EDEN: Efficient Dual-Layer Exploration Planning for Fast UAV Autonomous Exploration in Large 3-D Environments

Abstract: Efficient autonomous exploration in large-scale environments remains challenging due to the high planning computational cost and low-speed maneuvers. In this paper, we propose a fast and computationally efficient dual-layer exploration planning method. The insight of our dual-layer method is efficiently finding an acceptable long-term region routing and greedily exploring the target in the region of the first routing area with high speed. Specifically, the proposed method finds the long-term area routing through an approximate algorithm to ensure real-time planning in large-scale environments. Then, the viewpoint in the first routing region with the lowest curvature-penalized cost, which can effectively reduce decelerations caused by sharp turn motions, will be chosen as the next exploration target. To further speed up the exploration, we adopt an aggressive and safe exploration-oriented trajectory to enhance exploration continuity. The proposed method is compared to state-of-the-art methods in challenging simulation environments. The results show that the proposed method outperforms other methods in terms of exploration efficiency, computational cost, and trajectory speed. We also conduct real-world experiments to validate the effectiveness of the proposed method. The code will be open-sourced.

• The paper introduces a dual-layer planning approach that integrates heuristic region routing with curvature-penalized target optimization for enhanced UAV autonomy.
• The methodology employs the EOHC algorithm and ASEO trajectory planning, achieving up to 14% faster exploration and reducing computational costs by up to 74.5%.
• The study advances UAV exploration for applications such as search and rescue, environmental mapping, and industrial inspections in complex 3-D terrains.

EDEN: Efficient Dual-Layer Exploration Planning for UAVs

The paper "EDEN: Efficient Dual-Layer Exploration Planning for Fast UAV Autonomous Exploration in Large 3-D Environments", introduces a novel method designed to enhance the performance of UAVs in large-scale exploration tasks. Autonomous exploration by UAVs poses significant challenges, particularly in terms of computational costs and maneuvering efficiency. This paper proposes an innovative dual-layer exploration planning approach designed to address these challenges by optimizing both computation efficiency and UAV speed.

This research provides a detailed examination of the proposed EDEN system, introducing a dual-layer exploration method that highlights efficient routing and high-speed exploration. The paper makes a case for a dual-layer methodology that first determines large-scale region routing through a tailored heuristic algorithm and subsequently optimizes exploration targets within these areas. The approach employs a curvature-penalized criterion to minimize motion costs associated with sharp turns, thereby enhancing UAV trajectory speed.

Key Components of the EDEN System:

1. Exploration-Oriented Heuristic Christofides (EOHC) Algorithm: This modified algorithm is used to efficiently solve asymmetric traveling salesman problems (ATSP). The EOHC computes a suboptimal region routing that balances exploration cost with practical computational limits, capable of real-time adoption in expansive environments.
2. Curvature-Penalized Target Optimization: The method prioritizes targets that minimize curvature-penalized exploration gain, driving high-speed trajectory planning. By accounting for UAV kinematics, the system optimizes path selection to improve speed while ensuring flight stability and minimizing deceleration from sharp turns.
3. Aggressive and Safe Exploration-Oriented (ASEO) Trajectory: The trajectory model ensures exploration continuity while preserving safety. It encompasses exploring, continuous, and safety components to adaptively manage trajectory execution and provide backup pathways in the event of unexpected obstacles.

Performance Evaluation:

The proposed EDEN system was comparatively assessed against leading methods like FALCON and FUEL in multiple simulated environments. The experiments validated significant improvements in exploration efficiency, with exploration times reduced by 5.2%-14.0%, computational costs decreased by 28.9%-74.5%, and UAV speed improved by 10.4%-15.5%. Notably, the EDEN system proved effective in expansive environments where other methods faltered due to excessive computational resource requirements.

Implications and Future Directions:

This work proposes a substantial advancement in UAV autonomous exploration methodologies, opening avenues for practical deployment in search and rescue, environmental reconstruction, and industrial inspections. By streamlining exploration planning and trajectory execution, EDEN holds promise for reducing operational costs and increasing mission feasibility in complex 3-D terrains.

Future research may focus on extending adaptability and robustness of the EDEN framework to dynamic environmental changes, incorporating enhanced learning mechanisms for incremental improvement, and exploring multi-UAV coordination to expand exploration capacity and efficiency.

In conclusion, this paper contributes a significant step forward in UAV exploration planning, addressing critical limitations with a sophisticated dual-layer approach that optimizes efficiency, speed, and safety—advancing theoretical understanding and practical applications in the field.