Jihong Zhu’s research while affiliated with New York University and other places

What is this page?

This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (1)

The robot, in four steps performs the bagging task. In the first step, the robot unfolds the bag. In the second step, the bag is opened by the robot. The robot places the red cube in the bag's opening in the third step. In the fourth step, the robot carries the bag completing the task.
This figure illustrates the five states that comprise the bagging task, where the red and blue dots represent the grasping points the robot can select. In (a), the bag is folded such that its area is small, and the opening is not visible. In (b), the bag is unfolded, and the opening is visible. In (c), the bag's opening area is large enough to put an object inside. In (d), the object is in the bag's opening, distinguishing this state from the others. In (e), the task succeeded because no visible objects were left on the table, meaning that the robot carried both the bag and the object. Lastly, (f) shows a failure case when the robot took the bag, but the red cube was still on the table.
The left side of the figure illustrates the experimental setup comprising an object to be bagged (red cube), a Kuka® iiwa 1400TM robot, and an Intel® RealSenseTM. On the left side are the nine bags used during the experiments.
Proposed framework for learning to bag using reinforcement learning.
Implementation of the real‐world learning robot‐bagging framework.

+2

Learning to bag with a simulation‐free reinforcement learning framework for robots
  • Article
  • Full-text available

April 2024

·

62 Reads

·

1 Citation

·

Jihong Zhu

·

·

Bagging is an essential skill that humans perform in their daily activities. However, deformable objects, such as bags, are complex for robots to manipulate. A learning‐based framework that enables robots to learn bagging is presented. The novelty of this framework is its ability to learn and perform bagging without relying on simulations. The learning process is accomplished through a reinforcement learning (RL) algorithm introduced and designed to find the best grasping points of the bag based on a set of compact state representations. The framework utilises a set of primitive actions and represents the task in five states. In our experiments, the framework reached 60% and 80% success rates after around 3 h of training in the real world when starting the bagging task from folded and unfolded states, respectively. Finally, the authors test the trained RL model with eight more bags of different sizes to evaluate its generalisability.

Download