METHODS OF TASK AND MOTION PLANNING FOR ROBOTS: APPLICATIONS AND LIMITATIONS

Author:

Aygun Guseynova Aygun Guseynova,Javad Aliyev Javad Aliyev

Abstract

Robots are increasingly required to perform more complex tasks, which in turn demand advanced planning algorithms. Task and Motion Planning (TAMP) methods, studied for decades, have made significant progress but still face various challenges. This document provides an overview of TAMP's development, encompassing problem-solving, simulation environments, methods, and remaining limitations. It particularly compares different simulation environments and methods used in various tasks, offering a practical guide and overview for beginners. Task planning is typically seen as planning in discrete spaces, while motion planning deals with continuous spaces. Significant progress has been made in integrating discrete and continuous planning methods to address TAMP problems. A recent survey has focused on TAMP integration, summarizing various methods for solving multimodal motion planning and TAMP problems. It introduces general concepts but primarily focuses on methods that operate in fully observable environments, which are far from real-world applications. Additionally, it demonstrates TAMP problem-solving in a theoretical manner that may not be user-friendly for beginners looking to apply these methods in practice. Therefore, this article aims to provide a practical and broader overview to readers, facilitating an easy entry into the field of TAMP for solving various tasks. Keywords: task and motion planning, simulation environment, learning methods, TAMP.

Publisher

Education Support and Investment Fund NGO

Subject

Immunology

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3