Research Analyzer
← Back ICRA 2026

Autonomous Exploration with Terrestrial-Aerial Bimodal Vehicles

Yuman Gao, Ruibin Zhang, Tiancheng Lai, Yanjun Cao, Chao Xu, Fei Gao

PDF

AI summary

Key figure (auto-extracted from paper)
An energy- and time-aware hierarchical planning framework enables efficient autonomous exploration for terrestrial-aerial bimodal vehicles.
Autonomous exploration Terrestrial-aerial vehicles Bimodal planning Monte Carlo Tree Search Energy-aware robotics Mobile robotics

Problem

Single-platform robots struggle to balance mobility and endurance, while multi-robot systems introduce prohibitive coordination complexity. Practical exploration missions are also frequently constrained by finite energy and time budgets that prior work often overlooks.

Approach

The system generates candidate ground and aerial viewpoints, then uses an extended Monte Carlo Tree Search to optimally sequence them and switch locomotion modes while strictly respecting energy and time limits.

Key results

  • Hierarchical exploration framework with bimodal viewpoint generation and constraint-aware decision-making
  • Bimodal Monte Carlo Tree Search (BM-MCTS) for adaptive modality and viewpoint sequencing
  • Integration with an enhanced bimodal motion planner for terrain-aware trajectory generation
  • Validation through extensive simulations and real-world deployment on a customized TABV platform

Why it matters

Enables reliable, long-duration autonomous exploration in complex or communication-denied environments by leveraging complementary ground and aerial locomotion on a single platform.

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 infor- mation to identify informative regions, we generate a set of poten- tial 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.

Index terms

Aerial Systems: Applications Task and Motion Planning Aerial Systems: Perception and Autonomy

Related papers