Aerial Inspection Behaviors via RL-based Quadrotor Control for Under-canopy Forest Environments

2026-05-18

AI Summary

This study addresses the challenge of balancing low-level control accuracy with high-level path planning safety and efficiency for autonomous drone inspection in complex understory environments. The authors propose a hierarchical navigation architecture: at the low level, an end-to-end deep reinforcement learning controller directly maps environmental states to motor speeds, enabling joint tracking of position and yaw angle; at the high level, task assignment and path planning are achieved by integrating the Traveling Salesman Problem (TSP) with the RRT* algorithm. This work presents the first tightly integrated framework combining an end-to-end RL controller with TSP-based task sequencing and RRT*-generated trajectories. Experimental validation across five representative understory scenarios demonstrates that the approach achieves high trajectory tracking accuracy, robust obstacle avoidance, and computational efficiency in long-range inspection missions.

Abstract

This paper addresses the problem of using a deep Reinforcement Learning (RL)-based low-level Quadrotor controller within an autonomous Quadrotor navigation stack for aerial inspection missions in under-canopy forest environments. Specifically, the article presents an end-to-end (mapping states to RPMs) Quadrotor control policy that achieves inspection view-pose tracking (simultaneous position and yaw reference tracking), which is crucial for various target inspection behaviors and point-to-point navigation in forests. To ensure safe and reliable deployment of the end-to-end RL controller in long-range missions, this article utilizes a higher navigation guidance layer comprising of a Traveling Salesman Problem planner (TSP) and a Rapidly-exploring Random Tree Star (RRT*) planner. Over a known map of a forest and a set of user-specified inspection regions, the TSP planner finds the optimal visitation sequence. Between two target regions, collision-free paths that respect the tracking limitations of the lower end-to-end RL policy are generated by an RRT* planner. Through five target inspection scenarios, this article demonstrates that an RL-based motor-level stabilizing controller, supported by a navigation guidance layer, can be used effectively as the low-level inspection execution module for under-canopy forest inspection missions.

Problem

• aerial inspection
• under-canopy forest
• quadrotor control
• view-pose tracking
• autonomous navigation

Innovation

• Reinforcement Learning
• Quadrotor Control
• Under-canopy Navigation
• View-pose Tracking
• Hierarchical Planning