首页> 外国专利> METHOD FOR CREATING DYNAMIC MODELING OF DRONE BASED ON ARTIFICIAL INTELLIGENCE

METHOD FOR CREATING DYNAMIC MODELING OF DRONE BASED ON ARTIFICIAL INTELLIGENCE

机译:基于人工智能的无人机动态建模方法

摘要

The present invention relates to a method for generating dynamic modeling of an artificial intelligence-based drone, and more particularly, to a method for generating dynamic modeling of a drone expressing the movement of a drone, comprising: (1) generating a virtual dynamic modeling mathematically; (2) obtaining input data and output data from the virtual kinetic modeling generated in step (1); And (3) generating dynamic modeling by estimating the virtual dynamic modeling based on artificial intelligence using the input data and the output data obtained in step (2). According to the AI-based drone dynamic modeling generation method proposed in the present invention, a virtual dynamic model for learning and inference is generated in a situation where it is difficult to acquire data of a drone at the time of generating a dynamic modeling of a drone, and a virtual Virtual I / O data can be acquired from dynamic models to generate and verify more accurate drone dynamics modeling through AI-based model training. In addition, according to the AI-based drone dynamic modeling generation method proposed in the present invention, input data and output data are obtained from a virtual dynamic model so that the data need not be acquired in real time from the drone. You can create and verify dynamic modeling of the drone in its original state without mounting or modifying a separate sensor for acquisition.
机译:本发明涉及一种基于人工智能的无人机动态建模的生成方法,尤其涉及一种表达无人机运动的无人机动态建模的方法,包括:(1)生成虚拟动态建模。数学上(2)从步骤(1)中生成的虚拟动力学模型中获得输入数据和输出数据; (3)通过使用步骤(2)中获得的输入数据和输出数据,基于人工智能估计虚拟动态建模来生成动态建模。根据本发明提出的基于AI的无人机动态建模生成方法,在生成无人机的动态建模时难以获取无人机数据的情况下,生成用于学习和推理的虚拟动态模型。无人机和虚拟Virtual I / O数据可以从动态模型中获取,以通过基于AI的模型训练来生成和验证更准确的无人机动力学模型。另外,根据本发明提出的基于AI的无人机动态建模生成方法,从虚拟动态模型获得输入数据和输出数据,从而不需要从无人机实时获取数据。您可以创建和验证原始状态下的无人机动态建模,而无需安装或修改用于采集的单独传感器。

著录项

  • 公开/公告号KR102021384B1

    专利类型

  • 公开/公告日2019-09-20

    原文格式PDF

  • 申请/专利权人 주식회사 셈웨어;

    申请/专利号KR20170151450

  • 发明设计人 김광진;

    申请日2017-11-14

  • 分类号G06F17/50;B64C39/02;G06N99;

  • 国家 KR

  • 入库时间 2022-08-21 11:47:49

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号