Publication
Drone instrumentation for detection and monitoring of atmospheric pollution by microplastics
| Summary: | Aiming at the instrumentation of drones so that they could be used to detect microplastics in the atmosphere, this work consisted of the development of several methods that initially presented properties considered relevant for drone applications. Of the proposed methodologies, only two were subjected to experimental validation under controlled laboratory conditions. The remaining approaches were not considered for two reasons: first, because after theorization and simulations, it was concluded that they were of high complexity and instability for application in a drone; and second, because no sufficient evidence was found that would justify a more in-depth study. Of the two remaining methods, namely acoustic detection and optical detection, only the latter advanced to drone application, with the former falling short of initial expectations. Although it showed qualities for impact detection, laboratory tests did not reveal evidence that would support its use in drones. The optical system took advantage of phenomena such as reflection and transmission to identify particles. Initially, it was observed that this method was capable of distinguishing plastics from metals using statistical methods such as principal component analysis and clustering. The possibility of using machine learning to perform the distinction automatically was also observed. Following these tests, its capacity to detect particles was studied, and in the laboratory a minimum experimental limit of 200 μm was observed. However, the method showed strong indications of the ability to detect particles below this size. Initially, the method demonstrated the expected stability for the intended application context. Furthermore, after flight tests with the drone, it was verified that the prototype was able to detect particles during movement, having succeeded in detecting particles as small as 68 μm. This work demonstrated that the system composed of LEDs and photodetectors showed promising results that justify further development and optimization of this method with a view toward commercial-level applications. |
|---|---|
| Main Authors: | Pereira, Guilherme Ricardo Cardoso |
| Subject: | Environmental monitorization Microplastics Drone Detection Instrumentation |
| Year: | 2025 |
| Country: | Portugal |
| Document type: | master thesis |
| Access type: | embargoed access |
| Associated institution: | Universidade de Aveiro |
| Language: | English |
| Origin: | RIA - Repositório Institucional da Universidade de Aveiro |
| Summary: | Aiming at the instrumentation of drones so that they could be used to detect microplastics in the atmosphere, this work consisted of the development of several methods that initially presented properties considered relevant for drone applications. Of the proposed methodologies, only two were subjected to experimental validation under controlled laboratory conditions. The remaining approaches were not considered for two reasons: first, because after theorization and simulations, it was concluded that they were of high complexity and instability for application in a drone; and second, because no sufficient evidence was found that would justify a more in-depth study. Of the two remaining methods, namely acoustic detection and optical detection, only the latter advanced to drone application, with the former falling short of initial expectations. Although it showed qualities for impact detection, laboratory tests did not reveal evidence that would support its use in drones. The optical system took advantage of phenomena such as reflection and transmission to identify particles. Initially, it was observed that this method was capable of distinguishing plastics from metals using statistical methods such as principal component analysis and clustering. The possibility of using machine learning to perform the distinction automatically was also observed. Following these tests, its capacity to detect particles was studied, and in the laboratory a minimum experimental limit of 200 μm was observed. However, the method showed strong indications of the ability to detect particles below this size. Initially, the method demonstrated the expected stability for the intended application context. Furthermore, after flight tests with the drone, it was verified that the prototype was able to detect particles during movement, having succeeded in detecting particles as small as 68 μm. This work demonstrated that the system composed of LEDs and photodetectors showed promising results that justify further development and optimization of this method with a view toward commercial-level applications. |
|---|
Funded activities
No funded projects found.