-
公开(公告)号：KR101080073B1
公开(公告)日：2011-11-04
申请号：KR1020090016756
申请日：2009-02-27
Applicant: 숭실대학교산학협력단
IPC: G06T15/00
Abstract: 본발명은단일카메라를기반으로생성된이미지내에서특징점을추출하여가상의평면정보를구성한뒤 3차원대상물체의자세정보를획득한다. 가상의평면정보를적용함으로써 3차원자세정보에대한알려진변수와미지변수를분리시킬수 있으므로복잡한연산을간략화시킬수 있다. 영상에서정확한 3차원정보를검출할수 있으며, 수행시간을크게줄일수 있다. 또한각각의가상평면정보들을조합함으로써보다정확한자세정보를추출할수 있다.
-
Patent Agency Ranking
-
公开(公告)号：KR1020100097886A
公开(公告)日：2010-09-06
申请号：KR1020090016756
申请日：2009-02-27
Applicant: 숭실대학교산학협력단
IPC: G06T15/00
CPC classification number: G06T15/10 , G06K9/46 , G06T7/536 , G06T17/00 , G06T2207/10028 , G06T2219/028
Abstract: PURPOSE: A technique extracting geometry information of 3D object capable of simplifying a complex calculation by using a plurality of virtual plane information is provided to apply the virtual plane information, thereby dividing parameter informed about 3D pose information and an unknown parameter. CONSTITUTION: A specific feature about a target object which exist within an inputted image is extracted(11). A coordination relation of virtual plane about a target object fundamental coordinate system is composed(12). A relation about a plurality of virtual planes about the reference coordinate system is constructed(13). A depth map about each virtual plane is composed(14). The depth map about the object material is generated(16).
Abstract translation: 目的：提供能够通过使用多个虚拟平面信息来简化复杂计算的3D对象的几何信息提取技术，以应用虚拟平面信息，从而分配关于3D姿势信息和未知参数的参数。 构成：提取存在于输入图像内的目标对象的特定特征（11）。 虚拟平面关于目标对象基坐标系的协调关系由（12）组成。 关于围绕参考坐标系的多个虚拟平面的关系被构造（13）。 构成了关于每个虚拟平面的深度图（14）。 生成关于物体材料的深度图（16）。
-
公开(公告)号：KR1020100097881A
公开(公告)日：2010-09-06
申请号：KR1020090016749
申请日：2009-02-27
Applicant: 숭실대학교산학협력단
Abstract: PURPOSE: A balance control method using prediction of ZMP(Zero Moment Point) state of a Kalman filter is provided to ensure stable walking of a biped robot by compensating an error caused by complex dynamic analysis. CONSTITUTION: A balance control method using prediction of ZMP state of a kalman filter for stable walking of a biped robot is as follows. The actual dynamic ZMP is predicted through the kalman filter without performing complex dynamic analysis for control of the biped robot. The ZMP of the present walking period is converted into a COM(Center Of Mass) state value so as to calculate the exact ZMP of the next walking period of the biped robot(3).
Abstract translation: 目的：提供一种使用卡尔曼滤波器的ZMP（零矩点）状态预测的平衡控制方法，以通过补偿复杂动态分析引起的误差来确保双足机器人的稳定行走。 构成：使用卡尔曼滤波器的ZMP状态的预测的平衡控制方法，用于两足动物机器人的稳定行走如下。 实际动态ZMP是通过卡尔曼滤波器进行预测的，而不需要对双足机器人的控制进行复杂的动态分析。 本步行时段的ZMP被转换为COM（质心）状态值，以计算Biped机器人（3）的下一步行周期的确切ZMP。
-
公开(公告)号：KR100953431B1
公开(公告)日：2010-04-20
申请号：KR1020080018139
申请日：2008-02-28
Applicant: 숭실대학교산학협력단
Abstract: 본 발명은 복잡 다관절 2족보행로봇의 안정적 보행을 유지하는 기법을 제공한다. 보행의 안정성을 판별하기 위한 척도로서 ZMP(Zero Moment Point)가 주로 사용되고 있으며, 보행시 발바닥 접지면의 최외곽 영역에 의해 생성되는 안정영역(supported area) 안에 ZMP를 유지시켜 2족보행로봇을 안정화한다. 동적보행(Dynamic walking)을 수행하는 2족보행로봇은 균형를 쉽게 잃을 수 있기 때문에 다음 명령주기의 ZMP 위치를 Kalman 필터를 통해 예측하여 2족보행로봇의 보행을 안정화하는 균형 제어 기법을 제안한다. 예측된 ZMP 위치를 로봇의 역기구학 해석(Inverse Kinematics Analysis)을 통해 관절의 각속도로 보상함으로써 2족보행로봇의 균형 제어를 수행한다.
2족보행로봇, 동적보행, 보행 안정화, 균형제어, ZMP-
公开(公告)号：KR1020090092910A
公开(公告)日：2009-09-02
申请号：KR1020080018139
申请日：2008-02-28
Applicant: 숭실대학교산학협력단
Abstract: A balance control method of a biped robot using state prediction of ZMP(Zero Moment Point) is provided to minimize uncertain modeling by compensating the predicted ZMP location with the angular velocity of the joint through inverse kinematics analysis of robot. A balance control method of a biped robot using state prediction of ZMP includes a process of simplifying state values of a kinetic model of the biped robot to a three-dimensional linear inverse pendulum model and adaptively compensating the state values of the kinetic model using a Kalman filter, and a process(1) of predicting the ZMP state for the next instruction cycle of the biped robot using the Kalman filter.
Abstract translation: 提供使用ZMP（Zero Moment Point）状态预测的双足机器人的平衡控制方法，以通过机器人的逆运动学分析，通过用接头的角速度补偿预测的ZMP位置来最小化不确定建模。 使用ZMP的状态预测的双足机器人的平衡控制方法包括将两足动物机器人的动力学模型的状态值简化为三维线性逆摆模型的过程，并使用卡尔曼自适应地补偿动力学模型的状态值 过滤器以及使用卡尔曼滤波器预测用于两足动物机器人的下一个指令周期的ZMP状态的过程（1）。
-
-
-
-
-
Search Results
6
records
(took 0.017 seconds)
