Neighborhood global learning based flower pollination algorithm and its application to unmanned aerial vehicle path planning

Yang, Chen; Pi, Dechang; Xu, Yaoxian

doi:10.1016/j.eswa.2020.114505

Cited by 38 publications

(26 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to detecting the diversity of populations, we employed a mutation operation on individuals based on the probability of the solution. The strategy of randomly assigning positions to individuals in the search space has been used in the literature [39] and [41], but that strategy was unfocused. Gaussian mutation [42] is an effective way to improve the quality of individuals.…”

Section: B Algorithm Design and Implementationmentioning

confidence: 99%

HNIO: A Hybrid Nature-Inspired Optimization Algorithm for Energy Minimization in UAV-Assisted Mobile Edge Computing

Chen

Yang

et al. 2022

IEEE Trans. Netw. Serv. Manage.

Self Cite

View full text Add to dashboard Cite

Section: B Algorithm Design and Implementationmentioning

confidence: 99%

HNIO: A Hybrid Nature-Inspired Optimization Algorithm for Energy Minimization in UAV-Assisted Mobile Edge Computing

Chen

Yang

et al. 2022

IEEE Trans. Netw. Serv. Manage.

Self Cite

View full text Add to dashboard Cite

“…An appealing property of policy optimization methods is the local-search nature, which lends itself to an efficient implementation as a search over the whole MDP is avoided. However, this property also makes it difficult to obtain global optimality guarantees for these algorithms and a large portion of the literature postulates strong and often unrealistic assumptions to ensure global exploration (see e.g., Abbasi-Yadkori et al, 2019;Agarwal et al, 2020b;Neu and Olkhovskaya, 2021;Wei et al, 2021). Recently, the need for extra assumptions has been overcome by adding exploration bonuses to the update (Cai et al, 2020;Shani et al, 2020;Agarwal et al, 2020a;Zanette et al, 2020;Luo et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

“…However, this property also makes it difficult to obtain global optimality guarantees for these algorithms and a large portion of the literature postulates strong and often unrealistic assumptions to ensure global exploration (see e.g., Abbasi-Yadkori et al, 2019;Agarwal et al, 2020b;Neu and Olkhovskaya, 2021;Wei et al, 2021). Recently, the need for extra assumptions has been overcome by adding exploration bonuses to the update (Cai et al, 2020;Shani et al, 2020;Agarwal et al, 2020a;Zanette et al, 2020;Luo et al, 2021). These works demonstrate an additional robustness property of policy optimization, which is able to handle adversarial losses or some level of corruption.…”

Section: Introductionmentioning

confidence: 99%

Best of Both Worlds Policy Optimization

Dann¹,

Wei²,

Zimmert³

2023

Preprint

View full text Add to dashboard Cite

Policy optimization methods are popular reinforcement learning algorithms in practice. Recent works have built theoretical foundation for them by proving √ T regret bounds even when the losses are adversarial. Such bounds are tight in the worst case but often overly pessimistic. In this work, we show that in tabular Markov decision processes (MDPs), by properly designing the regularizer, the exploration bonus and the learning rates, one can achieve a more favorable polylog(T ) regret when the losses are stochastic, without sacrificing the worst-case guarantee in the adversarial regime. To our knowledge, this is also the first time a gap-dependent polylog(T ) regret bound is shown for policy optimization. Specifically, we achieve this by leveraging a Tsallis entropy or a Shannon entropy regularizer in the policy update. Then we show that under known transitions, we can further obtain a first-order regret bound in the adversarial regime by leveraging the log barrier regularizer.

show abstract

“…An improved fruit fly optimization algorithm is introduced in [22] to address the problem of path planning of multiple UAVs in 3D complicated environments with online changing tasks. In [23], a flower pollination algorithm based on neighborhood global learning is employed to complete route planning of a UAV. [24] offers an evolutionary algorithm based on a novel separate evolution strategy to plan an optimized path for a single UAV.…”

Section: Introductionmentioning

confidence: 99%

Path Planning and Control of a Quadrotor UAV Based on an Improved APF Using Parallel Search

Huang

Qin

et al. 2021

International Journal of Aerospace Engineering

View full text Add to dashboard Cite

Control and path planning are two essential and challenging issues in quadrotor unmanned aerial vehicle (UAV). In this paper, an approach for moving around the nearest obstacle is integrated into an artificial potential field (APF) to avoid the trap of local minimum of APF. The advantage of this approach is that it can help the UAV successfully escape from the local minimum without collision with any obstacles. Moreover, the UAV may encounter the problem of unreachable target when there are too many obstacles near its target. To address the problem, a parallel search algorithm is proposed, which requires UAV to simultaneously detect obstacles between current point and target point when it moves around the nearest obstacle to approach the target. Then, to achieve tracking of the planned path, the desired attitude states are calculated. Considering the external disturbance acting on the quadrotor, a nonlinear disturbance observer (NDO) is developed to guarantee observation error to exponentially converge to zero. Furthermore, a backstepping controller synthesized with the NDO is designed to eliminate tracking errors of attitude. Finally, comparative simulations are carried out to illustrate the effectiveness of the proposed path planning algorithm and controller.

show abstract

Neighborhood global learning based flower pollination algorithm and its application to unmanned aerial vehicle path planning

Cited by 38 publications

References 33 publications

HNIO: A Hybrid Nature-Inspired Optimization Algorithm for Energy Minimization in UAV-Assisted Mobile Edge Computing

HNIO: A Hybrid Nature-Inspired Optimization Algorithm for Energy Minimization in UAV-Assisted Mobile Edge Computing

Best of Both Worlds Policy Optimization

Path Planning and Control of a Quadrotor UAV Based on an Improved APF Using Parallel Search

Contact Info

Product

Resources

About