Data Fusion Research for Collaborative Robots-Manipulators within Industry 5.0

The article examines methods of data fusion (Data Fusion) for collaborative manipulator robots in the context of Industry 5.0. Three approaches are considered: Kalman filter, Bayesian estimation and Dempster-Shafer theory. The Kalman filter has proven to be effective for linear systems, but requires modification for nonlinear problems. Bayesian estimation provides accuracy for complex systems, although it requires more resources. The Dempster-Shafer theory is effective under data uncertainty, but has a high computational complexity. The conclusions indicate the importance of choosing a data fusion method depending on the requirements for accuracy and adaptability of robots in the production conditions of Industry 5.0.