RAPID: Robust and Agile Planner Using Inverse Reinforcement Learning for Vision-Based Drone Navigation

2025-02-04

AI Summary

Addressing the challenge of high-speed visual navigation for UAVs in complex, cluttered environments, this paper proposes an end-to-end vision-based planner grounded in inverse reinforcement learning (IRL), overcoming key limitations of behavioral cloning—namely, error accumulation—and of conventional reinforcement learning—including reward engineering difficulty and low sample efficiency. To our knowledge, this is the first application of IRL to high-speed visual navigation. Our approach jointly leverages expert trajectories encoded as motion primitives and agent interaction data to enable zero-shot sim-to-real transfer. The method operates without explicit mapping or modular components, generating collision-free waypoints in milliseconds. Experimental validation demonstrates average and peak flight speeds of 7 m/s and 8.8 m/s, respectively, with robust autonomous navigation across diverse real-world scenarios—including dense forests and multi-type urban structures—thereby confirming both efficacy and strong generalization capability.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This paper introduces a learning-based visual planner for agile drone flight in cluttered environments. The proposed planner generates collision-free waypoints in milliseconds, enabling drones to perform agile maneuvers in complex environments without building separate perception, mapping, and planning modules. Learning-based methods, such as behavior cloning (BC) and reinforcement learning (RL), demonstrate promising performance in visual navigation but still face inherent limitations. BC is susceptible to compounding errors due to limited expert imitation, while RL struggles with reward function design and sample inefficiency. To address these limitations, this paper proposes an inverse reinforcement learning (IRL)-based framework for high-speed visual navigation. By leveraging IRL, it is possible to reduce the number of interactions with simulation environments and improve capability to deal with high-dimensional spaces while preserving the robustness of RL policies. A motion primitive-based path planning algorithm collects an expert dataset with privileged map data from diverse environments, ensuring comprehensive scenario coverage. By leveraging both the acquired expert and learner dataset gathered from the agent's interactions with the simulation environments, a robust reward function and policy are learned across diverse states. While the proposed method is trained in a simulation environment only, it can be directly applied to real-world scenarios without additional training or tuning. The performance of the proposed method is validated in both simulation and real-world environments, including forests and various structures. The trained policy achieves an average speed of 7 m/s and a maximum speed of 8.8 m/s in real flight experiments. To the best of our knowledge, this is the first work to successfully apply an IRL framework for high-speed visual navigation of drones.

Problem

• Enables agile drone flight in cluttered environments.
• Reduces interactions with simulation via IRL.
• Achieves high-speed navigation without additional tuning.

Innovation

• Inverse Reinforcement Learning framework
• Motion primitive-based path planning
• Simulation-trained real-world application