Digital, autonomous, Intelligent and Synchronous system for Continuous identification, Optimization and Value Extraction of Resources from the end-of-use built environment

Project Information

DISCOVER

Grant agreement ID: 101129909

DOI

10.3030/101129909

EC signature date 29 May 2024

Start date 1 June 2024

End date 31 May 2028

Funded under

• Digital, Industry and Space

Total cost

No data

EU contribution

€ 5 997 208,50

Coordinated by

UNIVERSITAT POLITECNICA DE CATALUNYA
Spain

Periodic Reporting for period 1 - DISCOVER (Digital, autonomous, Intelligent and Synchronous system for Continuous identification, Optimization and Value Extraction of Resources from the end-of-use built environment)

Reporting period: 2024-06-01 to 2025-11-30

Summary of the context and overall objectives of the project

The DISCOVER project aims to develop an autonomous, synchronous, continuous and intelligent identification and data analysis system for materials and products in existing end-of-life built works.

As a result of the project, the following impacts are expected to be achieved:
1) Optimization of deconstruction processes by introducing new equipment for field sensing, and open databases.
2) Reduction of C&D waste and energy consumption, increase of reuse and recycling and better handling of hazardous materials, and lowering construction sector emissions.
3) Enhancement of the digital skills of construction sector workers, raising general public awareness on issues related to C&DW.
4) Enhancement of construction workers’ safety by reducing work-related injuries.

To achieve these impacts, the project will develop fully autonomous robotic platforms coupled with continuous, targeted identification tools to scan and test built works, to provide coordinated quantitative and qualitative data from different materials, including complex and concealed elements. Artificial intelligence algorithms will allow a rapid analysis of the properties and characteristics of components, and feed the information to an automated scan-to-BIM model creation. A multi-dimensional BIM, including selective demolition processes, labour productivity, and technical planning will become an on-demand digital twin of the demolition site. A user-friendly digital catalogue with a classification system will serve as a decision support tool and ensure asset traceability from the demolition site to their multi-cycle end of life. It will support the use of the secondary raw materials and will be destined at the agents of the construction sector. New professional development tools will guarantee maximum impact on society through the participation of the main stakeholders, such as on-site workers, contractors, designers, architects, engineers and developers. The tools will cover topics such as circularity in construction, material identification using robots, BIM modelling, sustainable demolition strategies, web3-based material traceability tools, and material quality assessment for reuse.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

During the first reporting period, the DISCOVER project has focused on the development of robotic systems, sensing technologies, artificial intelligence methods, digital data frameworks, and analytical approaches for demolition and reuse.
Below we describe the activities undertaken in the given period and illustrate the main achievements:
A mobile manipulator robot designed to operate in construction and demolition environments was fully engineered, built, and tested. The robot integrates multiple sensing technologies, including RGB cameras, LiDAR, ground-penetrating radar, and multispectral sensors, enabling it to capture detailed information about building elements and materials without physical intervention. Dedicated software was developed to allow the robot to navigate autonomously in complex environments, generate maps, localize itself, and follow planned paths while continuously collecting georeferenced data (WP1).
In parallel, the conceptual design of a complementary robotic system for discrete, localized sampling was initiated, following by the development and validation of a portable, automated system for in-situ identification of material composition. The system allows invasive but controlled measurements directly on site and is supported by dedicated data processing algorithms that translate raw measurements into material composition information. The prototyping and testing of the drilling platform was successfully performed (WP2).
Guidelines for ground-penetrating radar (GPR) were developed to ensure reliable data acquisition in demolition environments. An annotated database was created, containing both real and synthetic radar data representing different construction materials and embedded elements (WP1).
To enable coherent use of the diverse data generated by robots and sensors, a comprehensive data integration and governance framework was defined (WP3).
Beyond data acquisition and modelling, the project also advanced the scientific understanding of demolition and deconstruction practices. A comprehensive review of demolition methods currently used in Europe was conducted, covering technical approaches, operational constraints, waste separation practices, and levels of process control. A state-of-the-art review of life cycle assessment and life cycle costing approaches for demolition activities was completed (WP4).
The project initiated scientific research into the conditions required for the adoption of advanced digital and robotic tools in the construction and demolition sector. Bibliographic research on current deconstruction practices and digital tool uptake was combined with exploratory interviews and on-site observations (WP7).

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

A first major result is the demonstration of autonomous, robot-based pre-demolition inspection as a viable alternative to conventional manual surveys. This system is capable of safely operating in complex demolition environments, and continuously collecting georeferenced surface and subsurface data. The integration of multiple sensing modalities (e.g. RGB, LiDAR, GPR) within a single robotic workflow enables a level of coverage, consistency, and traceability that is not achievable with existing approaches.
A second key advancement lies in the automated interpretation of heterogeneous sensor data for material and element identification. The project applied AI-based methods capable of extracting meaningful material and structural information from these data sources and preparing them for fusion.
The project has also advanced the scientific understanding of demolition and deconstruction processes by systematically analysing current practices and their implications for material recovery and sustainability.
The expected impacts of these results have the potential to significantly improve the way existing buildings are assessed and deconstructed. More accurate material identification and quantification can improve the efficiency and quality of selective demolition, increasing the share of materials suitable for reuse and high-value recycling.
From an environmental perspective, the results pave the way for reducing construction and demolition waste, lowering resource consumption, and supporting circular construction strategies. Economically, improved information and planning can reduce unexpected costs, improve recovery value, and create new opportunities for secondary material markets. At a systemic level, the project contributes to the digital transformation of the construction and demolition sector by demonstrating how robotics, AI, and BIM can be combined into coherent operational workflows rather than remaining isolated innovations.
While the results achieved so far represent clear advances beyond the current state of the art, the following steps are required to enable their large-scale uptake and impact:
- Testing of the created robotic system in real demolition sites to further improvement and validation of its performance, reliability, and cost-effectiveness under operational constraints;
- More testing for the improvement of the robustness, scalability, and generalisation of the AI algorithms.