当前位置: X-MOL 学术IEEE Trans. Affect. Comput. › 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.)
Inconsistency-Based Multi-Task Cooperative Learning for Emotion Recognition
IEEE Transactions on Affective Computing ( IF 9.6 ) Pub Date : 8-9-2022 , DOI: 10.1109/taffc.2022.3197414
Yifan Xu 1 , Yuqi Cui 1 , Xue Jiang 1 , Yingjie Yin 2 , Jingting Ding 3 , Liang Li 2 , Dongrui Wu 1
Affiliation  

In this paper, we present a platform that can simulate Manned-UnManned Teaming (MUM-T) situations using various hardware and software being integrated. MUM-T, which combines unmanned aerial vehicles (UAVs) with an existing manned vehicle (MV), is growing into an indispensable strategy thanks to enormous advantages it provides. In order to test this indispensable strategy which involves flight control algorithms being applied for various MUM-T situations, it is certainly needed to run and evaluate three-dimensional (3D) MV-UAVs combined simulations in realistic environments. However, there is no such simulator that enables such 3D MUM-T simulations as well as the Human Interface (HI) for both MV pilot and the mission manager at a ground station. In this paper, an MUM-T integrated simulation platform (MUM-T ISP) is proposed that combines a single full-scale helicopter simulator (for MV simulation) and multiple Iris/PX4-Autopilot (for UAVs simulation), along with specially developed HI devices (TM and TV). Tactical Monitor (TM), one of the HI devices, is a touchable tablet which allows the MV pilot to interact with UAVs by simple actions (e.g. touching a button), and the other HI device called Tactical Viewer (TV) allows the mission manager at a ground station to closely monitor the MUM-T situation via realistic 3D displays of MV and UAVs. The proposed MUM-T ISP is tested on several mission scenarios and then evaluated by actual pilots and the mission manager based on sensible criteria. It is expected that the proposed simulation platform may serve well as a training tool for prospective MV pilots working with UAVs, and also as a research tool for flight engineers developing control algorithms that make MV and UAVs work together.

中文翻译:


基于不一致性的多任务协作学习的情感识别



在本文中,我们提出了一个平台,可以使用集成的各种硬件和软件来模拟有人-无人协作(MUM-T)情况。 MUM-T将无人机(UAV)与现有有人驾驶车辆(MV)相结合,由于其提供的巨大优势,正在发展成为一种不可或缺的战略。为了测试这一不可或缺的策略,其中涉及应用于各种 MUM-T 情况的飞行控制算法,当然需要在现实环境中运行和评估三维 (3D) MV-UAV 组合模拟。然而,还没有这样的模拟器能够为 MV 飞行员和地面站的任务经理提供此类 3D MUM-T 模拟以及人机界面 (HI)。本文提出了一种MUM-T集成仿真平台(MUM-T ISP),该平台结合了单个全尺寸直升机模拟器(用于MV仿真)和多个Iris/PX4-Autopilot(用于无人机仿真),以及专门开发的HI 设备(TM 和电视)。战术监视器 (TM) 是 HI 设备之一,是一种可触摸的平板电脑,允许 MV 飞行员通过简单的操作(例如触摸按钮)与无人机进行交互,而另一个称为战术查看器 (TV) 的 HI 设备允许任务经理在地面站通过 MV 和 UAV 的逼真 3D 显示密切监视 MUM-T 情况。拟议的 MUM-T ISP 在多个任务场景中进行了测试,然后由实际飞行员和任务经理根据合理的标准进行评估。预计所提出的仿真平台可以很好地作为使用无人机工作的未来MV飞行员的培训工具,也可以作为飞行工程师开发使MV和无人机协同工作的控制算法的研究工具。
更新日期:2024-08-26
down
wechat
bug