Continuous World Coverage Path Planning for Fixed-Wing UAVs using Deep Reinforcement Learning

2025-05-13

AI Summary

This work addresses the energy-optimal coverage path planning (CPP) problem for fixed-wing unmanned aerial vehicles (UAVs) in continuous real-world environments. We formulate continuous CPP as a curvature-constrained Bézier curve trajectory optimization problem, explicitly incorporating kinematic feasibility. To solve it, we propose an Action-Mapping Soft Actor-Critic (AM-SAC) algorithm and integrate an adaptive curriculum learning mechanism to enhance training stability and generalization. Environment modeling employs axis-aligned rectangles with variable dimensions, balancing geometric fidelity and computational efficiency. Experiments demonstrate near-100% coverage in both generated and hand-crafted scenarios, with significantly lower energy consumption compared to discrete CPP baselines. Generated paths exhibit continuous curvature, enabling smoother, more efficient flight. The framework provides a verifiable, end-to-end planning solution suitable for practical deployment of fixed-wing UAVs.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Mixed Discrete/Continuous Planning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Unmanned Aerial Vehicle (UAV) Coverage Path Planning (CPP) is critical for applications such as precision agriculture and search and rescue. While traditional methods rely on discrete grid-based representations, real-world UAV operations require power-efficient continuous motion planning. We formulate the UAV CPP problem in a continuous environment, minimizing power consumption while ensuring complete coverage. Our approach models the environment with variable-size axis-aligned rectangles and UAV motion with curvature-constrained B'ezier curves. We train a reinforcement learning agent using an action-mapping-based Soft Actor-Critic (AM-SAC) algorithm employing a self-adaptive curriculum. Experiments on both procedurally generated and hand-crafted scenarios demonstrate the effectiveness of our method in learning energy-efficient coverage strategies.

Problem

• Continuous coverage path planning for fixed-wing UAVs
• Minimizing power consumption in continuous environments
• Learning energy-efficient strategies using deep reinforcement learning

Innovation

• Continuous coverage planning with deep reinforcement learning
• Curvature-constrained Bézier curves for UAV motion
• Action-mapping-based Soft Actor-Critic (AM-SAC) algorithm