# UAV Trajectory Optimization via Improved Noisy Deep Q-Network

arXiv: https://arxiv.org/abs/2602.05644 · HTML: https://mlsanity.com/paper/2602.05644/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot navigation and UAV control, one of 15 directions in Deep reinforcement learning methods.
- Builds on 6 of its 18 references that are in the map.
- First posted on arXiv 5 Feb 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot navigation and UAV control](https://mlsanity.com/deeprl/rl-for-robot-navigation-and-uav-control/)

## Builds on (most cited first, 6 of 6)

- [Noisy Networks for Exploration](https://mlsanity.com/paper/1706.10295/) (2017)
- [Autonomous Unmanned Aerial Vehicle Navigation using Reinforcement Learning: A Systematic Review](https://mlsanity.com/paper/2208.12328/) (2022)
- [Autonomous quadrotor obstacle avoidance based on dueling double deep recurrent Q-learning with monocular vision](https://mlsanity.com/paper/2002.03510/) (2020)
- [Asynchronous Curriculum Experience Replay: A Deep Reinforcement Learning Approach for UAV Autonomous Motion Control in Unknown Dynamic Environments](https://mlsanity.com/paper/2207.01251/) (2022)
- [Vision-Based Deep Reinforcement Learning of UAV Autonomous Navigation Using Privileged Information](https://mlsanity.com/paper/2412.06313/) (2024)
- [Towards Robust Multi-UAV Collaboration: MARL with Noise-Resilient Communication and Attention Mechanisms](https://mlsanity.com/paper/2503.02913/) (2025)

## Abstract (arXiv)

This paper proposes an Improved Noisy Deep Q-Network (Noisy DQN) to enhance the exploration and stability of Unmanned Aerial Vehicle (UAV) when applying deep reinforcement learning in simulated environments. This method enhances the exploration ability by combining the residual NoisyLinear layer with an adaptive noise scheduling mechanism, while improving training stability through smooth loss and soft target network updates. Experiments show that the proposed model achieves faster convergence and up to +40 higher rewards compared to standard DQN and quickly reach to the minimum number of steps required for the task 28 in the 15 * 15 grid navigation environment set up. The results show that our comprehensive improvements to the network structure of NoisyNet, exploration control, and training stability contribute to enhancing the efficiency and reliability of deep Q-learning.