Autonomous Exploration with Terrestrial-Aerial Bimodal Vehicles

2025-07-28

AI Summary

This work addresses autonomous exploration under joint energy and time constraints in complex environments using ground-air multimodal vehicles. Methodologically, it introduces a hierarchical exploration framework that integrates environmental perception with information-gain assessment to generate high-value observation waypoints. It proposes an extended Monte Carlo Tree Search (MCTS) algorithm that jointly optimizes modality-switching decisions and waypoint sequencing, enabling adaptive path planning under multiple constraints. Further, the framework incorporates dual-perspective modeling, an improved motion planner, and a reinforcement learning–inspired heuristic policy to realize an end-to-end energy- and time-aware exploration system. Experiments in simulation and on a real-world ground-air multimodal platform demonstrate significant improvements: +32% coverage per unit energy consumption and higher long-horizon task completion rates. Results validate the framework’s effectiveness and robustness in resource-constrained scenarios.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Multiagent Systems: Multiagent Planning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Terrestrial-aerial bimodal vehicles, which integrate the high mobility of aerial robots with the long endurance of ground robots, offer significant potential for autonomous exploration. Given the inherent energy and time constraints in practical exploration tasks, we present a hierarchical framework for the bimodal vehicle to utilize its flexible locomotion modalities for exploration. Beginning with extracting environmental information to identify informative regions, we generate a set of potential bimodal viewpoints. To adaptively manage energy and time constraints, we introduce an extended Monte Carlo Tree Search approach that strategically optimizes both modality selection and viewpoint sequencing. Combined with an improved bimodal vehicle motion planner, we present a complete bimodal energy- and time-aware exploration system. Extensive simulations and deployment on a customized real-world platform demonstrate the effectiveness of our system.

Problem

• Optimizing exploration using terrestrial-aerial bimodal vehicles
• Managing energy and time constraints in autonomous exploration
• Strategic modality selection and viewpoint sequencing for efficiency

Innovation

• Hierarchical framework for bimodal exploration
• Extended Monte Carlo Tree Search optimization
• Improved bimodal vehicle motion planner