会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 5. 发明公开
    • 인공표식, 인공표식과 자연표식을 혼용한 네비게이션 장치 및 방법
    • 人造地标,使用人造地标和自然地名的导航方法
    • KR1020110049567A
    • 2011-05-12
    • KR1020090106636
    • 2009-11-05
    • 재단법인 포항지능로봇연구소
    • 홍영진이정우서진호최영호방석원강정원
    • G01C15/02G01C21/20G06T7/00
    • G01C15/02G01C21/3623G01C21/3626G06T7/00G08G1/096855
    • PURPOSE: An artificial mark, a navigation device using artificial marks and natural marks, and a method thereof are provided to enable accurate motion of a robot in a wide indoor space using artificial marks and natural marks. CONSTITUTION: A navigation device using artificial marks and natural marks comprises a camera(10), a differential image extracting unit(40), and a position detection unit(60). The camera takes an illumination image and a natural image. The differential image extracting unit calculates the difference image between the illumination image and the natural image. The position detection unit extracts the location information from natural marks extracted from the natural image. The position detection unit corrects the location information from the artificial marks extracted from the difference image. The artificial mark comprises a central body, a direction indicator, and a code part.
    • 目的:提供人造标记,使用人造标记和自然标记的导航装置及其方法,以使机器人能够使用人造标记和自然标记在宽的室内空间中进行准确的运动。 构成:使用人造标记和自然标记的导航装置包括相机(10),差分图像提取单元(40)和位置检测单元(60)。 相机拍摄照明图像和自然图像。 差分图像提取单元计算照明图像和自然图像之间的差分图像。 位置检测单元从从自然图像提取的自然标记中提取位置信息。 位置检测单元从从差分图像提取的人造标记中校正位置信息。 人造标记包括中心体,方向指示器和代码部分。
    • 6. 发明公开
    • 음성 기반 자연어 대화 장치 및 그 방법
    • 自然语言对话方式及其方法
    • KR1020100118744A
    • 2010-11-08
    • KR1020090037594
    • 2009-04-29
    • 재단법인 포항지능로봇연구소
    • 이청재정상근김경덕이성진이진식이근배오승섭최영호윤성용신규식염영일
    • G10L15/20G10L15/22
    • PURPOSE: A voice based natural language conversational device and a method thereof for reducing voice recognition error between a user and an automatic guide artificial intelligence agent are provided to estimate an agent by using voice recognition temporary installation of the high position. CONSTITUTION: An input unit inputs the user firing of the voice. A voice recognition unit(110) inputs the starting point of user voice from the input unit according to the recognition reliability score to the descending order. The voice recognition circuit is created the voice recognition temporary installation of the middle high position N. The sentence structure of N corresponding to the voice recognition temporary installation of an input understanding unit(120) becomes in the voice recognition circuit is created.
    • 目的:提供一种基于语音的自然语言会话装置及其方法,用于减少用户和自动引导人工智能代理之间的语音识别误差,以通过使用高位置的语音识别临时安装来估计代理。 构成:输入单元输入用户触发声音。 语音识别单元(110)根据识别可靠性分数将用户声音的起始点从输入单元输入到降序。 语音识别电路创建了中间高位置N的语音识别临时安装。对应于输入理解单元(120)的语音识别临时安装的N的句子结构变成语音识别电路。