APEX: A Decoupled Memory-based Explorer for Asynchronous Aerial Object Goal Navigation

2026-01-31

AI Summary

This work addresses the challenge of goal-oriented navigation for unmanned aerial vehicles (UAVs) in complex aerial environments, where the absence of effective spatial memory, interpretable decision-making, and efficient exploration mechanisms hinders performance. To overcome these limitations, the authors propose a hierarchical asynchronous parallel navigation agent that integrates a dynamic 3D semantic map, a reinforcement learning–driven control policy, and an open-vocabulary target recognition module. Leveraging a zero-shot vision-language model, the approach constructs an interpretable and decoupled semantic memory representation, while an asynchronous architecture mitigates inference latency, thereby enhancing exploratory initiative and generalization capability. Evaluated on the UAV-ON benchmark, the method achieves a 4.2% improvement in Success Rate (SR) and a 2.8% gain in Success weighted by Path Length (SPL), significantly outperforming existing approaches.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Planning with Language Models
• Search and Optimization: Learning to Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Aerial Object Goal Navigation, a challenging frontier in Embodied AI, requires an Unmanned Aerial Vehicle (UAV) agent to autonomously explore, reason, and identify a specific target using only visual perception and language description. However, existing methods struggle with the memorization of complex spatial representations in aerial environments, reliable and interpretable action decision-making, and inefficient exploration and information gathering. To address these challenges, we introduce \textbf{APEX} (Aerial Parallel Explorer), a novel hierarchical agent designed for efficient exploration and target acquisition in complex aerial settings. APEX is built upon a modular, three-part architecture: 1) Dynamic Spatio-Semantic Mapping Memory, which leverages the zero-shot capability of a Vision-Language Model (VLM) to dynamically construct high-resolution 3D Attraction, Exploration, and Obstacle maps, serving as an interpretable memory mechanism. 2) Action Decision Module, trained with reinforcement learning, which translates this rich spatial understanding into a fine-grained and robust control policy. 3) Target Grounding Module, which employs an open-vocabulary detector to achieve definitive and generalizable target identification. All these components are integrated into a hierarchical, asynchronous, and parallel framework, effectively bypassing the VLM's inference latency and boosting the agent's proactivity in exploration. Extensive experiments show that APEX outperforms the previous state of the art by +4.2\% SR and +2.8\% SPL on challenging UAV-ON benchmarks, demonstrating its superior efficiency and the effectiveness of its hierarchical asynchronous design. Our source code is provided in \href{https://github.com/4amGodvzx/apex}{GitHub}

Problem

• Aerial Object Goal Navigation
• spatial memory
• action decision-making
• exploration efficiency
• Embodied AI

Innovation

• Aerial Object Goal Navigation
• Vision-Language Model
• Hierarchical Asynchronous Architecture
• Dynamic Spatio-Semantic Mapping
• Open-Vocabulary Detection