AgilePilot: DRL-Based Drone Agent for Real-Time Motion Planning in Dynamic Environments by Leveraging Object Detection

2025-02-10

AI Summary

Addressing the challenge of achieving simultaneous safety and agility in real-time UAV motion planning within dynamic environments, this paper proposes an end-to-end navigation framework integrating Proximal Policy Optimization (PPO)-based deep reinforcement learning (DRL) with onboard YOLOv8 object detection. Methodologically, we introduce a Sim2Real transfer learning paradigm, design an environment-adaptive, safety-constrained decoupled dynamic potential field reward mechanism, and—novelty—we embed a velocity prediction module directly into the DRL policy to enhance dynamic target tracking accuracy. Experimental results demonstrate stable operation at 3.0 m/s, achieving threefold higher dynamic target tracking accuracy compared to conventional artificial potential field methods. Across 75 real-world flight trials, the system attains a 90% success rate, significantly improving responsiveness and robustness in complex, dynamic scenarios.

Technology Category

Computer Vision: Motion & TrackingHumans and AI: Human-Aware Planning and Behavior PredictionIntelligent Robots: Motion and Path Planning

Application Category

Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applicationsResponsible Web: Machine-in-the-loop, human agency and autonomySearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Autonomous drone navigation in dynamic environments remains a critical challenge, especially when dealing with unpredictable scenarios including fast-moving objects with rapidly changing goal positions. While traditional planners and classical optimisation methods have been extensively used to address this dynamic problem, they often face real-time, unpredictable changes that ultimately leads to sub-optimal performance in terms of adaptiveness and real-time decision making. In this work, we propose a novel motion planner, AgilePilot, based on Deep Reinforcement Learning (DRL) that is trained in dynamic conditions, coupled with real-time Computer Vision (CV) for object detections during flight. The training-to-deployment framework bridges the Sim2Real gap, leveraging sophisticated reward structures that promotes both safety and agility depending upon environment conditions. The system can rapidly adapt to changing environments, while achieving a maximum speed of 3.0 m/s in real-world scenarios. In comparison, our approach outperforms classical algorithms such as Artificial Potential Field (APF) based motion planner by 3 times, both in performance and tracking accuracy of dynamic targets by using velocity predictions while exhibiting 90% success rate in 75 conducted experiments. This work highlights the effectiveness of DRL in tackling real-time dynamic navigation challenges, offering intelligent safety and agility.

Problem

• Real-time drone motion planning in dynamic environments.
• Overcoming unpredictability with fast-moving objects.
• Enhancing adaptiveness and decision-making using DRL.

Innovation

• Deep Reinforcement Learning for drones
• Real-time Computer Vision integration
• Sim2Real training-to-deployment framework