Abstract
Autonomous motion planning (AMP) of unmanned aerial vehicles (UAVs) is aimed at enabling a UAV to safely fly to the target without human intervention. Recently, several emerging deep reinforcement learning (DRL) methods have been employed to address the AMP problem in some simplified environments, and these methods have yielded good results. This paper proposes a multiple experience pools (MEPs) framework leveraging human expert experiences for DRL to speed up the learning process. Based on the deep deterministic policy gradient (DDPG) algorithm, a MEP–DDPG algorithm was designed using model predictive control and simulated annealing to generate expert experiences. On applying this algorithm to a complex unknown simulation environment constructed based on the parameters of the real UAV, the training experiment results showed that the novel DRL algorithm resulted in a performance improvement exceeding 20% as compared with the state-of-the-art DDPG. The results of the experimental testing indicate that UAVs trained using MEP–DDPG can stably complete a variety of tasks in complex, unknown environments.
Highlights
The number of applications for unmanned aerial vehicles (UAVs) is widely increasing in the civil arena such as surveillance [1,2], delivery of goods [3,4], power line inspection [5,6], and mapping [7,8].In the majority of these applications, it is necessary for Unmanned aerial aerial vehicle vehicle (UAV) to plan their motion such that they can perform their tasks while avoiding threats in complex, unknown environments.Many traditional path planning algorithms, such as A* algorithm, visibility graph algorithm, and free space algorithm, are used to solve the motion planning problem of UAV, but these methods can usually only achieve good results when the environment or map is known
In the majority of these applications, it is necessary for UAVs to plan their motion such that they can perform their tasks while avoiding threats in complex, unknown environments
Simultaneous localization and mapping (SLAM) maps the unknown environment according to position and sensor information of the UAV during its movement in the environment so as to implement the automatic motion planning of the UAV according to the drawn map
Summary
The number of applications for unmanned aerial vehicles (UAVs) is widely increasing in the civil arena such as surveillance [1,2], delivery of goods [3,4], power line inspection [5,6], and mapping [7,8]. Many studies used DRL to solve the autonomous motion planning (AMP) problem of UAV and achieved good results, but these studies still have some shortcomings: (1) the models of the UAV and environment can be more complex and realistic; (2) the convergence speed and convergence results of the algorithm can be improved. To address these problems, explorations and experiments were conducted in this study.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
