http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
RGB-D 카메라와 관성 항법을 결합한 Visual-Inertial Odometry 설계 및 복합항법 성능검증
김연조(Yeonjo Kim),이병진(Byungjin Lee),이영재(Young Jae Lee),성상경(Sangkyung Sung) 제어로봇시스템학회 2018 제어·로봇·시스템학회 논문지 Vol.24 No.8
In recent years, visual navigation has been extensively studied for the indoor/urban environments where the GNSS signals are unavailable. Visual navigation has the advantage that the sensor is lightweight and low-cost while providing information. However, vision sensor is very vulnerable to the dynamic scene, illumination change and motion blur. To overcome these shortcomings, the Vision/IMU(Inertial Measurement Unit) integrated system has been also studied actively. Therefore, in this paper, we present EKF (Extended Kalman Filter) based visual-inertial odometry algorithm. Unlike the traditional EKF based algorithm, we use a camera and VO(Visual Odometry) algorithm as a filter measurement. For the VO algorithm, we employed essential part of the ORB-SLAM, which uses ORB features. We performed an outdoor experiment using an RGB-D camera to evaluate the accuracy of the presented algorithm. Also, we evaluated our algorithm using the public dataset to compare with other visual navigation systems.
속도증분벡터를 활용한 ORB-SLAM 및 관성항법 결합 알고리즘 연구
김연조(Yeonjo Kim),손현진(Hyunjin Son),이영재(Young Jae Lee),성상경(Sangkyung Sung) 대한전기학회 2019 전기학회논문지 Vol.68 No.1
In recent years, visual-inertial odometry(VIO) algorithms have been extensively studied for the indoor/urban environments because it is more robust to dynamic scenes and environment changes. In this paper, we propose loosely coupled(LC) VIO algorithm that utilizes the velocity vectors from both visual odometry(VO) and inertial measurement unit(IMU) as a filter measurement of Extended Kalman filter. Our approach improves the estimation performance of a filter without adding extra sensors while maintaining simple integration framework, which treats VO as a black box. For the VO algorithm, we employed a fundamental part of the ORB-SLAM, which uses ORB features. We performed an outdoor experiment using an RGB-D camera to evaluate the accuracy of the presented algorithm. Also, we evaluated our algorithm with the public dataset to compare with other visual navigation systems.
복합항법시스템을 위한 Levenberg-Marquardt 최적화 기법 기반 자세추정 성능연구
김연조(Yeonjo Kim),이병진(Byungjin Lee),성상경(Sangkyung Sung) 제어로봇시스템학회 2018 제어·로봇·시스템학회 논문지 Vol.24 No.3
The initial alignment procedure estimating initial vehicle attitude has a large impact on the quality of navigation solutions. However, the conventional initial alignment procedure should be performed under static condition. In this paper, we propose initial attitude estimation method for INS/GNSS integrated system under dynamic condition. The proposed method uses two velocity differential vectors, one is expressed in initial body frame and the other is expressed in local navigation frame. Each velocity vector is derived from traditional navigation equation and can be calculated using IMU and GNSS measurements. These two vectors comprise cost function to estimate initial attitude. The cost function, also called Wahba’s problem, is solved through the Levenberg-Marquardt algorithm which provide fair estimation results for dynamic motions of ground vehicle.