GAMMA: Graspability-Aware Mobile MAnipulation Policy Learning based on Online Grasping Pose Fusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Jiazhao, Gireesh, Nandiraju, Wang, Jilong, Fang, Xiaomeng, Xu, Chaoyi, Chen, Weiguang, Dai, Liu, Wang, He
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911787566235648
author Zhang, Jiazhao
Gireesh, Nandiraju
Wang, Jilong
Fang, Xiaomeng
Xu, Chaoyi
Chen, Weiguang
Dai, Liu
Wang, He
author_facet Zhang, Jiazhao
Gireesh, Nandiraju
Wang, Jilong
Fang, Xiaomeng
Xu, Chaoyi
Chen, Weiguang
Dai, Liu
Wang, He
contents Mobile manipulation constitutes a fundamental task for robotic assistants and garners significant attention within the robotics community. A critical challenge inherent in mobile manipulation is the effective observation of the target while approaching it for grasping. In this work, we propose a graspability-aware mobile manipulation approach powered by an online grasping pose fusion framework that enables a temporally consistent grasping observation. Specifically, the predicted grasping poses are online organized to eliminate the redundant, outlier grasping poses, which can be encoded as a grasping pose observation state for reinforcement learning. Moreover, on-the-fly fusing the grasping poses enables a direct assessment of graspability, encompassing both the quantity and quality of grasping poses.
format Preprint
id arxiv_https___arxiv_org_abs_2309_15459
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle GAMMA: Graspability-Aware Mobile MAnipulation Policy Learning based on Online Grasping Pose Fusion
Zhang, Jiazhao
Gireesh, Nandiraju
Wang, Jilong
Fang, Xiaomeng
Xu, Chaoyi
Chen, Weiguang
Dai, Liu
Wang, He
Robotics
Computer Vision and Pattern Recognition
Mobile manipulation constitutes a fundamental task for robotic assistants and garners significant attention within the robotics community. A critical challenge inherent in mobile manipulation is the effective observation of the target while approaching it for grasping. In this work, we propose a graspability-aware mobile manipulation approach powered by an online grasping pose fusion framework that enables a temporally consistent grasping observation. Specifically, the predicted grasping poses are online organized to eliminate the redundant, outlier grasping poses, which can be encoded as a grasping pose observation state for reinforcement learning. Moreover, on-the-fly fusing the grasping poses enables a direct assessment of graspability, encompassing both the quantity and quality of grasping poses.
title GAMMA: Graspability-Aware Mobile MAnipulation Policy Learning based on Online Grasping Pose Fusion
topic Robotics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2309.15459