclusive improvements. By addressing the specific concerns and needs of underrepresented groups, this study will contribute to the creation of more equitable and inclusive urban environments. Findings facilitate evidence-based decision-making, allowing policymakers to implement targeted interventions and policies that effectively address EDI challenges in road infrastructure planning and design.

Title: Measuring and enhancing the resilience of road infrastructure networks to climate change

Fellow: Dr Jie Liu

Theme: Sustainability

Abstract: Climate change has raised a series of problems for the environment and society, which causes transport unreliable and vulnerable. Therefore, measuring and enhancing the resilience of Road Infrastructure Networks (RINs) is significant in building resilient transport and society. The outcomes of the project and their application in transportation are: 1.Enriching and improving the research theories and methods on traffic resilience;2.Measuring the impact of climatic events on RINs’ operation and assessing their resilience to climate change.3.Optimizing the maintenance strategy and restoration strategy for improving the ability of RINs to resist climatic events, which guides the operators to reduce the climatic impact on RINs.

Title: Assessing road infrastructure resilience against extreme weather events and fostering climate change adaptation

Fellow: Dr Zizhen Xu

Theme: Sustainability

Abstract: Infrastructure resilience is increasingly recognized as a crucial aspect of national security, especially considering climate change threats. Interest and efforts in this field are escalating, offering solutions for climate change adaptation parallel to ongoing mitigation strategies such as decarbonization. There is imperative need of resilience engineering in critical infrastructure sectors that daily lives depend on. Conventional risk management strategies are showing limitations when facing the challenges -- the uncertainty of changing environment and the growing complexity of infrastructure systems. This project is proposed to tackle these challenges through a complex network approach that integrates risk assessment with resilience framework encompassing preparing, absorbing, and recovery phase. We will develop the approach with a case study of UK national highways. A spatial network model will be employed to assess the infrastructure resilience, including quantifying the infrastructure’s exposure to potential extreme weather events, examining climate hazard scenarios at all disruption levels, and simulating cascading failures within system and from other supporting infrastructures. The ultimate goal is to answer the question of how ready our road infrastructure is for the changing climate and develop quantitative analysis to inform future interventions.

Title: Digital twin-driven structural health monitoring of roads by using physics-based model and machine learning

Fellow: Dr Zhaojie Sun

Theme: Data Science

Abstract: To construct a digital twin of roads, the structural health information of roads is essential. This information can be obtained by conducting Non-Destructive Testing (NDT). A promising NDT method to evaluate road structural health is the Traffic Speed Deflectometer (TSD) test, which can continuously measure the surface response of roads caused by moving loads at normal driving speeds. In this project, a parameter identification technique for the TSD test will be developed by using physical modelling and machine learning. At first, an extensive literature review and in-depth discussions with stakeholders will be conducted to identify the requirements and expectations of final deliverables. Then, a physics-based model for the TSD test of roads will be developed, and the performance of the developed model will be validated by using available data. Next, based on the database generated by the developed physical model, a machine learning-based parameter identification technique will be developed. At last, the performance of the developed parameter identification technique for practical applications will be evaluated. The developed parameter identification technique for the TSD test can provide structural health information for the digital twin of roads, and further help formulate cost-effective maintenance and rehabilitation strategies of roads. In addition, by carrying out different dissemination activities, this project aims to promote the development of the whole road industry.

Title: Artificial intelligence assistant autonomous-vehicle-mounted sensors based road surface condition monitoring system

Fellow: Dr Xiang Wang

Theme: Automation&Robotics

Abstract: Road transportation is an important component of transportation. Numerous roads to be monitored are a challenge for the management by the authorities. Untimely road maintenance endangers the safety of drivers and vehicles. Low-cost and high-efficiency Road Surface Monitoring (RSM) becomes an important target for future roads. Conventional RSM systems have the disadvantages of high costs and difficulty to be improved. Automated driving shows potential for future road transportation. Will autonomous vehicles mounted with non-conventional sensors monitoring roads be a future RSM solution? This research will develop a low-cost RSM system mainly based on autonomous-vehicle-mounted Inertial Measurement Units (IMUs) sensors. The onset of aquaplaning is a key research point and is expected to be monitored by the RSM system. Information processing in complex road environments is a major difficulty in this research. Data fusion methods and artificial intelligence will be developed for the RSM system. This research mainly consists of three stages. The first stage is to study the scenarios where there is a high risk of aquaplaning for vehicles. The second stage is to develop the RSM system for these scenarios and to study the signal acquisition of the onset of aquaplaning in the laboratory. The third stage is to conduct a field study in multiple real scenarios in Cambridge. This research will realize a low-cost vehicle-mounted-sensor mobile RSM system which is helpful to improve the driving safety of drivers and vehicles and improve road maintenance efficiency.