基于自主导航和感知抓取的智能照护机器人系统研究
Design of an Intelligent Care Robot System Based on Autonomous Navigation and Perceptual Grasping
雷浩 1杨涵 2王金武3
作者信息
- 1. 上海徐汇卓越医工交叉与医学人工智能技术创新中心,上海 200233
- 2. 上海交通大学生物医学工程学院,上海 200030
- 3. 上海交通大学生物医学工程学院,上海 200030;上海交通大学医学院附属第九人民医院骨科,上海 200011
- 折叠
摘要
该文设计了一款带有机械臂,可以自主导航、感知抓取的智能照护机器人,以满足老年人对照护的多样化和复杂性需求.首先,应用Cartographer算法实现照护机器人在养老场景下的同时定位与地图构建(SLAM)功能.其次,利用构建好的SLAM地图,将分段混合搜索(SHS)等法用于自主导航.此外,利用Y-GGCNN网络进行精确的物体姿态分析和抓取.因此,照护机器人能够在复杂的社区照护任务(如搀扶老人、端茶倒水、开门、取快递、盖被子等)中展现出卓越的自适应能力和优秀的操作性能.实验表明,基于自主导航和感知抓取的智能照护机器人系统在真实世界环境中的操作精度与响应速度能得到有效提高.该系统为老龄化社会的社区养老提供了一个高效和智能的照护解决方案,解决了人力护工短缺的问题,适用于医院、养老院、家庭等场所.
Abstract
This article designs an intelligent care robot with a robotic arm that can autonomously navigate and perceive grasping to meet the diverse and complex needs of the elderly for care.Firstly,the simultaneous localization and map building(SLAM)function is implemented using the Cartographer algorithm.Secondly,the SHS path planning method is implemented for autonomous navigation.Furthermore,the Y-GGCNN network is used for precise object pose analysis and grasping.Those above enable care robots to demonstrate excellent adaptive ability and efficient operational performance in handling complex community care tasks,such as assisting the elderly,serving tea and water,opening doors,picking up parcels,and covering blankets.Experiments have shown that autonomous navigation and perceptual grasping can effectively improve the operational accuracy and response speed of care robots in real-world environments.This system provides an efficient and intelligent care solution for community elderly care in an aging society,solving the shortage of human caregivers.It is suitable for hospitals,nursing homes,homes and other places.
关键词
社区照护机器人/自主导航/感知视觉/移动App端Key words
Community Care Robots/Autonomous Navigation/Perceived Vision/Mobile App引用本文复制引用
出版年
2024
NETL
NSTL
万方数据