Reinforcement Learning for Connected Autonomous Vehicle Localization via UAVs

Adaptive Stochastic ADMM for Decentralized Reinforcement Learning in Edge Industrial IoT

“…We modify the environment in [50] to a decentralized RL setting where 𝑁 agents (UAVs) aim to work together to reach a specific target. Each agent can choose a set of four actions {north, south, west, east} as shown in Fig.…”

“…We assume the scenario contains only light-of-sight components. The estimated position of agents can be obtained as in [50]. The reward function is defined as:…”

Real-Time Learning for THZ Radar Mapping and UAV Control

“…Indeed, time is a key aspect for UAV networks because of their limited energy autonomy [5][6][7] and, thus, it should be properly accounted for when designing the UAV control for time-critical applications (e.g., search-and-rescue). In [5], an information-seeking algorithm is developed for extraterrestrial exploration and return-to-base application, whereas in [8,9] a similar problem is solved using RL for source localization. Algorithms for UAVs formation, navigation and self-localization have been proposed in [10][11][12][13][14], and RL for enhancing communications has been studied in [15][16][17][18].…”

Multi-Agent Q-Learning in UAV Networks for Target Detection and Indoor Mapping

“…For example, UAVs have played a central role in emergency situations in hazardous environments, for post natural disasters, or for search-and-rescue operations. In such events, UAVs have been used as a temporary network infrastructure for localization, communications, and for delivering items [1]- [3].…”

“…In this sense, machine learning (ML) can help in acquiring a knowledge of the model through experience. To that end, we adopt reinforcement learning (RL), which is based on the "trial-and-error" philosophy that allows to choose actions in order to maximize the sum of the discounted rewards over the future [3], [6]- [8]. In such settings, UAV navigation is driven by the balance between "exploration" and "exploitation".…”