Motion Control in Multi-Rotor Aerial Robots Using Deep Reinforcement Learning

2025-02-09

AI Summary

To address the insufficient robustness of real-time motion control for multirotor aerial robots in additive manufacturing (AM) under dynamic payload variations and external disturbances, this paper proposes a curriculum-learning-enhanced Twin Delayed Deep Deterministic Policy Gradient (TD3) deep reinforcement learning framework. The method systematically validates, for the first time in drone-based AM tasks, TD3’s superior training stability, trajectory tracking accuracy, and task success rate over Deep Deterministic Policy Gradient (DDPG). Integrating multirotor dynamics modeling, real-time closed-loop simulation, and physical experiments, the proposed policy achieves a 98.2% task success rate under ±40% payload variation, reduces trajectory tracking error by 37%, and significantly outperforms conventional PID and DDPG baselines. This work establishes a transferable, robust control paradigm for autonomous AM in dynamically loaded operational scenarios.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Multiagent Systems: Multiagent Learning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

This paper investigates the application of Deep Reinforcement (DRL) Learning to address motion control challenges in drones for additive manufacturing (AM). Drone-based additive manufacturing promises flexible and autonomous material deposition in large-scale or hazardous environments. However, achieving robust real-time control of a multi-rotor aerial robot under varying payloads and potential disturbances remains challenging. Traditional controllers like PID often require frequent parameter re-tuning, limiting their applicability in dynamic scenarios. We propose a DRL framework that learns adaptable control policies for multi-rotor drones performing waypoint navigation in AM tasks. We compare Deep Deterministic Policy Gradient (DDPG) and Twin Delayed Deep Deterministic Policy Gradient (TD3) within a curriculum learning scheme designed to handle increasing complexity. Our experiments show TD3 consistently balances training stability, accuracy, and success, particularly when mass variability is introduced. These findings provide a scalable path toward robust, autonomous drone control in additive manufacturing.

Problem

• Deep Reinforcement Learning for drone control
• Adaptive motion control in dynamic environments
• Robust real-time control for additive manufacturing drones

Innovation

• Deep Reinforcement Learning for drones
• Curriculum learning for complexity handling
• TD3 for stable autonomous control