Automation in Construction in brief – 25/04/2026
Automation and Digitization
Scientists have unveiled a soil-powered fuel cell that generates electricity using microbes, offering a battery-free energy source for underground sensors. The device operates reliably in both dry and wet conditions and outlasts comparable technologies. The innovation could enable sustainable, low-maintenance monitoring of soil moisture and environmental conditions, with promising applications in underground infrastructures.
Source: https://www.sciencedaily.com/releases/2026/04/260419054821.htm
OpenAI unveiled GPT-5.5, its most advanced AI model to date, touting improvements in speed, intuition, and versatility across enterprise and scientific applications. The release marks a step toward OpenAI’s ambition of creating a unified “super app” combining multiple AI services.
Source: https://techcrunch.com/2026/04/23/openai-chatgpt-gpt-5-5-ai-model-superapp/
Researchers released CUBIT-InSeg, a high-resolution UAV-acquired dataset comprising 6,996 images and 62,187 annotated façade defects (cracks and spalling) for high-rise building inspection. Over 80 segmentation models were benchmarked, with cross-domain transferability demonstrated. The dataset enables automated, quantitative defect assessment, integrating instance segmentation with digital twin workflows for severity grading, supporting reproducible, scalable, and standard-aligned façade condition monitoring and maintenance planning.
Source: https://doi.org/10.1016/j.autcon.2026.106980
A systematic review of 281 publications examines localization and navigation (L&N) technologies for construction robots, highlighting the challenges posed by dynamic, unstructured sites. The study classifies L&N approaches, assesses human involvement in robot deployment, and identifies gaps in human-centric design. Despite diverse solutions, no consensus has emerged, underscoring the need for further research to ensure safe and efficient human-robot collaboration on construction sites.
Source: https://doi.org/10.1016/j.autcon.2026.106953
Researchers developed an automated method to generate Building Information Models (BIMs) from inconsistent 2D floor plans and elevations, overcoming challenges of missing dimensions and misaligned drawings. Using deep learning and a large multimodal model for scale extraction, plus the Hungarian algorithm for cross-view matching, the system achieved high accuracy (mAP@0.5 of 0.934 for floor plans) and outperformed traditional approaches, enabling scalable BIM creation for legacy buildings.
Construction Robotics
SusHi Tech Tokyo 2026, running April 27-29, cements Tokyo’s status as a global tech hub, showcasing advances in AI, robotics, climate resilience, and entertainment. The event features live demos, industry leaders from firms such as Nvidia and Nissan, and a city leaders’ summit on urban resilience. Remote participation options expand access.
Source: https://sushitech-startup.metro.tokyo.lg.jp/en/
Researchers demonstrated an integrated system for autonomous on-site assembly of cross-laminated timber panels using a digitally controlled tower crane and distributed robotic modules embedded in the panels. The robots provide local vision-based sensing and relay data to a central coordinator, enabling real-time multi-view pose estimation and precise crane guidance. Field experiments show robust, markerless assembly under realistic conditions, indicating distributed robotic perception can address key bottlenecks in timber construction automation.
Source: https://doi.org/10.1016/j.autcon.2026.106973
Researchers introduced STF-Inspector, a multi-sensor, vision-guided framework for mobile robots to inspect building plenum spaces. By fusing RGB, depth, thermal, and environmental data, and employing advanced segmentation and tracking models, the system detects faults such as structural deformation and water leakage with improved robustness and accuracy. Field tests demonstrated superior performance over traditional methods, reducing reliance on manual inspections and enhancing safety and efficiency.
Source: https://doi.org/10.1016/j.autcon.2026.106889
Researchers have developed an integrated framework for autonomous mobile robots (AMRs) in construction, combining resource scheduling, delivery, and enhanced 6-DoF grasping. Using reinforcement learning and advanced pose estimation, the system optimises task sequencing and item handling, reducing total waiting time by up to 39.2%. The approach addresses site-specific constraints and improves operational efficiency, offering a scalable solution for automating construction logistics.
Source: https://doi.org/10.1016/j.autcon.2026.106968
Space Construction
NASA’s Artemis II mission concluded with a successful splashdown off San Diego after a 694,481-mile lunar journey. Initial assessments show Orion’s heat shield and SLS rocket performed as expected, with minimal char loss and precise landing. Ground systems sustained little damage. Engineers are analysing post-flight data to inform Artemis III, scheduled for 2027, as NASA advances plans for lunar surface missions and future Mars exploration.
Source: https://www.nasa.gov/missions/nasa-on-track-for-future-missions-with-initial-artemis-ii-assessments/
NASA has released its advanced Launch, Ascent, and Vehicle Aerodynamics (LAVA) software to the US aerospace sector. Previously used for critical missions such as Artemis and Mars landers, LAVA offers high-fidelity, GPU-accelerated simulations of airflow, pressure, and structural dynamics. The tool enables faster, more accurate modeling for rockets, aircraft, and spacecraft, supporting innovation across industry and research with improved efficiency and accessibility.
Source: https://www.nasa.gov/aeronautics/nasa-releases-powerful-lava-software-to-us-aerospace-industry/
NASA will supply a Neutron Spectrometer System (NSS) to the Japan-India LUPEX lunar rover, set to launch no earlier than 2028, to map subsurface ice at the Moon’s south pole. The NSS detects hydrogen, indicating water, without drilling. This initiative supports future lunar exploration by identifying accessible water deposits for use in life support and fuel, advancing NASA’s Artemis programme.
Source: https://www.nasa.gov/solar-system/moon/nasas-water-hunting-tool-will-help-scout-moons-south-pole/
German, Norwegian, and Swedish space industry representatives convened at the DLR in Bremen on April 14th for the German-Nordic Space Industry Dialogue. The event featured keynote addresses on sector strategy and international cooperation, as well as company pitches and networking sessions. Participants discussed strengthening cross-border ties and advancing collaboration in the space sector.
Source: https://www.dlr.de/en/irs/news/news/german-nordic-space-industry-dialogue-in-bremen
