Intelligent Additive Construction using Geomaterials

Modern developments have led to the availability of an extremely high level of computing power at low cost, while at the same time, material resources and CO2 are becoming increasingly expensive both economically and in terms of environmental impact. This has created the opportunity to envision a 'high intelligence, low resource' construction technology using computing power and robotics to allow complex adaptive assembly of unprocessed raw materials into suitably performing structures. This PhD project will pioneer practical aspects of this approach utilising little more than in-situ geo-materials i.e soils and raw cobbles/rocks, combined with minimal but strategically placed reinforcing materials where appropriate.

Geo-materials are generally good in compression and shear but lack tensile capacity. However, when strongly interlocked they show significant strength. Historic structures have worked with these materials and properties for millennia and there are many examples around the world of long-lived structures using soil and mortarless rock from the ubiquitous dry stone wall, through stone arches to large scale castles and fortresses. The challenge for modern construction using these approaches is speed, cost, precision and low risk.

The overarching hypothesis in this project is that it can be cost effective to take this 'high intelligence, low resource' approach. Specifically in this project the hypothesis is that such precision assembled systems (optimised both locally and globally) can outperform conventional low technology bulk earthworks approaches or high resource input approaches (e.g. use of concrete) in one or more construction scenarios.

The hypothesis will be tested through investigation of:

(i) optimal dry packing of cobbles/gravel and characterisation of their performance through conventional geotechnical testing
(ii) performance enhancement with minimal bonding applied to contact points using fine grained soils with/without additional binding agents
(iii) strategic inclusion of reinforcement elements

Experiments will be conducted on soil elements using conventional geotechnical test methods and later on, scale model earthwork constructions. At all stages the ability to automate the processes adopted will be evaluated and specifications written. Underpinning theory will also be developed, together with pilot systems demonstrating the automation concepts, adapting existing robotic equipment.

Co-supervisors both from within the Civil Engineering Department (Dr Jonathan Black) and ACSE (Dr Jonathan Aitken) will give strong support to the cross-disciplinary aspects of this project.

Applications span a large range from low cost, locally sourced construction in developing countries (including use of demolition rubble), through low carbon resource construction in developed countries, to planetary applications (e.g. construction on the moon where raw material transport would be prohibitive).

A brief timetable of envisaged work is as follows:

Year 1: Literature review, testing of unbonded optimally packed materials
Year 2: Theory development. Investigation of bonding agents and tensile reinforcement
Year 3: Engineering application and automation model demonstrators. Thesis write up.

This project links in with and builds on existing projects and expertise within the Department, Faculty and University in masonry arch construction, chemical and biological stabilization of soils, robotics, soil reinforcement and engineering optimization, building a critical mass from which to springboard major projects in this area.