Cohort 1 projects

Title: Maximizing Recycled Materials for Sustainable Pavement Construction 

Fellow: Anand Sreeram

Theme: Smart Materials

Abstract: The construction of pavements utilizes a significant amount of non-renewable natural resources that generate substantial amount of greenhouse gases. One way to mitigate this issue would be using waste materials for new construction which will lower its overall carbon footprint, and in turn provide secondary benefits by diverting material from overburdened landfills and other processing facilities. However, the maximum utilization of the most relevant and compatible national waste streams into the pavement infrastructure with appropriate design methods is still missing. The goal of this research is to maximize the use of the most pertinent recyclable materials such as reclaimed asphalt pavement (RAP), recycled concrete aggregates (RCA) and waste plastics for new pavement construction through evaluating its potential to be incorporated into the roadway infrastructure from multi-scale perspectives. Based on an extensive review of literature, assessment of local construction practices and discussions with stakeholders, the maximum usage scenario and most compatible waste streams will be firstly identified. Following this, the structural performance of rigid pavements incorporated with these materials will be comprehensively evaluated, and its design optimized using machine learning (ML) approaches. To prepare for field evaluation, the use of low-cost radio frequency identification (RFID) sensors will be explored in order to integrate it into the modified pavement for its structural health monitoring. The outcomes of the study are expected to advance the understanding of this highly relevant and critical infrastructural issue and provide a field ready technology for potential large-scale adoption.

Title: Human-Robot Cooperation for Maintenance and Construction of Future Roads  

Fellow: Arsen Abdulali

Theme: Automation & Robotics

Abstract: Construction and maintenance of roads are tedious processes involving a considerable amount of manual work. The personnel working at the construction site usually perform difficult tasks, e.g., lifting and moving heavy loads, and are often exposed to severe weather conditions. In this project, we propose a novel approach of human-robot social cooperation, where the human operator remotely orchestrates the robot. The robot performs the manipulation routines with objects and the roadside environment in a semi-automated manner. The proposed human-in-loop design will move the roadside workers to the remote offices, which reduces the negative effects associated with health and safety. The applicant plans to develop the project in three steps. First, the applicant will develop the haptic-enabled simulator that allows the human operator to manipulate objects simulated in a digital twin road construction or maintenance. The proposed simulator can be later used as a training platform to teach new staff members. In the second stage, the algorithms developed for manipulation with a virtual environment will be implemented to operate a physical robot in a teleoperation setting. At this point, the main objective is to design control algorithms, as well as to develop an end-effector capable of sensing the physical contact and executing task-specific actuation. The last stage is the development of the concept of human-robot cooperation, where the operator controls the robot through abstract motion patterns(orchestration) rather than by employing direct coordination through teleoperation. The proposed project has potentially a high social and scientific impact and involves multidisciplinary and international collaboration.  

Title: Current and future material flows and stocks in roads

Fellow: Daniel Grossegger 

Theme: Sustainability

Abstract: Construction and maintenance of roads are tedious processes involving a considerable amount of manual work. The personnel working at the construction site usually perform difficult tasks, e.g., lifting and moving heavy loads, and are often exposed to severe weather conditions. In this project, we propose a novel approach of human-robot social cooperation, where the human operator remotely orchestrates the robot. The robot performs the manipulation routines with objects and the roadside environment in a semi-automated manner. The proposed human-in-loop design will move the roadside workers to the remote offices, which reduces the negative effects associated with health and safety. The applicant plans to develop the project in three steps. First, the applicant will develop the haptic-enabled simulator that allows the human operator to manipulate objects simulated in a digital twin road construction or maintenance. The proposed simulator can be later used as a training platform to teach new staff members. In the second stage, the algorithms developed for manipulation with a virtual environment will be implemented to operate a physical robot in a teleoperation setting. At this point, the main objective is to design control algorithms, as well as to develop an end-effector capable of sensing the physical contact and executing task-specific actuation. The last stage is the development of the concept of human-robot cooperation, where the operator controls the robot through abstract motion patterns(orchestration) rather than by employing direct coordination through teleoperation. The proposed project has potentially a high social and scientific impact and involves multidisciplinary and international collaboration.  

Title: Data science and advanced technologies for sustainability-orientated decision-making  

Fellow: Jinying Xu

Theme: Sustainability

Abstract: Sustainability is the continued protection of human health and the environment while fostering economic prosperity and societal wellbeing. It is a hit and urgent topic in both industry and academia. For future roads, sustainability is an inevitable theme to be addressed. One of the most intriguing problems in sustainability is its measurement and analysis. This project will focus on data-informed sustainability decision making in future roads. It aims to study what advanced technologies/techniques could be utilized and how they could be integrated and employed to enable data science which will then inform sustainability-orientated decision making for future roads. To fulfil this aim, the project will be organized along the four major stages of data science, i.e., data collection, data analysis, data communication, and decision making. Accordingly, six research objectives are outlined: (1) to expand data availability with advanced sensing and communicating technologies; (2) to develop a standardised data protocol; (3) to develop an advanced big data analytics solution that accounts for different aspects of sustainability in road system; (4) to communicate data analysis results using advanced digital twin and advanced data presentation technologies to decision makers timely, comprehensively, and concisely; (5) to develop appropriate decision making criteria; and (6) to develop an application system framework on top of the above data science technologies for data-informed sustainability-oriented decision making. Research methods to be employed for each of the research objectives and a quarterly research plan are also proposed.  

Title: Low Carbon Self-healing Concrete Pavements 

Fellow: Vahid Afroughsabet

Theme: Smart Materials

Abstract: The objective of this work is to introduce low carbon concrete into everyday concrete construction in UK. Portland cement production accounts for 5-8% of total global greenhouse gas emissions. This research addresses directly research priorities on low carbon concrete as established by the Low Carbon Concrete Routemap, for a path to net zero by 2050. Within the context of UK concrete construction, very little attention has been paid to exploit the current-generation alkali-activated concretes that make use of slag and fly ash and meet necessary requirements of standards and specifications. The proposed research will address this gap. This project also offers a value-added solution to convert local excavated waste shale and readily available clays into calcined clay supplementary cementitious materials that can partially replace Portland cement in concrete. Moreover, the road infrastructure sector remains largely traditional in its deployment of materials and design processes as well as in maintenance procedures. The road closures for repair and maintenance alone costing the UK £26.2 m/year. Therefore, the proposed research aims to revolutionise road infrastructure materials by development of effective self-healing systems for concrete pavements, and transit form traditional roads to more efficient, sustainable, and smart roads. This research explores the properties of concrete with superabsorbent polymers for the prevention of freeze-thaw damage, and expansive minerals and fibres for the prevention of early-age thermal shrinkage. Low carbon self-healing concretes will be developed, and laboratory material characterisation and performance will be assessed with a range of state-of-the-art  facilities at University of Cambridge.  

Title: Digital twin-driven Bayesian dynamic linear model and reinforcement learning for predictive maintenance of pavements 

Fellow: Yiming Zhang

Theme: Data Science

Abstract: Highway agencies are facing with challenges of ageing pavements, deteriorating networks, and limited budgets, highlighting the importance of effective maintenance strategies. Predictive maintenance is a prominent strategy to deal with the above issues, which enables pre-failure interventions based on the condition monitoring and prediction models. The digital twin (DT) integrates a physical-virtual connection that paves the way to predictive maintenance by monitoring, modelling, and analyzing through the lifecycle of pavements. However, the DT-based predictive maintenance suffers the following limitations:  

(1) anomalies that could affect the reliability of pavement performance prediction inevitably exist in the measurements; (2) the computationally expensive black-box models with inefficient interpretability and uncertainty consideration are usually adopted to predict pavement performance; and (3) it is challenging to design an efficient reinforcement learning (RL) architecture for maintenance planning optimization. This project aims to develop an autonomous and efficient decision-making tool through the DT, RL, and Bayesian dynamic linear model involving computational efficiency, readily interpretability, and uncertainty quantification. It will focus on establishing the system for automatic data anomaly detection and correction, developing an effective probabilistic prediction model for pavement performance, and presenting the RL framework with efficient modelling and training for maintenance decision-making. The technical results will be finally integrated into the DT platform. Three outcomes will be (i) work toward developing autonomous  and maintenance and communicated through publications, conferences, and meetings; (ii) development of the applicant’s modelling, analysis, and communication skills; and (iii) strategic  collaborations between partner organizations.