Three-Dimensional Surface Area Computation and Coverage Optimization in Directional Sensor Networks: With Applications for Vertical Farming
Translated title
Three-Dimensional Surface Area Computation and Coverage Optimization in Directional Sensor Networks
Authors
Madsen, Mads Bjørn ; Aver, Lau Christian ; Nygaard, Frederick Alexander Bue
Term
4. term
Education
Publication year
2024
Submitted on
2024-06-03
Pages
33
Abstract
As more tasks are automated, monitoring becomes essential for consistent product quality. Sensors are central to this, including in vertical farms where crops are grown indoors in stacked layers. This study examines directional sensors—devices that measure in a specific direction—in such a setting. We apply a Particle Swarm Optimization algorithm, a nature-inspired search method that mimics group movement, to place a sensor within a confined area so that the monitored surface area is as large as possible. Starting from a random placement, the algorithm improves coverage and finds near-optimal positions. We also find that fixed (static) sensors are significantly outperformed by sensors that can rotate and scan different directions.
I takt med at flere opgaver automatiseres, bliver overvågning afgørende for ensartet produktkvalitet. Sensorer er centrale, også i vertikalt landbrug, hvor afgrøder dyrkes indendørs i stablede lag. Dette studie undersøger retningsbestemte sensorer, altså sensorer der måler i en bestemt retning, i denne sammenhæng. Vi anvender en Particle Swarm Optimization-algoritme, en naturinspireret søgemetode der efterligner flokadfærd, til at placere en sensor i et afgrænset område, så den overvågede flade bliver størst mulig. Ud fra en tilfældig startplacering forbedrer algoritmen dækningsarealet og finder næsten optimale positioner. Vi ser også, at faste (statiske) sensorer bliver markant overgået af sensorer, der kan rotere og scanne i forskellige retninger.
[This apstract has been rewritten with the help of AI based on the project's original abstract]