Deep Reinforcement Learning for Vision-Based Navigation of UAVs in Avoiding Stationary and Mobile Obstacles

Review Automation & Control Systems

Autonomous Unmanned Aerial Vehicle navigation using Reinforcement Learning: A systematic review

Fadi AlMahamid et al.

Summary: There is a growing demand for using drones in various applications, and autonomous UAV navigation is commonly achieved using reinforcement learning algorithms. Understanding the navigation environment and algorithm limitations is crucial in selecting the right algorithms to solve navigation problems effectively. This study identifies the main UAV navigation tasks, discusses navigation frameworks and simulation software, and classifies RL algorithms based on their environment, characteristics, abilities, and applications.

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE (2022)

添加到收藏夹

Review Remote Sensing

Towards the Unmanned Aerial Vehicles (UAVs): A Comprehensive Review

Syed Agha Hassnain Mohsan et al.

Summary: This article discusses the importance and potential of drones in various application fields, as well as the challenges and security issues they face. It also presents future research directions and areas that need to be addressed.

DRONES (2022)

添加到收藏夹

Article Remote Sensing

UAV Obstacle Avoidance Algorithm to Navigate in Dynamic Building Environments

Enrique Aldao et al.

Summary: This work presents a real-time collision avoidance algorithm for autonomous navigation in building environments where fixed and moving obstacles are present for UAV. The algorithm considers the geometry of the environment and obstacles not registered in the original point cloud. It has been successfully tested in different scenarios, offering robust results in all avoidance maneuvers, and is computationally feasible for onboard implementation on UAVs.

DRONES (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Quadrotor Path Following and Reactive Obstacle Avoidance with Deep Reinforcement Learning

Bartomeu Rubi et al.

Summary: This paper proposes a deep reinforcement learning approach to solve the quadrotor path following and obstacle avoidance problem by using two agents. A novel structure is introduced to interpret training outcomes, achieve faster training, and ensure safety in real quadrotor training. The simulation results validate the effectiveness of the proposed method.

JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Vision Based Drone Obstacle Avoidance by Deep Reinforcement Learning

Zhihan Xue et al.

Summary: Research on autonomous obstacle avoidance of drones has attracted much attention, with an increasing number of researchers exploring machine learning for training. While supervised learning and reinforcement learning have limitations, the soft-actor-critic algorithm shows promise in training drones for autonomous obstacle avoidance in continuous action space using image data.

AI (2021)

添加到收藏夹

Proceedings Paper Computer Science, Information Systems

Computer vision based obstacle detection and target tracking for autonomous vehicles

Ruoyu Fang et al.

Summary: This paper introduces a new scheme for target tracking and obstacle detection using deep learning neural networks, which can be deployed on an autonomous vehicle to achieve stable and precise tracking by adjusting the vehicle's steering and movement according to the PID algorithm.

2020 2ND INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE COMMUNICATION AND NETWORK SECURITY (CSCNS2020) (2021)

添加到收藏夹

Article Chemistry, Multidisciplinary