RoBétArmé is a project that aims to revolutionise Construction 4.0 by automating time-consuming tasks related to shotcrete application. The goal is to enhance safety, productivity, and sustainability in the construction industry. The project will introduce a collaborative construction system that combines human and robot efforts to digitalize and automate shotcrete applications using advanced perception, cognition, mobility, and additive manufacturing skills. With this system, laborious tasks will be automated, reducing the risk of accidents and injuries, while simultaneously increasing productivity and sustainability in the construction industry. 

In this presentation, we will discuss the practical implementation of use cases in the RobétArmé system. Each use case represents a user-centric scenario that showcases interactions between the robotic system, user, and environment. The talk will cover high-level use cases that reflect the RobétArmé system's vision, as well as sub-use cases that specify goal-oriented interactions in various project phases. We will explore these use cases in detail to gain a better understanding of how the system operates and how it benefits users. 

This presentation gives an overview of the digital tools for understanding construction site shotcreting and modelling. The tools are governed by digital twin with the aim of providing a platform for the simulation of mobile robot platform and shotcrete process simulation alongside visualization and construction process diagnostics dashboard. The digital twin is fuelled through IoT sensors and 3D information models resulting in a decision support system and AR-enabled human-machine interface.  

The main focus of this presentation is showcasing the perception and cognition skills of the RoBétArmé robots. A range of techniques, including AI-based scanning, semantic understanding, and 3D reconstruction, have been examined, and tailored for construction requirements. Additionally, developments include volumetric strategies and slicing for construction 3D printing. The focus extends further to the creation of an advanced vision system utilizing multimodal sensor fusion, ensuring real-time monitoring during shotcrete and surface finishing applications for quality assurance. A pivotal aspect involves high-level cognitive planning of robotic tasks. To validate and optimise these technologies, distinct data acquisition campaigns have been conducted at end-users' premises and lab environments, specifically testing various sensors in challenging shotcrete conditions. 

Currently, only a few highly automated solutions have been applied in the construction domain and such solutions have only scratched the surface of shotcrete automation. This task is to deliver a robotic-enabled solution, which is robust and versatile enough, to perform autonomous shotcrete operations (under human supervision), from spraying to surface scraping, for diverse construction environments e.g., tunnels, buildings, etc. In this presentation, you will hear about how fundamental construction skills are learned from human workers, and then I explain how this knowledge is transferred to robots. I will further talk about control and planning strategies for safe and reactive robotic operation during construction tasks. 

During this presentation, we will discuss the RoBétArmé construction robots that we are developing. These robots feature heavy-duty mobile platforms, arm manipulators, and extrusion nozzles for metal additive manufacturing and shotcrete applications. They automate on-site construction and repair of reinforced concrete structures.