PARRoT: Predictive Ad-hoc Routing Fueled by Reinforcement Learning and Trajectory Knowledge

TL;DR

PARRoT is a novel machine learning-based routing protocol for UAV swarms that leverages trajectory knowledge and reinforcement learning to improve robustness and reduce latency in highly mobile ad-hoc networks.

Contribution

This paper introduces PARRoT, a new routing protocol that integrates mobility control information and reinforcement learning for better decision-making in dynamic UAV networks.

Findings

PARRoT significantly reduces end-to-end latency.

PARRoT demonstrates higher robustness compared to existing protocols.

Simulation results validate the effectiveness of PARRoT in mobile UAV networks.

Abstract

Swarms of collaborating Unmanned Aerial Vehicles (UAVs) that utilize ad-hoc networking technologies for coordinating their actions offer the potential to catalyze emerging research fields such as autonomous exploration of disaster areas, demanddriven network provisioning, and near field packet delivery in Intelligent Transportation Systems (ITSs). As these mobile robotic networks are characterized by high grades of relative mobility, existing routing protocols often fail to adopt their decision making to the implied network topology dynamics. For addressing these challenges, we present Predictive Ad-hoc Routing fueled by Reinforcement learning and Trajectory knowledge (PARRoT) as a novel machine learning-enabled routing protocol which exploits mobility control information for integrating knowledge about the future motion of the mobile agents into the routing process. The performance of the proposed routing approach is evaluated using comprehensive network simulation. In comparison to established routing protocols, PARRoT achieves a massively higher robustness and a significantly lower end-to-end latency.