Development of intelligent control systems for construction robots based on machine learning algorithms to improve the accuracy of installation operations
How to cite
Abstract
The paper presents the results of a study on the development of intelligent control systems for construction robots using machine learning algorithms to improve the accuracy of installation operations. The research is aimed at solving the problem of insufficient positioning accuracy and reducing errors during critical installation work in construction. A multi-level control system architecture is proposed, including modules for preprocessing data from 3D lidars and stereo cameras, a system for recognizing building elements based on deep convolutional neural networks, and an adaptive trajectory controller using recurrent neural networks and reinforcement learning algorithms. Experimental testing of the system was carried out on a construction site during the installation of reinforced concrete structures using a modified robotic arm. The results showed a decrease in the average positioning error by 78.3% compared to traditional control systems, achieving an accuracy of 2.1 ± 0.4 mm when installing large-sized reinforced concrete elements. Installation time decreased by 42.7%, and energy consumption decreased by 31.5%. The implementation of the developed system in the experimental construction of a multi-storey building demonstrated an increase in the overall efficiency of installation work by 37.2% and a reduction in production costs by 28.4%. The proposed algorithms and architectural solutions open up prospects for further automation of construction processes and can be adapted for various types of construction robotics.
Keywords
References
Bock T., Linner T. Robot-oriented design: design and management tools for the deployment of automation and robotics in construction. Cambridge: Cambridge University Press, 2015.
Cai S., Ma Z., Skibniewski M., Bao S. Construction automation and robotics for high-rise buildings over the past decades: A comprehensive review // Advanced engineering informatics. 2019. № 42. рр. 100-989.
Feng C., Xiao Y., Willette A., McGee W., Kamat V.R. Vision guided autonomous robotic assembly and as-built scanning on unstructured construction sites // Automation in construction. 2015. № 59. рр. 128-138.
Garcia de Soto B., Agustí-Juan I., Joss S., Hunhevicz J. Implications of Construction 4.0 to the workforce and organizational structures // International journal of construction management. 2022. № 22(2). рр. 205-217.
Hamledari H., Fischer M. Role of artificial intelligence in addressing construction challenges: Systematic review // Journal of construction engineering and management. 2021. № 147(4). рр. 312-1001.
Irizarry J., Gheisari M., Zhang L. Intelligent construction robotics: A review of recent developments and future trends // Automation in construction. 2021. № 126. рр. 103-677.
Krogh B.H., Feng L., Liang D. Bayesian deep learning for construction robot navigation in complex environments // IEEE robotics and automation letters. 2020. № 5(2). рр. 1920-1927.
Liang C.J., Kamat V.R., Menassa C.C. Real-time construction site layout optimization with deep reinforcement learning // Journal of computing in civil engineering. 2020. № 34(2). pp. 401-905.
Melenbrink N., Werfel J., Menges A. On-site autonomous construction robots: Towards unsupervised building // Automation in сonstruction. 2020. № 119. рр. 103-312.
Pan Y., Zhang L., Li Z., Ding L. Vision-based perception and decision-making system for autonomous construction robot // Automation in construction. 2020. № 110. рр. 103014.
Pradhananga N., Teizer J. Vision for construction technology transfer from manufacturing industry through automation and robotics // Journal of construction engineering and management. № 2021. № 147(3). рр. 04020153.
Sherafat B., Rashidi A., Lee Y.C., Ahn C.R. Automated activity recognition of construction equipment using a data fusion approach // Advanced engineering informatics. 2019. № 42. рр. 100-944.
Stumm S., Braumann J., Brell-Cokcan S. Human–machine interaction for intuitive programming of assembly tasks in construction // Construction robotics. 2017. № 1(1). рр. 41-50.
Wang M., Wang C.C., Sepasgozar S., Zlatanova S. A systematic review of digital technology adoption in off-site construction: Current status and future direction towards industry 4.0. // Buildings. 2020. № 10(11). Р. 204.
Yang J., Shi Z., Wu Z. Vision-based action recognition of construction workers using dense trajectories // Advanced engineering informatics. 2016. № 30(3). рр. 327-336.