Efficient 3D object tracking approach based on convolutional neural network and Monte Carlo algorithms used for a pick and place robot

Abstract

Currently, Deep Learning (DL) shows us powerful capabilities for image processing. But it cannot output the exact photometric process parameters and shows non-interpretable results. Considering such limitations, this paper presents a robot vision system based on Convolutional Neural Networks (CNN) and Monte Carlo algorithms. As an example to discuss about how to apply DL in industry. In the approach, CNN is used for preprocessing and offline tasks. Then the 6-DoF object position are estimated using a particle filter approach. Experiments will show that our approach is efficient and accurate. In future it could show potential solutions for human-machine collaboration systems.

Motion Capture Technology in Industrial Applications: A Systematic Review

“…The increased use of IMUs promoted the development of advanced algorithmic methods (e.g., Kalman filters and machine learning, Table 7 ) for data processing and estimation of parameters that are not directly measured with inertial systems [ 17 ]. Optoelectronic technologies performed better and with higher tracking accuracy in human–robot collaboration tasks [ 33 ] and robot trajectory planning [ 32 , 46 , 52 ], due to the favourable conditions in such applications (e.g., the limited working volume and the known robot configurations) which allowed cameras to avoid obstructions. In general, hybrid systems that incorporate both vision and inertial sensors were found to have improved tracking performance in noisy and highly dynamic industrial environments, compensating for drift issues of inertial sensors and long-term occlusions which can effect camera-based systems [ 15 , 40 ].…”