Research on Autonomous Obstacle Avoidance and Target Tracking of UAV Based on Improved Dueling DQN Algorithm

Jiang, Weilai; Bao, Chong; Xu, Guoqiang; Wang, Yaonan

doi:10.1109/cac53003.2021.9728707

Cited by 4 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dueling DQN is widely considered as a signifcant improvement to conventional DQN. Diferent from the natural DQN, dueling DQN divides the Q-network into two parts, action advantage function with independent of state A(s t , a t ; ω, ξ) and state-value function V(s t ; ω, θ), which are calculated separately [20,21]. It is easy to fnd which action has better feedback by learning A(s t , a t ; ω, ξ).…”

Section: Problem Solutionmentioning

confidence: 99%

Design and Optimization in MEC‐Based Intelligent Rail System by Integration of Distributed Multi‐Hop Communication and Blockchain

Tian

et al. 2023

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Mobile edge computing technology has emerged as a novel computing paradigm that makes use of resources close to the devices of the smart rail system. Nevertheless, it is difficult to support data offloading to the stations directly from different trains due to the limited coverage of the stations equipped with MEC servers. Therefore, multi-hop ad hoc network is considered and introduced in this case. In this paper, an improved architecture is proposed for the MEC-based smart rail system by blockchain and multi-hop data communication. The requesting trains can offload the tasks to MEC servers by multi-hop transmission between trains, even when requesting trains are not covered by servers. Furthermore, we utilize the blockchain technology for the authenticity and anti-falsification of information during multi-hop transmission. Then, the offloading routing path and offloading strategy are co-optimized to minimize both delay and cost of the system. The proposed majorization problem is formulated as a Markov decision process (MDP) and solved by deep reinforcement learning (DRL). In comparison to other existing schemes, simulation results demonstrate that the proposed scheme can greatly improve system performance.

show abstract

Section: Problem Solutionmentioning

confidence: 99%

Design and Optimization in MEC‐Based Intelligent Rail System by Integration of Distributed Multi‐Hop Communication and Blockchain

Tian

et al. 2023

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…Compared to the classical reinforcement learning models, the A2C algorithm has a more powerful learning ability, since it includes two neural networks: Actor and Critic [30,31], and A2C supports synchronous parallel sampling training, on the one hand, to ensure diversity of data, on the other hand, to improve learning efficiency [32]. Moreover, compared to the Q-learning [33,34] or DQN [35,36] algorithm, it is more suitable for continuous space problems, which is applicable to the vehicle swarm control problem in this study [37,38]. To reflect the cooperative obstacle-avoidance behaviour of the automated vehicle swarm, the optimization target in the A2C algorithm not only incorporates the safety and efficiency of an individual vehicle but also considers the efficiency of the vehicle swarm.…”

Section: Introductionmentioning

confidence: 99%

Model for the cooperative obstacle‐avoidance of the automated vehicle swarm in a connected vehicles environment

Zhai

Yang

Chen

et al. 2023

IET Intelligent Trans Sys

View full text Add to dashboard Cite

The obstacle‐avoidance problem of automated vehicles is a hot topic in the community of autonomous driving. The majority of the existing studies focused on the obstacle‐avoidance of a single automated vehicle. The connected vehicles technology provides the possibility of controlling a vehicle swarm to avoid the obstacle cooperatively. Through cooperation, the vehicle swarm not only can avoid an obstacle safely but also can minimize traffic delays. Therefore, this paper proposes a cooperative obstacle‐avoidance model for the automated vehicle swarm driving on the freeway based on the A2C reinforcement learning. The proposed model considers the efficiencies of both the individual and swarm in the learning, and a cooperative lane‐changing execution model is proposed to ensure that the optimal decision made by the A2C algorithm can be performed by the vehicles. Furthermore, simulations are conducted to verify the proposed model. The results indicate that the proposed model can significantly improve the overall traffic efficiency compared with the existing models. In a congested state, when the proposed model is applied to control vehicles, an optimal control range can be found (i.e. 700 m here), and within this optimal range, the traffic efficiency increases with the increment of the number of the vehicles controlled by the proposed model.

show abstract

The application of path planning algorithm based on deep reinforcement learning for mobile robots

Tian

Lei

Huang

et al. 2022

2022 International Conference on Culture-Oriented Science and Technology (CoST)

View full text Add to dashboard Cite

Research on Autonomous Obstacle Avoidance and Target Tracking of UAV Based on Improved Dueling DQN Algorithm

Cited by 4 publications

References 11 publications

Design and Optimization in MEC‐Based Intelligent Rail System by Integration of Distributed Multi‐Hop Communication and Blockchain

Design and Optimization in MEC‐Based Intelligent Rail System by Integration of Distributed Multi‐Hop Communication and Blockchain

Model for the cooperative obstacle‐avoidance of the automated vehicle swarm in a connected vehicles environment

The application of path planning algorithm based on deep reinforcement learning for mobile robots

Contact Info

Product

Resources

About