To Risk or Not to Risk: Learning with Risk Quantification for IoT Task Offloading in UAVs

TL;DR

This paper introduces a deep reinforcement learning method that uses risk quantification via CVaR to improve task offloading decisions in UAV-based IoT networks, effectively reducing critical deadline violations in smart farms.

Contribution

It presents a novel CVaR-based risk-aware reinforcement learning approach for UAV task offloading in IoT environments, enhancing safety and reliability.

Findings

CVaR-based method reduces deadline violations significantly.

Approach effectively avoids risky actions leading to irreversible damage.

Negligible increase in energy consumption compared to other methods.

Abstract

A deep reinforcement learning technique is presented for task offloading decision-making algorithms for a multi-access edge computing (MEC) assisted unmanned aerial vehicle (UAV) network in a smart farm Internet of Things (IoT) environment. The task offloading technique uses financial concepts such as cost functions and conditional variable at risk (CVaR) in order to quantify the damage that may be caused by each risky action. The approach was able to quantify potential risks to train the reinforcement learning agent to avoid risky behaviors that will lead to irreversible consequences for the farm. Such consequences include an undetected fire, pest infestation, or a UAV being unusable. The proposed CVaR-based technique was compared to other deep reinforcement learning techniques and two fixed rule-based techniques. The simulation results show that the CVaR-based risk quantifying method eliminated the most dangerous risk, which was exceeding the deadline for a fire detection task. As a result, it reduced the total number of deadline violations with a negligible increase in energy consumption.