The integrated inertial system, typically integrating an IMU and an exteroceptive sensor, such as radar, light detection and ranging (LiDAR), and camera, has been widely accepted and applied in modern robotic applications for ego-motion estimation, motion control, or autonomous exploration. To improve system accuracy, robustness, and further usability, both multiple and various sensors are generally resiliently integrated, which benefits the system performance regarding failure tolerance, perception capability, and environment compatibility. For such systems, accurate and consistent spatiotemporal calibration is required to maintain a unique spatiotemporal framework for multisensor fusion. Considering that most existing calibration methods first, are generally oriented to specific integrated inertial systems, second, often focus on spatial-only determination, and third, usually require artificial targets, lacking convenience and usability, we proposeiKalibr:a unified targetless spatiotemporal calibration framework for resilient integrated inertial systems, which overcomes the above issues, and enables both accurate and consistent calibration. Altogether four commonly employed sensors are supported iniKalibrcurrently, namely, IMU, radar, LiDAR, and camera. The proposed method starts with a rigorous and efficient dynamic initialization, where all parameters in the estimator would be accurately recovered. Subsequently, several continuous-time batch optimizations are conducted to refine the initialized parameters toward better states. Sufficient real-world experiments were conducted to verify the feasibility and evaluate the calibration performance ofiKalibr. The results demonstrate thatiKalibrcan achieve accurate resilient spatiotemporal calibration.

For advanced autonomous systems, precise and consistent ego-motion estimation is critical to ensure system safety and performant closed-loop control. Unfortunately, traditional discrete-time approaches generally rely on interpolation to fuse asynchronous measurements, which would inevitably lose accuracy in high-speed motions. To this end, we propose a tightly coupled LiDAR/IMU/Radar odometry tailored for accurate and smooth state estimation using continuous-time trajectory optimization. Specifically, a novel coarse-to-fine dynamic initialization method is first proposed to obtain system states, IMU biases, and optional intersensor spatiotemporal parameters without requiring specific artificial targets and movements. After performing data association for LiDAR scans and outlier removal for Radar scans, LiDAR point-to-plane factor, angular velocity factor, linear acceleration factor, and Radar Doppler velocity factor are jointly minimized in a factor graph framework to derive optimal estimates. A series of dedicated experiments on the real-world data sets were conducted to evaluate the proposed method. The results demonstrate that the proposed method exhibits superior navigation performance in complex environments than the state-of-the-art methods in terms of position, velocity, and attitude.

Nowadays, LiDAR-IMU systems have progressively prevailed in mobile robotic applications due to their excellent complementary characteristics. With the steady decline and shrinking in cost and size of these sensors, it has become feasible and even imperative to further leverage multiple sensor units for better accuracy and robustness. In such a fusion-based system, accurate spatiotemporal calibration is a fundamental prerequisite. However, existing calibration methods generally necessitate artificial targets or an overlapping field-of-view (FoV) between LiDARs. To this end, we propose an accurate and easy-to-use and spatiotemporal calibration approach tailored to multi-LiDAR multi-IMU systems based on continuous-time batch estimation, which supports both mechanical spinning LiDARs and small FoV solid-state LiDARs. Inspired by classical hand–eye calibration, a stepwise multistage nonlinear optimization problem is first built to recover the rotational extrinsic and poses of control points in the spline without using any artificial targets, special movements, or manual intervention. Meanwhile, a virtual IMU and LiDAR are constructed to bridge all IMUs and LiDARs based on the centralized principles in both temporal and spatial domains. Subsequently, the point-to-plane factors associated by different LiDARs, gyroscope, and accelerometer factors are jointly minimized to optimize all spatiotemporal parameters over multiple batches. Extensive simulation tests and real-world experiments were conducted to quantitatively evaluate the feasibility of the proposed method. Meanwhile, the proposed method is compared with the state-of-the-art methods and shows superior calibration performance.

With significant advances in manufacturing technology, multiple sensors such as global navigation satellite system (GNSS) devices, inertial measurement units (IMU) and cameras have become affordable in intelligent vehicle systems and are generally integrated for motion estimation and environmental perception. In such integrated systems, accurate and flexible sensor calibration is a basic foundation for optimal information fusion. However, traditional calibration methods are generally labor-intensive or only applicable to handheld platforms. To address this issue, we propose an accurate and plug-and-play self-calibration method for in-vehicle GNSS/inertial navigation system (INS)/Vision system. Without relying on specially-designed targets and initial guess of sensor setups, all involved sensing parameters, including GNSS-IMU and Camera-IMU spatial-temporal extrinsics, are properly initialized and estimated online via a centralized Extended Kalman Filter (EKF). Besides, an observability constrained module is also developed to address degradation cases through detailed observability analysis. The results in both Monte-Carlo simulations and real-world experiments indicate that the proposed method enables accurate online calibration of all spatiotemporal parameters involved in GNSS/INS/Vision systems and shows superior calibration performance over the state-of-the-art methods. Furthermore, for the first time, we prove the superiority of using raw GNSS observations to enhance camera-IMU calibration than using processed GNSS positions or not using any GNSS information.