Domain Independent Unsupervised Learning to grasp the Novel Objects
This work addresses the problem of efficient and accurate grasping for novel objects in robotics, offering an incremental improvement over prior learning-based methods by reducing computational demands.
The paper tackles the challenge of selecting feasible grasp regions for novel objects in vision-based grasping by proposing an unsupervised learning algorithm using k-means clustering and a Grasp Decide Index, achieving robust and adaptive performance validated through experiments on standard datasets like Amazon Robotics Challenge 2017.
One of the main challenges in the vision-based grasping is the selection of feasible grasp regions while interacting with novel objects. Recent approaches exploit the power of the convolutional neural network (CNN) to achieve accurate grasping at the cost of high computational power and time. In this paper, we present a novel unsupervised learning based algorithm for the selection of feasible grasp regions. Unsupervised learning infers the pattern in data-set without any external labels. We apply k-means clustering on the image plane to identify the grasp regions, followed by an axis assignment method. We define a novel concept of Grasp Decide Index (GDI) to select the best grasp pose in image plane. We have conducted several experiments in clutter or isolated environment on standard objects of Amazon Robotics Challenge 2017 and Amazon Picking Challenge 2016. We compare the results with prior learning based approaches to validate the robustness and adaptive nature of our algorithm for a variety of novel objects in different domains.