Abstract
This paper proposes the object localization and depth estimation to select and set goals for robots via machine vision. An algorithm based on a deep region-based convolution neural network (R-CNN) will recognize targets and non-targets. After the targets are recognized, we employed both the k-nearest neighbors (kNN) and the fuzzy inference system (FIS) to localize two-dimension (2D) positions. Moreover, based on the field of view (FoV) and a disparity map, the depth is estimated by a mono camera mounted on the end-effector with an eye-in-hand manipulator structure. Although using a single mono camera, the system can easily find the camera baseline by only shifting the end-effector a few millimeters towards the x-axis. Thus, we can obtain and identify the depth of the layered environment in 3D points, which form a dataset to recognize the junction box covers on the table. Experimental tests confirmed that the algorithm could accurately distinguish junction box covers or non-targets and could estimate whether the targets are within the depth for grasping by three-finger grippers. Furthermore, the proposed optimized depth error of -0.0005%, and localization method could precisely position the junction box cover with recognizing and picking error rates 0.993 and 98.529% respectively.
Highlights
Nowadays, no doubt that Artificial Intelligence (AI) has been a fundamental portion of industrial robots
Our challenge lies in using a mono camera with basic field of view (FoV) and disparity point clustering, so we need to blend several approaches such as k-nearest neighbors (kNN), fuzzy inference system (FIS), and the disparity map
We employ deep learning to recognize the targeted junction box covers, which depend on the layered environment, and we propose a localization method based on region-based convolution neural network (R-convolutional neural network (CNN))
Summary
No doubt that Artificial Intelligence (AI) has been a fundamental portion of industrial robots. Muslikhin et al.: Object Localization and Depth Estimation for Eye-in-Hand Manipulator Using Mono Camera its applications in pick-and-place robots have been limited published. Many of these approaches centered on image processing and were not appropriate for a particular industrial robot system. In our previous work [34], object localization and depth estimation have adopted machine learning systems based on eye-to-hand using a stereo camera with a color thresholding method. Our challenge lies in using a mono camera with basic FoV and disparity point clustering, so we need to blend several approaches such as kNN, FIS, and the disparity map These works have focused on deep learning, which is applied to industrial robot picking, as previously mentioned.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have