Development of Automatic Water Sampler for Aerial Drones

Sampling water in Arctic environments is a long, cumbersome task that takes a lot of equipment and expertise to carry out on a regular basis. Currently, data acquisition in Arctic regions is carried out via two methods;

1. Satellite information regarding water properties.
2. Field work for real-time sampling and investigation of sites.

Each of the said tasks are difficult for operation and it is imperative that a solution be created around the problem of water sampling. Considering the nature of the task, to improve how researchers’ work can be made more efficient, AF AB initiated the project, “Drones in Arctic Environments.” Drones are a priority approach for this task given their ability to easily maneuver through different areas and effectively collect samples on the ground before they leave back for their dispatch point.

Exploring and researching this problem, a thesis titled, “Drones in Arctic Environments: Development of Automatic Water Sampler for Aerial Drones“, was published at KTH Royal Institute Of Technology School Of Industrial Engineering And Management. The purpose of the thesis was to study the current water sampling process of the with a mechatronic perspective to investigate how to improve the process. This was be done by developing an automatic water sampling equipment based on their needs, to be used with a drone for remote water sampling. In the thesis, the researchers explore the question as they put it, “can a water sampler be designed to a given drone to automate the water sampling process?”

The thesis is published as a feasibility report as most of the work has been done with rapid and frequent prototyping of drones programmed for water sampling. The publication mentions the benefits of using this approach:

• Quick analysis of physical working of water sampling systems in real time.
• Physical demonstration assessing the viability of the proposed solutions in practice.

On the other hand, the author mentions the flaws with using the said approach of study,

• The experiments conducted with developed prototypes are dependent on the assumptions made by the author.
• Quantification of design parameters and simulation results have been ignored.

A small quad-copter was used for the field tests to test its working as the drone sampled water from the Arctic.

Autonomous drone with the three water sample containers attached on
the bottom of the drone.

It was observed that upon collecting water, the weight carried by the drone and therefore the thrust provided by the rotors increased, reducing the battery time of the drone to a mere 7.5 minutes; which severely limits the distance the drone can travel for collecting water samples as well as the number of samples it can collect in one battery cycle. This problem was solved with using octa-rotor drones that although cost much more than quad-copters, do offer better battery performance.

At the same time, another problem was observed, due to the low density of water and its fluid properties, its center of gravity shifted rapidly during the flight. This caused serious stability errors in the drone, continuously misaligning its optimal roll angle and pitch. After a number of iterations with prototypes, the said problem still persisted.

Citation: [1]S. Olsson, ‘Drones in Arctic Environments: Development of Automatic Water Sampler for Aerial Drones’, Dissertation, 2018. http://www.diva-portal.org/smash/record.jsf?pid=diva2%3A1257208&dswid=-3637