WLTCL: Wide Field-of-View 3-D LiDAR Truck Compartment Automatic Localization System

TL;DR

This work tackles the problem of automatic, accurate localization of fenced truck compartments to enable autonomous loading. It presents a hardware-software solution based on a rotating wide-field-of-view 3-D LiDAR and a geometry-driven key-point localization pipeline that yields a unified loading-space coordinate frame across varying truck sizes. The approach combines world-coordinate establishment, robust vehicle point-cloud segmentation, and contour-based localization to identify eight compartment corners, demonstrated across large, medium, and small trucks with low CPU-time requirements. The results show strong localization accuracy and robustness in cluttered environments, highlighting practical applicability for logistics automation and automated loading systems.

Abstract

As an essential component of logistics automation, the automated loading system is becoming a critical technology for enhancing operational efficiency and safety. Precise automatic positioning of the truck compartment, which serves as the loading area, is the primary step in automated loading. However, existing methods have difficulty adapting to truck compartments of various sizes, do not establish a unified coordinate system for LiDAR and mobile manipulators, and often exhibit reliability issues in cluttered environments. To address these limitations, our study focuses on achieving precise automatic positioning of key points in large, medium, and small fence-style truck compartments in cluttered scenarios. We propose an innovative wide field-of-view 3-D LiDAR vehicle compartment automatic localization system. For vehicles of various sizes, this system leverages the LiDAR to generate high-density point clouds within an extensive field-of-view range. By incorporating parking area constraints, our vehicle point cloud segmentation method more effectively segments vehicle point clouds within the scene. Our compartment key point positioning algorithm utilizes the geometric features of the compartments to accurately locate the corner points, providing stackable spatial regions. Extensive experiments on our collected data and public datasets demonstrate that this system offers reliable positioning accuracy and reduced computational resource consumption, leading to its application and promotion in relevant fields.

Figures (19)

• Figure 1: Localization system execution steps. (a) The vehicle enters the parking area. (b) Wide-field-of-view LiDAR scans the scene point cloud. (c) Establish world coordinate system. (d) Vehicle point cloud segmentation and compartment key point localization.
• Figure 2: System overview. First, during the startup phase, the system needs to load the parameter configuration of the software and hardware modules. Second, the vehicle enters the parking area, and the wide-field-of-view LiDAR starts scanning to generate the scene point cloud. Then, the system establishes a loading coordinate system and segments the vehicle point cloud. Finally, the compartment key point localization algorithm identifies the key points of the compartment and outputs the loadable area, completing the compartment localization.
• Figure 3: Establish world coordinate system. (a) Installation location. Four reflective boards are installed at the corners of the parking area, with red dots indicating the center of each reflective board. (b) World coordinate system. The global coordinate system $(O)$ is established based on these reflective boards. Blue dots represent the compartment corner points, and the blue area denotes the loading area, all indicated under coordinate system $(A)$.
• Figure 4: Visualization of public dataset vehicles. The red box area shows that the vehicles have occlusions and missing point clouds at the rear.
• Figure 5: Illustration of the contour fused and completion strategy. (a) Three cases of line segment fusion within the plane. The blue and red segments represent two line segments that satisfy the merging criteria, while the black points are the endpoints of the new line segment. (b) Three cases of truck compartment contour line completion. The green and yellow segments represent the compartment edge segments, and the black lines and points indicate the newly grown line segments and endpoints, respectively.