Search papers, labs, and topics across Lattice.
This paper introduces WinTA-GIL, a novel framework for heading refinement that integrates GNSS, IMU, and LiDAR data through a temporal window-based optimization strategy, addressing the challenges of heading estimation in intermittent signal environments. By transforming heading estimation into a trajectory-based consistency optimization problem, the method dynamically adapts to refine estimates and suppress inertial drift during navigation. Experimental results show that WinTA-GIL significantly enhances both the accuracy and robustness of heading estimation compared to existing state-of-the-art techniques.
Heading estimation can now be dynamically refined in challenging environments, drastically reducing drift and improving accuracy during GNSS outages.
Although multi-source fusion positioning systems have achieved significant progress, accurate and reliable heading estimation remains a critical challenge due to the lack of gravitational constraints and the inherent weak observability of heading in complex environments. Most existing methodologies are specifically tailored for the startup phase, relying on a singular initial alignment to establish the heading reference. Consequently, these approaches lack the adaptability required to refine heading estimates dynamically, which renders the system highly vulnerable to accumulated drift and observation noise during prolonged navigation or immediately following GNSS signal outages. To address these limitations, this paper proposes WinTA-GIL, a novel heading refinement framework that integrates information from Global Navigation Satellite System (GNSS), Inertial Measurement Unit (IMU), and Light Detection and Ranging (LiDAR) through a temporal window-based optimization strategy. Unlike conventional alignment methods restricted to the startup phase, WinTA-GIL leverages high-precision local trajectories from LiDAR-Inertial Odometry (LIO) to register against filtered GNSS observations. This approach transforms heading estimation into a repeatable, trajectory-based consistency optimization problem. In particular, an adaptive re-estimation mechanism based on state discrimination is incorporated to trigger heading corrections whenever necessary, thereby effectively suppressing the inertial drift accumulated during challenging conditions. Extensive experiments on both open-source and self-collected datasets demonstrate that WinTA-GIL significantly outperforms state-of-the-art approaches in both estimation accuracy and system robustness.
