The 6D pose estimation is important for the safe take-off and landing of the aircraft using a single RGB image. Due to the large scene and large depth, the exiting pose estimation methods have unstratified performance...The 6D pose estimation is important for the safe take-off and landing of the aircraft using a single RGB image. Due to the large scene and large depth, the exiting pose estimation methods have unstratified performance on the accuracy. To achieve precise 6D pose estimation of the aircraft, an end-to-end method using an RGB image is proposed. In the proposed method, the2D and 3D information of the keypoints of the aircraft is used as the intermediate supervision,and 6D pose information of the aircraft in this intermediate information will be explored. Specifically, an off-the-shelf object detector is utilized to detect the Region of the Interest(Ro I) of the aircraft to eliminate background distractions. The 2D projection and 3D spatial information of the pre-designed keypoints of the aircraft is predicted by the keypoint coordinate estimator(Kp Net).The proposed method is trained in an end-to-end fashion. In addition, to deal with the lack of the related datasets, this paper builds the Aircraft 6D Pose dataset to train and test, which captures the take-off and landing process of three types of aircraft from 11 views. Compared with the latest Wide-Depth-Range method on this dataset, our proposed method improves the average 3D distance of model points metric(ADD) and 5° and 5 m metric by 86.8% and 30.1%, respectively. Furthermore, the proposed method gets 9.30 ms, 61.0% faster than YOLO6D with 23.86 ms.展开更多
舰载机位姿实时检测对于甲板上的舰载机的运动控制、轨迹规划与防撞等具有重要意义。传统的舰载机调度主要依靠人工判断舰载机位置与航向角进行调度,传统方法不能得出准确数据,还易因为操作员的疏忽与疲劳发生碰撞事故。针对该问题,提...舰载机位姿实时检测对于甲板上的舰载机的运动控制、轨迹规划与防撞等具有重要意义。传统的舰载机调度主要依靠人工判断舰载机位置与航向角进行调度,传统方法不能得出准确数据,还易因为操作员的疏忽与疲劳发生碰撞事故。针对该问题,提出了舰载机位姿实时视觉测量算法。基于YOLO-V4(you only look once version 4)网络以及Canny边缘提取算法对舰载机进行识别分割。创新性地提出一种线框模板匹配算法,通过计算舰载机边缘轮廓与线框模板的匹配度获取最佳位姿。通过并行化与GPU(graphics processing unit)加速,使其满足实时性要求,并在1∶70与1∶14的实物模拟环境中完成测试。结果表明,该算法识别率在95%以上,位置精度在8 mm以内,姿态精度在0.7°以内,速度可达8 Hz。展开更多
A 6-degree of freedom （6-DOF） aircraft wing position and pose automatic adjustment method is presented to improve ARJ21 wing-fuselage connection precision and efficiency. Wing position and pose are adjusted by three...A 6-degree of freedom （6-DOF） aircraft wing position and pose automatic adjustment method is presented to improve ARJ21 wing-fuselage connection precision and efficiency. Wing position and pose are adjusted by three pillars which are driven by six high-precision servo motors. During the adjustment process, wing is tracked and positioned by laser tracker. Wing initial position and pose are calibrated by using the measurement coordinates of assembly reference points. Wing target position and pose are calculated according to wing initial, fuselage position and pose, and relative position and pose requirements between wing and fuselage for the connection. Combining Newton-Euler method with quaternion position and pose analyzing method, the inverse kinematics of servo motors, together with the adjustment system dynamics is obtained. Wing quintic polynomial trajectory planning algorithm based on quatemion is proposed; the initial, target position and pose need to be solved and the intermediate moving path is uncertain. Simulation results show that the adjustment method has good dynamic characteristics and satisfies engineering requirements. Preliminary engineering application indicates that ARJ21 wing adjustment efficiency and precision are improved by using the proposed method.展开更多
基金co-supported by the Key research and development plan project of Sichuan Province,China(No.2022YFG0153).
文摘The 6D pose estimation is important for the safe take-off and landing of the aircraft using a single RGB image. Due to the large scene and large depth, the exiting pose estimation methods have unstratified performance on the accuracy. To achieve precise 6D pose estimation of the aircraft, an end-to-end method using an RGB image is proposed. In the proposed method, the2D and 3D information of the keypoints of the aircraft is used as the intermediate supervision,and 6D pose information of the aircraft in this intermediate information will be explored. Specifically, an off-the-shelf object detector is utilized to detect the Region of the Interest(Ro I) of the aircraft to eliminate background distractions. The 2D projection and 3D spatial information of the pre-designed keypoints of the aircraft is predicted by the keypoint coordinate estimator(Kp Net).The proposed method is trained in an end-to-end fashion. In addition, to deal with the lack of the related datasets, this paper builds the Aircraft 6D Pose dataset to train and test, which captures the take-off and landing process of three types of aircraft from 11 views. Compared with the latest Wide-Depth-Range method on this dataset, our proposed method improves the average 3D distance of model points metric(ADD) and 5° and 5 m metric by 86.8% and 30.1%, respectively. Furthermore, the proposed method gets 9.30 ms, 61.0% faster than YOLO6D with 23.86 ms.
文摘舰载机位姿实时检测对于甲板上的舰载机的运动控制、轨迹规划与防撞等具有重要意义。传统的舰载机调度主要依靠人工判断舰载机位置与航向角进行调度,传统方法不能得出准确数据,还易因为操作员的疏忽与疲劳发生碰撞事故。针对该问题,提出了舰载机位姿实时视觉测量算法。基于YOLO-V4(you only look once version 4)网络以及Canny边缘提取算法对舰载机进行识别分割。创新性地提出一种线框模板匹配算法,通过计算舰载机边缘轮廓与线框模板的匹配度获取最佳位姿。通过并行化与GPU(graphics processing unit)加速,使其满足实时性要求,并在1∶70与1∶14的实物模拟环境中完成测试。结果表明,该算法识别率在95%以上,位置精度在8 mm以内,姿态精度在0.7°以内,速度可达8 Hz。
基金Basic Scientific Research Projects of Nanjing University of Aeronautics & Astronautics (NS 2010128)
文摘A 6-degree of freedom （6-DOF） aircraft wing position and pose automatic adjustment method is presented to improve ARJ21 wing-fuselage connection precision and efficiency. Wing position and pose are adjusted by three pillars which are driven by six high-precision servo motors. During the adjustment process, wing is tracked and positioned by laser tracker. Wing initial position and pose are calibrated by using the measurement coordinates of assembly reference points. Wing target position and pose are calculated according to wing initial, fuselage position and pose, and relative position and pose requirements between wing and fuselage for the connection. Combining Newton-Euler method with quaternion position and pose analyzing method, the inverse kinematics of servo motors, together with the adjustment system dynamics is obtained. Wing quintic polynomial trajectory planning algorithm based on quatemion is proposed; the initial, target position and pose need to be solved and the intermediate moving path is uncertain. Simulation results show that the adjustment method has good dynamic characteristics and satisfies engineering requirements. Preliminary engineering application indicates that ARJ21 wing adjustment efficiency and precision are improved by using the proposed method.
