How AI could make “self-healing” roads a reality
The UK has a pothole problem, costing millions annually in repairs and causing frustration for drivers. But groundbreaking research, supported by Google Cloud’s artificial intelligence (AI), may offer a solution: self-healing roads made from biomass waste.
Researchers from King's College London and Swansea University, in collaboration with scientists in Chile, have developed a new type of asphalt that can repair its own cracks over time, eliminating the need for manual maintenance. This innovative material is inspired by nature, mimicking the self-healing abilities of trees and some animals. By replicating these regenerative qualities in asphalt, the team aims to create more durable and sustainable roads.
The production of asphalt for construction and maintenance in the UK is a massive undertaking, with over 20 million tonnes produced annually. While the industry is moving towards more sustainable practices by incorporating recycled materials like food waste, the issue of cracks and potholes remains.
AI paves the way
The exact causes of cracks are not fully understood, but they often stem from the hardening of bitumen, the sticky black substance used in asphalt mixtures, due to oxidation. Scientists are actively investigating this process to develop ways to reverse it and effectively "stitch" asphalt back together.
In laboratory experiments, the researchers demonstrated that the new asphalt material could heal a microcrack in under an hour. This self-healing process is facilitated by natural spore microcapsules and waste-based rejuvenators.
The team used machine learning, a type of AI, to study organic molecules in complex fluids like bitumen. The team developed a new data-driven model to accelerate atomistic simulations, advancing research into bitumen oxidation and crack formation. This approach is significantly faster and more cost-effective than traditional computational models.
Additionally, the researchers collaborated with Google Cloud to develop tools that can identify chemical properties and enable the creation of virtual molecules designed for specific purposes, similar to techniques used in drug discovery.
Dr. Francisco Martin-Martinez, a computational chemistry expert at King's College London who was part of Google Cloud’s Research Innovators Programme in 2022, was one of the researchers working on this project. He points out that mimicking nature in its ability to heal will expand the lifetime of our roads to pave the way towards a more sustainable and resilient road infrastructure. This is in part thanks to the collaboration with Google Cloud in developing AI tools for the rapid development of these self-healing pavements.
Dr. Jose Norambuena-Contreras, an expert in self-healing asphalt at Swansea University and winner of the prestigious RILEM Robert L’Hermite medal in 2024, sees this research as a significant contribution to the development of net-zero asphalt roads with enhanced durability.
While still under development, self-healing asphalt holds immense promise for improving infrastructure and promoting sustainability worldwide. This innovation aligns with the UK government's net-zero emissions goals and addresses the significant financial burden of road maintenance. By reducing the need for repairs and extending the lifespan of roads, self-healing asphalt could offer a cost-effective and environmentally friendly solution to the UK's pothole problem.