Hierarchical Policy for Non-prehensile Multi-object Rearrangement with Deep Reinforcement Learning and Monte Carlo Tree Search

1 code implementation • 18 Sep 2021 • Fan Bai, Fei Meng, Jianbang Liu, Jiankun Wang, Max Q. -H. Meng

Non-prehensile multi-object rearrangement is a robotic task of planning feasible paths and transferring multiple objects to their predefined target poses without grasping.

Object

A Large-Scale Dataset for Benchmarking Elevator Button Segmentation and Character Recognition

1 code implementation • 16 Mar 2021 • Jianbang Liu, Yuqi Fang, Delong Zhu, Nachuan Ma, Jin Pan, Max Q. -H. Meng

Human activities are hugely restricted by COVID-19, recently.

Benchmarking Position +1

Reinforcement Learning for Contact-Rich Tasks: Robotic Peg Insertion Strategies

1 code implementation • CUHK Course IERG5350 2020 • Jianbang Liu, Ang ZHANG

In this project, to connect vision and touch, we first equip robots with both visual and tactile sensors and collect a large-scale dataset of corresponding vision and tactile sequences.

reinforcement-learning Reinforcement Learning (RL)

Autonomous Removal of Perspective Distortion of Elevator Button Images based on Corner Detection

no code implementations • 23 Jul 2020 • Nachuan Ma, Jianbang Liu, Delong Zhu

First, we leverage a novel image segmentation model and the Hough Transform method to obtain button segmentation and button corner detection results.

Image Segmentation Segmentation +1

Autonomous Removal of Perspective Distortion for Robotic Elevator Button Recognition

no code implementations • 26 Dec 2019 • Delong Zhu, Jianbang Liu, Nachuan Ma, Zhe Min, Max Q. -H. Meng

To verify the effectiveness of the algorithm, we collect an elevator panel dataset of 50 images captured from different angles of view.