Abstract:Exploring an unknown environment without colliding with obstacles is one of the essentials of autonomous vehicles to perform diverse missions such as structural inspections, rescues, deliveries, and so forth. Therefore, unmanned aerial vehicles (UAVs), which are fast, agile, and have high degrees of freedom, have been widely used. However, previous approaches have two limitations: a) First, they may not be appropriate for exploring large-scale environments because they mainly depend on random sampling-based path planning that causes unnecessary movements. b) Second, they assume the pose estimation is accurate enough, which is the most critical factor in obtaining an accurate map. In this paper, to explore and map unknown large-scale environments rapidly and accurately, we propose a novel exploration method that combines the pre-calculated Peacock Trajectory with graph-based global exploration and active loop-closing. Because the two-step trajectory that considers the kinodynamics of UAVs is used, obstacle avoidance is guaranteed in the receding-horizon manner. In addition, local exploration that considers the frontier and global exploration based on the graph maximizes the speed of exploration by minimizing unnecessary revisiting. In addition, by actively closing the loop based on the likelihood, pose estimation performance is improved. The proposed method's performance is verified by exploring 3D simulation environments in comparison with the state-of-the-art methods. Finally, the proposed approach is validated in a real-world experiment.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to perform fast and accurate large - scale 3D mapping in an unknown environment, especially the main challenges faced when using unmanned aerial vehicles (UAVs) to perform this task. Specifically, the paper makes improvements on two main limitations in existing methods: 1. **Inapplicability of large - scale environmental exploration**: Most of the existing methods rely on path planning based on random sampling, which will lead to unnecessary movement and thus reduce the exploration speed. 2. **Accuracy assumption of pose estimation**: Most methods assume that pose estimation is accurate enough, but in practical applications, especially in large - scale environmental exploration, this assumption is often not valid, resulting in inaccurate map construction. To solve these problems, the paper proposes a new exploration method - **Rapid Exploration with Active Loop - Closing (REAL)**. This method combines the pre - calculated peacock trajectory, graph - based global exploration and active loop - closing techniques, aiming to achieve fast, efficient and accurate large - scale 3D environmental mapping. ### Main contributions 1. **Controlling efficient peacock trajectory**: Considering the dynamic characteristics of UAVs, it avoids inefficient random sampling and optimizes obstacle avoidance and frontier exploration through the local exploration scoring matrix. 2. **Graph - based global exploration planner**: Using the optimal A* algorithm, it prevents unnecessary round - trip movements by considering the cost function of the frontier cluster, maximizes the planning efficiency and improves the exploration speed. 3. **Active loop - closing**: It reduces the drift of pose estimation by maximizing the likelihood, which is particularly important for accurate mapping in large - scale environments. 4. **Experimental verification**: The effectiveness of the proposed method is verified in simulation and real - world experiments. ### Solutions 1. **Local exploration**: Local exploration is carried out through the peacock trajectory and the scoring matrix to ensure the efficiency of obstacle avoidance and frontier exploration. 2. **Global exploration**: Graph - based global exploration optimizes path planning through the gain function to prevent repeated visits and dead - end situations. 3. **Active loop - closing**: The loop is actively closed through likelihood selection to improve the accuracy of pose estimation. ### Experimental results The paper conducted experiments in small - scale and large - scale simulation environments and compared with existing methods. The results show that the REAL method is superior to existing methods in terms of exploration speed and pose estimation accuracy, especially in large - scale environments. In conclusion, by proposing the REAL method, this paper effectively solves the key problems in large - scale environmental exploration and provides a new solution for UAVs to efficiently map in unknown environments.

REAL: Rapid Exploration with Active Loop-Closing toward Large-Scale 3D Mapping using UAVs

Peacock Exploration: A Lightweight Exploration for UAV using Control-Efficient Trajectory

Autonomous 3D Exploration of Large Structures Using an UAV Equipped with a 2D LIDAR

An autonomous UAV system based on adaptive LiDAR inertial odometry for practical exploration in complex environments

Autonomous Exploration Method for Fast Unknown Environment Mapping by Using UAV Equipped with Limited FOV Sensor

Multi-Drone Collaborative Trajectory Optimization for Large-Scale Aerial 3D Scanning

Multi-map Collaborative Exploration Toward Large-Scale 3D Mapping Using UAV *

RACER: Rapid Collaborative Exploration With a Decentralized Multi-UAV System

A Motion Camouflage-Inspired Path Planning Method for UAVs Based on Reinforcement Learning

EPIC: A Lightweight LiDAR-Based UAV Exploration Framework for Large-Scale Scenarios

Aerial path planning for online real-time exploration and offline high-quality reconstruction of large-scale urban scenes

Autonomous Exploration of Unknown 3D Environments Using a Frontier-Based Collector Strategy

STExplorer: A Hierarchical Autonomous Exploration Strategy with Spatio-Temporal Awareness for Aerial Robots

EFP: Efficient Frontier-Based Autonomous UAV Exploration Strategy for Unknown Environments

Deep-learning based autonomous-exploration for UAV navigation

Efficient and High Path Quality Autonomous Exploration and Trajectory Planning of UAV in an Unknown Environment

Efficient Submap-based Autonomous MAV Exploration using Visual-Inertial SLAM Configurable for LiDARs or Depth Cameras

Autonomous Area Exploration and Mapping in Underground Mine Environments by Unmanned Aerial Vehicles

Towards Multi-robot Exploration: A Decentralized Strategy for UAV Forest Exploration

A Fast Planning Approach for 3D Short Trajectory with a Parallel Framework

Onboard Real-Time Dense Reconstruction in Large Terrain Scene Using Embedded UAV Platform