A Tightly-Coupled Framework for Large-Scale Map Construction With Multiple Non-Repetitive Scanning LiDARs

With the ability of providing direct and accurate enough range measurements, light detection and ranging (LiDAR) is playing an essential role in localization and detection for autonomous vehicles. Since single LiDAR suffers from hardware failure and performance degradation intermittently, we present a multi-LiDAR integration scheme in this article. Our framework tightly couples multiple non-repetitive scanning LiDARs with inertial, wheel encoder, and real-time kinematic (RTK) into pose estimation and simultaneous global map generation. Primarily, we formulate a precise synchronization strategy to integrate isolated sensors, and the extracted feature points from separate LiDARs are merged into a single sweep. The fused scans are introduced to compute the scan-matching correspondences, which can be further refined by additional RTK measurements. Based thereupon, we construct a factor graph along with the inertial preintegration result, estimated ground constraints, and RTK data. For the purpose of maintaining a restricted number of poses for estimation, we deploy a keyframe based sliding-window optimization strategy in our system. The real-time performance is guaranteed with multi-threaded computation, and extensive experiments are conducted in challenging scenarios. Experimental results show that the utilization of multiple LiDARs boosts the system performance in both robustness and accuracy.

I. Introduction

Robust global localization is a fundamental property of building high definition (HD) maps for autonomous driving [1]. Current vehicular positioning schemes are monopolized by global navigation satellite systems (GNSS), for instance, the global positioning system (GPS). Previous studies have reported that the accuracy of single-point positioning (SPP) is about 10 m considering clock errors, satellite orbit, ionospheric and tropospheric delays. These errors can be calibrated out utilizing external correction sources, with precise point positioning (PPP) and real-time kinematic (RTK) two approaches in this category. However, these approaches are limited by GNSS signal quality, and require external sensors to compensate for signal blockage and multipath problems [2]–[5]. As a direct measurement of 3D environment and invulnerable to illumination variations, light detection and ranging (LiDAR) is a promising sensor towards this demanding task.