当前位置: X-MOL 学术Comput. Commun. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Precise localization for achieving next-generation autonomous navigation: State-of-the-art, taxonomy and future prospects
Computer Communications ( IF 6 ) Pub Date : 2020-06-16 , DOI: 10.1016/j.comcom.2020.06.007
Rathin Chandra Shit

Achieving full autonomy in navigation is a complicated problem. The most widely used solution takes up the modular framework for sensing and information processing such as perception, mapping, control, planning and decision making. However, this approach misses the capability of environmental understanding. Hence, to achieve full autonomy in navigation a computing model with self-learning capability inspired by biological intelligence such as memorizing, inferring and experience update is essential for dynamic and noisy environments. Recent advanced sensing, communication and hardware miniaturization technologies achieved few autonomous operations in commercial systems but the full autonomy has not been attained yet. In this paper, the effect of precise and accurate localization for autonomous navigation technologies is extensively studied and the problems and limitations of the related algorithms are analyzed. The major limitations for precise localization are computational complexity, sensor noise and communication delays. These limitations further reduce perception and planning capabilities of autonomous navigation systems. From this study, the future prospects are outlined to achieve a higher level of autonomy by precise localization.



中文翻译:

实现下一代自主导航的精确定位:最新技术,分类学和未来前景

实现导航的完全自治是一个复杂的问题。使用最广泛的解决方案采用了用于感测和信息处理的模块化框架,例如感知,映射,控制,规划和决策。但是,这种方法缺少对环境的理解能力。因此,要在导航中实现完全自主,具有动态学习能力的计算模型将受到动态智能(例如记忆,推断和经验更新)的启发,对于动态和嘈杂的环境至关重要。最近的先进传感,通信和硬件小型化技术在商业系统中几乎没有实现自主操作,但尚未实现完全自主。在本文中,深入研究了精确定位对自主导航技术的影响,分析了相关算法存在的问题和局限性。精确定位的主要限制是计算复杂度,传感器噪声和通信延迟。这些限制进一步降低了自主导航系统的感知能力和计划能力。这项研究概述了通过精确定位实现更高水平的自治权的未来前景。

更新日期:2020-06-24
down
wechat
bug