skip to content
 

Infrastructure Digitization

Current Projects

AUTOMATED BIM GENERATION OF EXISTING AND UNDER CONSTRUCTION RAILWAYS October 2017 – September 2021

Sponsor: Bentley Systems UK Ltd. Grant, PI: Ioannis Brilakis, £93,574

The process is of as-is infrastructure modelling involves the generation of infrastructure objects and their relationships from low-level point cloud datasets and the design BIM file.This research is aimed at creating a viable approach to automate the generation of as-is geometric railway Building Information Modelling (BIM).The key novel idea that makes this possible is that railways follow a set of engineering design assumptions which can be used as guides for segmentation by Markov Chain Monte Carlo (MCMC) methods, region growing, or octrees.The assumptions can be formulated as rules-sets can be used to recursively segment the data multiple times down to the individual component level through deep learning methods, such as recurrent neutral networks and conditional random fields (CRF). Template priors and fitting methods could then be used to replace segmented high-level primitives with actual objects and their relationships. Leading researcher: Mahendrini Fernando. 

 

INTERACTIVE POINT CLOUD AND IMAGE DATA GENERATION IN INFRASTRUCTURE SCENES October 2017 – September 2021

Sponsor: Trimble Europe BV Grant, PI: Ioannis Brilakis, £32,000

The research is mainly focused on increasing the efficiency of geometry and visual data collection. Leading researcher: Gulsum Sevde Baltasi. 

 

Past Projects

DIGITALLY ENABLING THE DESIGN FOR MANUFACTURE, ASSEMBLY AND MAINTENANCE OF BRIDGES April 2015 – March 2017

Sponsor: UK Technology Strategy Board Grant, PI/Grant Funding: Laing O’ Rourke £1.2 mil, Co-I/Subamount: Ioannis Brilakis, £365.000

Bridges are largely designed as bespoke solutions with the majority of the work being carried out on site and in the case of improvements to and replacement of existing bridges involves disruption through lane closures and detours. The objective of the project is to develop an integrated digital delivery process for bridges and bridge parts. It will address the whole lifecycle of bridges from identification and rationalisation of needs to manufacture, assembly, operation, maintenance and decommissioning.  The output will be an interoperable set of digital tools, data schema and virtual prototyping processes that lead to the automated manufacture of a set of standardised, validated parts and sub-assemblies at a controlled price, configured virtually and in reality that are capable of meeting the requirements of the most common bridge types. 

 

CIG: AUTOMATED AS-BUILT MODELLING OF THE BUILT INFRASTRUCTURE September 2013 – August 2017

European Commission Grant, PI: Ioannis Brilakis, €100,000

Current automated methods and technologies for mapping and labeling existing infrastructure are not able to achieve a fully automated process that can generate as-built geometric infrastructure models. Although as-built modelling is significantly assisted by recent technological advancements, most of it heavily relies on manual inspection and analysis, which are error-prone and time-consuming. Further automation could be achieved with the help of object detection concepts. This research aims to automate the detection of the objects by creating new object detection methods from image and spatial data using computer vision and match the detected objects with standardized ones used in Building Information Modelling. Leading researchers: Marianna Kopsida, Ioannis Anagnostopoulos.

 

URBAN SCALE BUILDING ENERGY NETWORK October 2014 – September 2015

Sponsor: Engineering and Physical Sciences Council, PI / Grant Funding: James Keirstead, CE, Imperial, £28,868, Co-PI / Sub-amount: Ioannis Brilakis, £6,131

The Climate Change Act 2008 requires a 34% cut in 1990 greenhouse gas emissions by 2020 and at least an 80% reduction in emissions by 2050. Residential and commercial buildings account for 25% and 18% of the UK's total CO2 missions respectively and therefore have a significant role to play in a national decarbonisation strategy. As the UK has some of the oldest and least efficient buildings in Europe, there is substantial scope for improving the efficiency of energy end-use within UK buildings. The overall aim of this project was to establish a network of academics and practitioners to discover the knowledge gaps and practical obstacles that inhibit the rapid improvement of the thermal performance of the UK building stock and to devise hypotheses of theoretically-feasible solutions that could be used to solve these problems. 

 

FP7-PEOPLE-2009-IRSES: BIMAUTOGEN September 2012 – August 2015

European Commision Grant, PI: Ioannis Brilakis, €300,510

The purpose of this project was to facilitate international collaboration and transfer of knowledge between the AutoBIM consortium members (I.Brilakis, Cambridge; R.Sacks, Technion; S.Christodoulou, UofCyprus; M.Lourakis, FORTH; S.Savarese, UofM; J.Teizer, GATech) and to implement and test whether a novel framework can be successfully used to generate parametric building models of buildings, ranging from residential housing to industrial facilities, almost entirely automatically. The project’s most significant contributions was, not only the automation of several mundane and repetitive processes with the addition of visual and spatial pattern recognition concepts in the modelling workflow, but also the exchange of knowledge and the building of transatlantic research collaborations on cutting-edge and high-impact scientific projects through the exchange of interdisciplinary staff among partner institutions and joint training of research teams on thematic areas of common research interest.

 

RECIPROCAL RECONSTRUCTION AND RECOGNITION FOR MODELING OF CONSTRUCTED FACILITIES September 2010 – August 2015

NSF Grant #1031329, PI: Ioannis Brilakis, $306,043

The research objective of this project was to evaluate whether a novel framework proposed by the PIs can progressively reconstruct a reinforced concrete frame structure into an object-oriented geometric model, for the purpose of automating the Building Information Model (BIM) making process of constructed facilities in a cost-effective manner. According to the proposed framework, the modeler videotapes the structure from all accessible angles to minimise occlusions. During this stage, the structural members (concrete columns and beams in this study) in the resulting stream of images are detected and their occupying region is marked in all images. These regions are used to establish correspondence at the object level across images, and solve the rough registration problem efficiently. Line-based structure from motion is then applied to the result to produce a rendered 3D view of the structure with the recognized regions marked. This loops back to the detection of structural members, which can now be also performed on the spatial data covered by the visually marked regions. The result is more robust element detection (by combining visual and spatial detection results), and consequently improved element matching and reconstruction. The resulting object-oriented model is an accurate 3D representation of the structure with the load bearing linear members detected. This model is provided to the modeler, who can then use it to complete the model making process. Leading researchers: Abbas Rashidi and Guangcong Zhang

 

SBIR: A NOVEL VIDEO BASED SOFTWARE APPLICATION FOR AUTOMATIC ACCURATE ROOF SURVEYING January 2013 – June 2013

NSF Grant #1248784, PI: Metalforming Inc., $150,000

This project investigated the technical and commercial feasibility of designing a video-based roof surveying software. Roof surveying is an essential activity in sheet metal roofing projects. Several technologies have been evolved over the years for this purpose; however, none of them are safe, inexpensive, automatic, and accurate enough at the same time. Tape measurement is therefore still the standard practice despite its apparent limitations. This project addressed this need by automatically generating an accurate 3Dwirediagram of a roof which includes necessary measurements. It is the world’s first video-based roof surveying software that fulfills all industry requirements (accuracy, simplicity, cost, safety, and efficiency). Compared to tape measuring, this project significantly reduces measuring costs and also eliminates the exposure of employees to fall hazards, thereby decreasing the very high number of occupational injuries and fall deaths which occur in the roofing industry (7% of private construction fatalities in 2009). The simplicity of the application removes the need for trained surveyors which are required for surveying with a total station. Leading researcher: Habib Fathi.

 

VIDEOGRAMMETRIC BUILDING WIREFRAME CALCULATOR FOR SHEET METAL ROOFING January 2011 – December 2012

PI: Ioannis Brilakis, $123,825

Metalforming Inc. is the North American leader in metal building and architectural sheet metal technology. The company sells and services a CNC machine, called CINCO, which is able to automatically cut roll sheet metal into different pieces appropriate to cover a roof. The only input data to this machine is a 3D wirediagram which includes the perimeter of the roof in 3D space. Currently, this information is manually collected using a tape measure while total station surveying is also used in some projects. Our collaborative research aimed to completely automate this data collection process using a videogrammetric surveying technology. This way, the entire process of cutting roll metal sheets, from data collection to end product, was automated. Leading researcher: Habib Fathi.

 

VIDEOGRAMMETRIC ROOF SURVEYING SYSTEM FOR DIGITAL FABRICATION OF SHEET METAL ROOF PANELS

I-CORPS Grant #1217201, PI: Ioannis Brilakis, $50,000

A structured hypothesis/validation approach was being investigated in this project to develop a disposition plan for a videogrammetric surveying technology. The primary focus of this effort was on the possible use of this technology for as-built 3D documentation of construction sites; however, other potential markets such as on-site measurement of buildings, 3D visualization, augmented reality, etc. were also investigated. Leading researcher: Habib Fathi.

 

CAREER: VISUAL PATTERN RECOGNITION MODELS FOR REMOTE SENSING OF CIVIL INFRASTRUCTURE August 2010 – August 2013

NSF Grant #0948415, PI: Ioannis Brilakis, $402,729

This was a project focused on fundamental research that enabled automated, model based recognition of construction objects. It entailed the creation of visual pattern recognition models for a variety of construction object types. The purpose was to assist as-built modelers of facilities by automatically recognizing the common and more frequent objects automatically, leaving only the specialty items to the hands of the modeler. The validation of this project was based on the software platform of Bentley Inc. , who was the industrial collaborator to this project. Results from this work can be found here. Leading researcher: Stefania Radopoulou.