Air-Ground Collaboration with SPOMP: Semantic Panoramic Online Mapping and Planning

TL;DR

SPOMP addresses the challenge of coordinating heterogeneous air-ground robot teams under GPS-denied conditions by leveraging semantic panoramas and depth-based local planning. The framework integrates ASOOM-based aerial mapping, cross-view semantic localization, and a traversability-graph–based global planner, with online learning that extrapolates ground-truth traversability from aerial context. Real-world experiments across urban and rural environments show high mission success, with 96% of goals reached and substantial autonomous distance traveled, validating scalability and robustness. The work demonstrates the practical viability of semantic representations for large-scale multi-robot collaboration under real communication and perception constraints.

Abstract

Mapping and navigation have gone hand-in-hand since long before robots existed. Maps are a key form of communication, allowing someone who has never been somewhere to nonetheless navigate that area successfully. In the context of multi-robot systems, the maps and information that flow between robots are necessary for effective collaboration, whether those robots are operating concurrently, sequentially, or completely asynchronously. In this paper, we argue that maps must go beyond encoding purely geometric or visual information to enable increasingly complex autonomy, particularly between robots. We propose a framework for multi-robot autonomy, focusing in particular on air and ground robots operating in outdoor 2.5D environments. We show that semantic maps can enable the specification, planning, and execution of complex collaborative missions, including localization in GPS-denied settings. A distinguishing characteristic of this work is that we strongly emphasize field experiments and testing, and by doing so demonstrate that these ideas can work at scale in the real world. We also perform extensive simulation experiments to validate our ideas at even larger scales. We believe these experiments and the experimental results constitute a significant step forward toward advancing the state-of-the-art of large-scale, collaborative multi-robot systems operating with real communication, navigation, and perception constraints.

Figures (21)

• Figure 1: The SPOMP architecture described in terms of ROS nodes. Sensors and actuators are shown in orange and visualization outputs in green.
• Figure 2: Simplified state machine determining the UAV high level goal.
• Figure 3: State machine describing the algorithm for assigning goals to ground vehicles.
• Figure 4: The Semantic Integrated LiDAR Labelling (SILL) tool and output labelled depth panoramas.
• Figure 5: Visualization of the panorama after each processing step, and the final projected obstacle point cloud.