Autonomous Maintenance Plant & Innovative Materials

Highways form a critical part of modern infrastructure, enabling economic activity, mobility, and access to essential services. However, maintaining an extensive and ageing road network remains a complex and resource-intensive challenge. Conventional maintenance practices are often time-consuming, carbon-intensive, and disruptive to road users and surrounding communities. 

Road maintenance activities can contribute to traffic congestion, increased vehicle emissions, and operational safety risks for both maintenance personnel and road users. As road networks continue to age and traffic demand increases, there is a growing need for maintenance approaches that can deliver faster, more precise, and less disruptive interventions. 

To address these challenges, this work presents the Autonomous Maintenance Plant (AMP), a digitally enabled and semi-autonomous highway maintenance system developed within the Digital Roads Prosperity Partnership. The AMP combines robotic automation, intelligent control, digital sensing, and low-carbon repair technologies to support more efficient road maintenance operations. 

By automating key maintenance activities, the AMP aims to reduce intervention and repair times, improve the safety of maintenance operations, minimise traffic disruption, and reduce the environmental impact associated with conventional road maintenance. The system provides a platform for exploring how automation and digital technologies can contribute to more sustainable, efficient, and resilient highway infrastructure. 

The Autonomous Maintenance Plant (AMP) forms the third component of the Digital Roads cyber-physical platform, operating in close integration with the Digital Twin. The Digital Twin continuously analyses the condition of the road in real time, identifying emerging or incipient pavement damage. Based on this information, the AMP receives maintenance requests and autonomously/semi-autonomously performs the required investigation and repair interventions.

The AMP prototype is designed to detect and seal transverse, longitudinal, and randomly oriented cracks in asphalt concrete and asphalt pavement, targeting crack widths greater than 3 mm. 

Following each maintenance intervention, the AMP transmits updated road-condition data to the cloud, enabling the Digital Twin to be updated with the latest information on pavement condition and repair status. This continuous feedback mechanism progressively enhances the digital representation of the road network, supporting improved condition assessment, maintenance planning, and data-driven decision-making for future interventions. 

Innovative Materials

The AMP design utilises particle-based simulations of the enhanced repair materials developed by the Digital Roads Materials research team.  

The materials consisted of three commercial repair material products enhanced through microfibre addition:  

The three selected materials were; 

• Rapid Set Cement All (RS), a cement-based material;  
• Roadware 10 Minute Concrete Mender (RM), a polymer-based material; and 
• Polycot External Crack Filler (Pc), a polymer-based repair material.  

RS, RM and Pc are suitable for concrete pavements and Pc for asphalt pavements. 

The results proved that, mechanistically, short fibres (~1–2 mm) allow more fibres to be added while maintaining extrusion and enabling crack filling (6 mm wide). Fibres enhance mechanical properties like U through stable crack-bridging and progressive fibre pull-out. Despite their higher unit costs, fibre-enhanced systems demonstrate strong whole-life value. For a 6-km lane over a 10-year period, RS- and Pc-reinforced repairs yield substantial economic savings—more than halving total costs compared with hot-applied bitumen. Comparable reductions in whole-life global warming potential are also anticipated.  

Research papers have been published on this research; see below. The materials portfolio is ready for manual deployment today and is fully compatible with AMP extrusion in the near term, using existing supply chains and placement methods.  

The evidence base supports departures from current specifications toward the adoption of performance criteria that prioritise compatibility, toughness, and durability over peak strength alone. 

These materials have been used to capture the physical liquid flow behaviour, which was then transformed into a position-based physics simulation. The results of these studies validate its use for general robotic liquid manipulation applications, such as the AMP simulation.