Towards autonomous robot collectives delivering collaborative tasks in dynamic unstructured construction environments

Scope

Background and scope

Robotic automation offers significant advantages to several sectors, yet on-site construction robotics is amongst the most challenging and least understood fields in robotics. The unstructured, dynamic environment with human presence makes navigation and automation of the many concurrent construction tasks deeply challenging. Further, the current state-of-the-art solutions focus on adding higher degrees of automation to legacy tools, such as heavy equipment designed for diesel engines and human operators.

Radical innovations are essential for the sector to address the unprecedented wave of building growth, especially in the context of labour shortages, a productivity gap between the construction and other industries, rising expectations for occupational health and safety, and the need for healthier and more affordable living environments.

Through the collaboration of multiple agents (both humans and machines) construction processes can be accelerated, enabling more complex processes with multiple tasks to be performed simultaneously and collaboratively. Multi-robotic collaboration, where robotic agents support and complement each other’s tasks and skill sets within the same workspace, may unlock entirely new processes that are not possible using single robotic machines. This approach could involve multiple distributed “swarms” of collaborative robots using distributed control algorithms and robot learning systems, which may be better suited to large, spatially distributed tasks and can adapt to unpredictable environments. Doing so while also supporting the electrification of the (legacy suite of) construction equipment, will help break with the need to go for ever larger machines and facilitate the development of novel technologies that enable efficient accurate and reliable control, and the adoption of collaborative robots that are suitable for commercial on-site construction environments.

Realising the disruptive potential of novel emerging technology paradigms that reconsider construction processes from the fundamentals can help supplant and substitute the legacy suite of tools with novel autonomous collaborative construction robots in an integrated, “designed-for-robotics” digital production and assembly chain.

Such developments could also further enhance an emerging paradigm shift from today’s complex mix of on-site construction tasks, towards a future of off-site fabrication and on-site assembly. Off-site fabrication offers industrial economic advantages of producing modularized building elements at scale in a controlled, digitalized and automated factory environment. For the construction sector this paradigm shift can deliver demand-side emissions reductions, by implementing strategies of digitalized structural efficiency and novel materials, as well as of zeroemission construction sites through electrification.

This Pathfinder Challenge aims to address all construction tasks typically required for site preparation, substructure, and superstructure, as well as the coordination between 43 these tasks to support a transition towards building with autonomous electrified construction equipment. It includes the role of human agents in construction processes, as even high degrees of multi-robotic autonomy with low degrees of supervision will require a collaborative connection between human and robotic agents, ensuring they can safely collaborate and share the same workspace.

Specific objectives

The overall objective of this Challenge is the development of breakthrough technologies in the domain of autonomous collaborative on-site construction robots for an integrated, designed-for-robotics, digital production and assembly chain.

The Challenge is open to the 3 main construction tasks applied to the 2 main construction segments of buildings and infrastructure. Innovative application in adjacent construction segments (for example coastal protection foundations for energy infrastructure) also fall within scope.

Each funded project shall deliver the following 3 specific objectives:

Objective 1:

Development of a simplified structural, load-bearing, material-robot building system to assemble a representative and future-relevant structure (pavilion) using a multitude of discrete modules (elements, segments, blocks, voussoirs). This system must demonstrate TRL4 (validation in laboratory environment) of the autonomous collaborative multirobotic assembly. The structure can represent an infrastructure (for example a bridge, tunnel, culvert, conduit), a building (for example a tower, vault, dome, arch, multi-story skeleton, wall) or other construction elements (for example a foundation, secant wall, barrier, sea wall). The building system can also integrate unprocessed and pre-processed in-situ building materials (rocks, sand, natural materials, demolition materials, disassembled elements). Projects are expected to demonstrate the technologies at least at a relevant human scale in terms of volume, mass and moment of inertia, and ideally at a larger real-world architectural scale, rather than at a laboratory desktop scale.

Solutions are expected to incorporate “design-for-robotic-assembly” aspects, such as the robot-material interfaces, module interfaces and connectors, and may include innovative approaches such as embedded sensing in the modules.

A virtual simulation of the disassembled state, various intermediate assembly stages (including temporary (robotic) support measures if necessary) and final assembled state is expected to be part of the systems development process. The project should include a documented validation of key design decisions (for example materials used 44 or configurations that simulate scaled behaviour) against the minimal requirements of the TRL4 demonstration objectives of the autonomous mobile multi-robotic collaborative platform.

Objective 2:

Development of an autonomous mobile multi-robotic collaborative platform using at least two, preferably more, mutually aware collaborative robotic systems specifically designed for the assembly tasks outlined in Objective 1. This objective requires a structured systems engineering approach to conduct a thorough functional system analysis and to allocate system-level functions between humans and machines within the target autonomous mobile multi-robotic collaborative platform.

The design should include the definition of system states and modes, along with the transitions between them, to ensure safe autonomous operations and effective demonstration of robot-robot and human-robot collaborations and interactions (passive, active, adaptive) at TRL4.

The project should also describe how the proposed technology can be scaled to meet the full dimensions of the intended commercial application in future.

Utilizing existing industrial robots or modifying suitable existing construction tools is allowed. However, these approaches may face workspace limitations when scaled to full commercial dimensions. Conversely, novel relative multi-robotic platforms could make full use of the opportunities of the material-robot system independent of scaling limitations in future.

Objective 3:

Achieve a TRL4 demonstration of an autonomous assembly sequence using the demonstration building system developed in Objective 1, executed by the autonomous mobile multi-robotic collaborative platform developed in Objective 2. The demonstration of a subsequent disassembly sequence is optional but encouraged if the building system is designed for disassembly. The demonstration will take place in a laboratory environment, including tests that explore the system’s resilience and limits under controlled unstructured real-world conditions (for example fault tolerance, granular uneven surfaces, environmental obstacles). These tests aim to identify key weaknesses and recommend future technology developments.

The specific objective of this challenge is to advance the digitalized chain of off-site modular production with on-site autonomous mobile multi-robotic collaborative assembly. Therefore, on-site 3D-printing of cementitious materials or polymers as a primary construction task is outside the scope of this challenge.

Expected Outcomes and Impacts:

This Challenge contributes to the European Green Deal, the European AI Strategy, and the key strategic orientations of Horizon Europe for the digital and green transitions of the construction sector. The anticipated impacts of this Challenge include addressing likely shortages and competition in the labour markets, enhancing productivity and competitiveness within the construction industry, and improving worker safety. It will facilitate a shift towards offsite industrial fabrication coupled with onsite assembly and disassembly, reducing emissions from on-site construction activities, and lowering costs and mitigating risks associated with construction projects. This Challenge will also serve as a lighthouse for industrialization in important policy areas, such as affordable housing, the renovation wave, circular construction, and infrastructure development.

The field of mobile construction robotics, in particular heterogeneous collaborative robots assembling discrete building elements, is challenging and multi-disciplinary. Given the nascent state of the enabling technologies, the cumulative impact of the portfolio of Pathfinder projects is expected to surpass that of individual projects. Consortia will benefit from mutual learning and the exchange of approaches and expertise in areas such as mapping, navigating and building awareness of unstructured environments, force-aware manipulation, swarm collectives, as well as commercialisation pathways.

Furthermore, consortia will be encouraged to collaborate on developing performance metrics and communicate their outputs to the broader public with a view to accelerating the adoption of these radical innovations by the sector. Such valuable joint portfolio activities are anticipated to be discussed and agreed upon by the funded projects.

The portfolio of projects selected will aim to cover a complementary set of projects that span the “application” and “approach” fields specified below and combinations thereof:

- Applications fields (super-structure, sub-structure, site-preparation, building, infrastructure, other construction, target type of environment).

- Approach (type of robot, number of agents, coordination strategy, level of autonomy, strategy for stability during assembly sequence, multi-modal sensors, resilience strategy for environmental variability, type of discrete building elements and fixations, level of integration of material-robot system).

Specific conditions

Applications for this Challenge with elements that concern the evolution of European communication networks (5G, post-5G and other technologies linked to the evolution of European communication networks) will be subject to restriction for the protection of European communication networks (see Annex II – Section B1)