首页|New Robotics Research Reported from Jeonbuk National University (Uncertainty-Awa re Depth Network for Visual Inertial Odometry of Mobile Robots)
New Robotics Research Reported from Jeonbuk National University (Uncertainty-Awa re Depth Network for Visual Inertial Odometry of Mobile Robots)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
024 OCT 31 (NewsRx)-By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News-Researchers detail new data in robotic s. According to news reporting originating fromJeonju, South Korea, by NewsRx c orrespondents, research stated, "Simultaneous localization and mapping,a critic al technology for enabling the autonomous driving of vehicles and mobile robots, increasinglyincorporates multi-sensor configurations."Funders for this research include Institute of Information & Commu nications Technology Planning &Evaluatio; National Research Found ation of Korea.Our news reporters obtained a quote from the research from Jeonbuk National Univ ersity: "Inertialmeasurement units (IMUs), known for their ability to measure a cceleration and angular velocity, are widelyutilized for motion estimation due to their cost efficiency. However, the inherent noise in IMU measurementsnecess itates the integration of additional sensors to facilitate spatial understanding for mapping. Visualinertialodometry (VIO) is a prominent approach that combin es cameras with IMUs, offering high spatialresolution while maintaining cost-ef fectiveness. In this paper, we introduce our uncertainty-aware depthnetwork (UD -Net), which is designed to estimate both depth and uncertainty maps. We propose a novelloss function for the training of UD-Net, and unreliable depth values a re filtered out to improve VIOperformance based on the uncertainty maps."
Jeonbuk National UniversityJeonjuSou th KoreaAsiaEmerging TechnologiesMachine LearningNano-robotRobotics
NETL
NSTL
