J Shanghai Jiaotong Univ Sci ›› 2023, Vol. 28 ›› Issue (1): 70-76.doi: 10.1007/s12204-023-2570-6

Previous Articles     Next Articles

Lidar-Visual-Inertial Odometry with Online Extrinsic Calibration


MAO Tianyang (茅天阳), ZHAO Wentao (赵文韬), WANG Jingchuan∗ (王景川), CHEN Weidong (陈卫东)   

  1. (Department of Automation, Institute of Medical Robotics, Shanghai Jiao Tong University, Shanghai 200240, China; Key Laboratory of System Control and Information Processing of Ministry of Education, Shanghai 200240, China; Shanghai Engineering Research Center of Intelligent Control and Management, Shanghai 200240, China)
  2. (上海交通大学 医疗机器人研究院 自动化系,上海200240;系统控制与信息处理教育部重点实验室,上海200240;上海工业智能管控工程技术研究中心,上海200240)
  • Received:2021-12-10 Online:2023-01-28 Published:2023-02-10

Abstract: To achieve precise localization, autonomous vehicles usually rely on a multi-sensor perception system surrounding the mobile platform. Calibration is a time-consuming process, and mechanical distortion will cause extrinsic calibration errors. Therefore, we propose a lidar-visual-inertial odometry, which is combined with an adapted sliding window mechanism and allows for online nonlinear optimization and extrinsic calibration. In the adapted sliding window mechanism, spatial-temporal alignment is performed to manage measurements arriving at different frequencies. In nonlinear optimization with online calibration, visual features, cloud features, and inertial measurement unit (IMU) measurements are used to estimate the ego-motion and perform extrinsic calibration. Extensive experiments were carried out on both public datasets and real-world scenarios. Results indicate that the proposed system outperforms state-of-the-art open-source methods when facing challenging sensor-degenerating conditions.

Key words: autonomous vehicles, multi-sensor fusion, online calibration

摘要: 为了实现精确定位,自动驾驶汽车通常依赖分布于移动平台周围的多传感器感知系统。标定是一个耗时的过程,机械形变会导致外参标定误差。因此,我们提出了一种激光-视觉-惯性里程计,它与自适应滑动窗口机制相结合,允许在线非线性优化和外参标定。在自适应滑动窗口机制中,利用时空对齐管理到达不同频率的测量。在具有在线标定的非线性优化中,视觉特征、点云特征和惯性测量单元(IMU)测量用于估计自身运动并执行外参标定。在公共数据集和真实场景中进行了大量实验。结果表明,当面临具有挑战性的传感器退化条件时,所提出的系统优于最先进的开源方法。

关键词: 自动驾驶,多传感器融合,在线标定

CLC Number: