Automation in Construction in brief – 24/05/2025
Automation and Digitization
Researchers developed a constraint programming-based relax-and-solve algorithm (CP-R&S) to cut costs in repetitive construction project scheduling. By modeling crew allocation strategies through mixed-integer linear programming and constraint programming, the method outperformed traditional models in case studies. CP-R&S offers construction managers a flexible, efficient tool to optimize schedules and resource use across complex, multi-crew projects.
Source: https://doi.org/10.1016/j.autcon.2025.106272
A Chilean research team developed an augmented reality tool with low-cost ultrasonic sensors to automate rebar inspection on construction sites. Integrated with BIM, the system improves accuracy, reduces human error, and enhances workflow efficiency. Tested in real conditions, the tool exemplifies Construction 4.0’s promise, linking digital models with physical structures. Future upgrades may include LiDAR and AI for predictive inspections.
Source: https://doi.org/10.1016/j.autcon.2025.106288
Hasani et al. propose an AI-powered system for structural health monitoring that detects damage in infrastructure without human input. By combining machine learning with signal processing and probabilistic models, the framework identifies structural changes even under varying environmental and operational conditions. Tested on a lab bridge, it pinpointed stiffness losses with high accuracy—promising safer, smarter, and more autonomous infrastructure maintenance.
Source: https://doi.org/10.1016/j.autcon.2025.106222
Xu et al. developed multi-GeoLLM, a framework using multiple AI agents to automate complex geotechnical design. It combines text and image inputs, performs calculations, generates drawings, and includes a human-in-the-loop for risk review. Tested on 160 cases, it achieved near-perfect accuracy, promising to cut repetitive work and bring intelligent automation to tasks like footing and slope design.
Source: https://doi.org/10.1016/j.autcon.2025.106257
A new multi-domain learning method offers a breakthrough in computer vision for construction, balancing generalization across domains with domain-specific accuracy. The model, trained on significantly less data, outperformed traditional approaches in robustness and versatility. It proved effective across varied construction environments, paving the way for more efficient, universal AI models in the sector despite data scarcity.
Construction Robotics
Colombia’s Robot.com, formerly Kiwibot, has quietly become a global player in service robotics. With over 500 robots designed in Medellín, it operates across 30 U.S. states, Dubai, and Saudi Arabia, serving clients like Amazon Web Services and Careem. Despite limited local investment, the firm exemplifies Latin America’s emerging tech talent, leveraging skilled engineers to compete in a $40 billion global robotics market.
Beyond Imagination, a robotics startup co-founded by AI futurist Ray Kurzweil, secured $100 million in Series B funding from Gauntlet Ventures, valuing the firm at $500 million. The company’s humanoid robot, Beyond Bot, is designed for industrial use in sectors like pharmaceuticals and semiconductors. It is also developing Aura, an operating system to integrate humans, robots, and legacy machinery in manufacturing.
Source: https://www.fastcompany.com/91337758/beyond-imagination-raises-100-million-humanoid-robots
South Korea is deploying robot chefs in highway restaurants to combat labor shortages amid an aging population. These machines can prepare up to 150 meals per hour, boosting efficiency but sparking worker unease and customer complaints over food quality and limited menus. While automation addresses staffing gaps, it struggles to match the skill and nuance of human cooks, highlighting tensions between technology and tradition in the kitchen.
The US-China trade war is stalling the development of humanoid robots by disrupting supply chains. The US’s 145% tariff on Chinese imports and tightened export controls on AI chips restrict China’s access to crucial technology. As China produces 63% of the world’s robot hardware, these measures threaten delays in mass production, potentially hindering advancements in factory automation, elder care, and domestic robotics.
Source: https://restofworld.org/2025/china-us-humanoid-robot-trade-war-2/
Amazon’s claim that robots don’t eliminate jobs contrasts with data revealing a 25% drop in Fulfillment Center employment from 2022 to 2024, despite increased productivity. While automation has introduced over 700 new job categories, it has also led to significant labor reductions in highly automated centers. This trend underscores the complex impact of automation on the workforce.
Source: https://www.theguardian.com/commentisfree/2025/may/08/amazon-jobs-robotics
Amazon has introduced Vulcan, a robot equipped with tactile sensing capabilities, marking a significant advancement in warehouse automation. Unlike previous models relying on suction and vision, Vulcan utilizes AI to identify and manipulate objects by touch, enabling it to handle approximately 75% of items in Amazon’s warehouses. This development aims to reduce physical strain on human workers and improve efficiency in sorting and stowing items. While Amazon emphasizes that robots will augment human roles rather than replace them, the introduction of Vulcan raises concerns about potential job displacement due to increased automation.
Space Construction
NASA is advancing 3D printing technologies to enable construction on the Moon and Mars using local materials like regolith. Through projects such as MMPACT and partners like ICON and Dr. Behrokh Khoshnevis, techniques like Contour Crafting and laser sintering are being developed for habitats and infrastructure. These innovations promise sustainable off-world living—and are already influencing construction on Earth.
NASA has released Ogma, an open-source tool designed to generate runtime monitoring applications for flight and robotics systems. Written in Haskell, Ogma translates high-level specifications into hard real-time C99 code, facilitating integration with platforms like NASA’s Core Flight System and the Robot Operating System (ROS). By automating the creation of runtime monitors, Ogma aims to enhance the reliability and safety of autonomous aerospace operations .
Source: https://github.com/nasa/ogma
NASA’s VIPER rover, a $450 million project designed to explore lunar water ice, remains grounded after the agency canceled its mission in 2024 due to delays and cost overruns. Attempts to partner with private firms for a lunar launch have faltered, leaving the fully assembled rover in storage. NASA has now abandoned its solicitation for commercial proposals and is seeking alternative methods to deliver VIPER to the Moon.
Source: https://gizmodo.com/nasa-has-a-badass-moon-rover-but-no-way-to-get-it-to-the-moon-2000600189
Aubrie Henspeter leads procurement for NASA’s Commercial Lunar Payload Services (CLPS), a key initiative under Artemis to deliver science payloads to the Moon via private firms. With two decades at NASA, she blends strategic acumen and adaptability, guiding a lean team through seven missions. Her work streamlines lunar deliveries, boosts industry collaboration, and embodies the agency’s push for faster, cost-effective space exploration.