Distributed Reinforcement Learning for Flexible and Efficient UAV Swarm Control

Venturini, Federico; Mason, Federico; Pase, Francesco; Chiariotti, Federico; Testolin, Alberto; Zanella, Andréa; Zorzi, Michele

doi:10.1109/tccn.2021.3063170

Cited by 32 publications

(13 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The infrastructural damages are usually assessed using the fragility curves predictions that are built on statistical and historical data. However, now, the utilization of Random Forest, One-shot learning, neural networks, and Support Vector Machine (SVM)-based AI methods can be used effectively in the detection of damages during a flood event [103]. Particularly, for the aged care facilities, the application of AI techniques will not only assist in damage detection but will also help in obtaining the forecasts of upcoming flood events and the provision of early warning for the safe evacuation of the aged care facilities.…”

Section: Flood Risk Management and Evacuation Strategies For The Aged Care Facilitiesmentioning

confidence: 99%

An AI/ML-Based Strategy for Disaster Response and Evacuation of Victims in Aged Care Facilities in the Hawkesbury-Nepean Valley: A Perspective

et al. 2022

View full text Add to dashboard Cite

The Hawkesbury-Nepean Valley, Australia’s longest coastal catchment, is spanned by a river system of more than 470 km, that runs from Goulburn to Broken Bay, covering a total area of over 2.2 million hectares. This region has remained prone to flood events, with considerable mortalities, economic impacts and infrastructural losses occurring quite regularly. The topography, naturally variable climatic conditions and the ‘bathtub’ effect in the region are responsible for the frequent flood events. In response, the Government at the national/federal, state and local level has focused on the design of efficient flood risk management strategies with appropriate evacuation plans for vulnerable communities from hospitals, schools, childcare and aged care facilities during a flood event. Despite these overarching plans, specialized response and evacuation plans for aged care facilities are critical to reducing the loss incurred by flood events in the region. This is the focus of this present paper, which reviews the history of flood events and responses to them, before examining the utilization of artificial intelligence (AI) techniques during flood events to overcome the flood risks. An early flood warning system, based on AI/Machine Learning (ML) strategy is being suggested for a timely decision, enhanced disaster prediction, assessment and response necessary to overcome the flood risks associated with aged care facilities within the Hawkesbury-Nepean region. A framework entailing AI/ML methods for identifying the safest route to the destination using UAV and path planning has been proposed for timely disaster response and evacuation of the residents of aged care facilities.

show abstract

Section: Flood Risk Management and Evacuation Strategies For The Aged Care Facilitiesmentioning

confidence: 99%

An AI/ML-Based Strategy for Disaster Response and Evacuation of Victims in Aged Care Facilities in the Hawkesbury-Nepean Valley: A Perspective

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Figure 1 plots the accumulated reward J(ω t ) v.s. communication and sample complexity 3 . Each curve includes 10 repeated experiments, and its upper and lower envelopes denote the 95% and 5% percentiles of the 10 repetitions, respectively.…”

Section: Methodsmentioning

confidence: 99%

“…Multi-agent reinforcement learning (MARL) has achieved great success in various application domains, including control [1][2][3], robotics [4], wireless sensor networks [5,6], intelligent systems [7], etc. In MARL, a set of fully decentralized agents interact with a dynamic environment following their own policies and collect local rewards, and their goal is to collaboratively learn the optimal joint policy that achieves the maximum expected accumulated reward.…”

Section: Introductionmentioning

confidence: 99%

Sample and Communication-Efficient Decentralized Actor-Critic Algorithms with Finite-Time Analysis

Chen

Zhou

Chen

et al. 2021

Preprint

View full text Add to dashboard Cite

Actor-critic (AC) algorithms have been widely adopted in decentralized multi-agent systems to learn the optimal joint control policy. However, existing decentralized AC algorithms either do not preserve the privacy of agents or are not sample and communication-efficient. In this work, we develop two decentralized AC and natural AC (NAC) algorithms that are private, and sample and communicationefficient. In both algorithms, agents share noisy information to preserve privacy and adopt mini-batch updates to improve sample and communication efficiency. Particularly for decentralized NAC, we develop a decentralized Markovian SGD algorithm with an adaptive mini-batch size to efficiently compute the natural policy gradient. Under Markovian sampling and linear function approximation, we prove the proposed decentralized AC and NAC algorithms achieve the stateof-the-art sample complexities O( −2 ln −1 ) and O( −3 ln −1 ), respectively, and the same small communication complexity O( −1 ln −1 ). Numerical experiments demonstrate that the proposed algorithms achieve lower sample and communication complexities than the existing decentralized AC algorithm.Preprint. Under review.

show abstract

“…In [24], a distributed reinforcement learning (RL) approach is proposed with an algorithmic framework that relies on the possibility of drones exchanging some information through communication channels to achieve context awareness and implicitly coordinate the actions of UAV swarms. In [25], an end-to-end collaborative multiagent reinforcement learning (MARL) scheme is presented that enables UAVs to make intelligent flight decisions for collaborative target tracking based on the past and current state of the target.…”

Section: Related Workmentioning

confidence: 99%

An Information-Entropy-Based Hierarchical Serialization Allocation Method for UAV Tracking in 6G Networks

Zhong

Yang

et al. 2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

Unmanned aerial vehicles (UAVs) play an important role in future 6G networks, which can be used to assist cellular networks in setting up temporary networks to provide communication services when network access demand is intense. It is critical to design a UAV tracking method with high efficiency and high precision under active sensor radiation control to build a reliable network of UAVs. Multisensor cooperative multitarget tracking for UAVs with high accuracy is an alternative solution to meet the performance requirements of 6G. In this paper, an information-entropy-based multisensor to multitarget allocation method under low-radiation control is proposed for UAV tracking. Specifically, the knowledge-map-based demand assessment of integrated track detection is first present; and then, fuzzy-decision-tree-based generation of sensor management plan is given; the next, the basic linear programming model for multitarget to multi-UAV allocation is utilized based on cooperative tracking under radiation control; finally, extensive experiments are conduct, and the experimental results show that the proposed method has a good performance.

show abstract

Distributed Reinforcement Learning for Flexible and Efficient UAV Swarm Control

Cited by 32 publications

References 27 publications

An AI/ML-Based Strategy for Disaster Response and Evacuation of Victims in Aged Care Facilities in the Hawkesbury-Nepean Valley: A Perspective

An AI/ML-Based Strategy for Disaster Response and Evacuation of Victims in Aged Care Facilities in the Hawkesbury-Nepean Valley: A Perspective

Sample and Communication-Efficient Decentralized Actor-Critic Algorithms with Finite-Time Analysis

An Information-Entropy-Based Hierarchical Serialization Allocation Method for UAV Tracking in 6G Networks

Contact Info

Product

Resources

About