GuangMing-Explorer: A Four-Legged Robot Platform for Autonomous Exploration in General Environments

TL;DR

Autonomous exploration in unknown environments demands tight integration of perception, planning, and control. The authors present GuangMing-Explorer, a fully integrated four‑legged platform with dual LiDARs, a camera, an onboard AI computer, and a robotic arm to enable robust exploration and mapping. They enhance a hierarchical exploration framework derived from TARE with spatial-temporal calibration and LiDAR-based odometry (Fast-LIO2) to reduce drift and improve map accuracy, validating the system in office and outdoor settings. Results show high environment coverage, real-time planning, and accurate mapping, underscoring the platform's practicality for deployment in unstructured environments and outlining future work on more complex scenarios and learning-based locomotion control.

Abstract

Autonomous exploration is a fundamental capability that tightly integrates perception, planning, control, and motion execution. It plays a critical role in a wide range of applications, including indoor target search, mapping of extreme environments, resource exploration, etc. Despite significant progress in individual components, a holistic and practical description of a completely autonomous exploration system, encompassing both hardware and software, remains scarce. In this paper, we present GuangMing-Explorer, a fully integrated autonomous exploration platform designed for robust operation across diverse environments. We provide a comprehensive overview of the system architecture, including hardware design, software stack, algorithm deployment, and experimental configuration. Extensive real-world experiments demonstrate the platform's effectiveness and efficiency in executing autonomous exploration tasks, highlighting its potential for practical deployment in complex and unstructured environments.

Figures (5)

• Figure 1: Pipeline for autonomous exploration on our four-legged robot platform.
• Figure 2: Configuration of the robotic platform.
• Figure 3: Point cloud of the explored environments: office (left) and parking lot (right).
• Figure 4: Temporal evolution of Test 3. The x-axis represents time while the y-axes (from top to bottom) shows the explored volume, traveled distance, and runtime per planning iteration.
• Figure 5: Evaluation of map accuracy. The red lines with numbers are selected lines for accuracy evaluation.