Grantee Research Project Results
Final Report: Mapping Air Quality with Kite-Based Sensors
EPA Grant Number: SU839465Title: Mapping Air Quality with Kite-Based Sensors
Investigators: Cairncross, Richard A. , DeCarlo, Peter , Terranova, Brandon , Vanderkluysen, Loyc , Lipscomb, Myles , Olega, Darius , Efymow, Jesse , O’Dwye, Shannon , Lim, Jesse , Kloiber, Anna , Bhagwat, Atharva , Omo-Lamai, Darrell , Hudson, Kira , Reina, Nicolas , Clark, Rasheem , Patel, Mikin , Lo, James
Institution: Drexel University
EPA Project Officer: Page, Angela
Phase: I
Project Period: January 1, 2019 through December 31, 2019
Project Amount: $15,000
RFA: P3 Awards: A National Student Design Competition Focusing on People, Prosperity and the Planet (2018) RFA Text | Recipients Lists
Research Category: P3 Awards , P3 Challenge Area - Air Quality
Objective:
The overall objective of this Phase I EPA P3 project is design of Kite-based Environmental Monitoring and Mapping Systems (KEMMS) to enable three-dimensional mapping of environmental air quality metrics. KEMMS merges the low-technology aspects of kites with high-technology aspects of drones to provide a means of environmental monitoring that is user-friendly, low-energy, educational, and suitable for citizen-science. This Phase I P3 project enlisted twelve undergraduate students from varying disciplines to develop and evaluate multiple aspects of KEMMS. The design and research activities focused on four general areas:
- Evaluating the performance of various kites as vehicles for environmental sensing
- Implementing sensor platforms for datalogging of GPS and ambient weather conditions as prototypes for environmental monitoring
- Testing several options for obtaining high-quality photo and video imaging during flights
- Developing a novel air-sampling system using suspended tubing to deliver air to heavy, expensive air testing equipment on the ground
Field testing the components of KEMMS occurred approximately monthly at several locations in the Philadelphia metropolitan area. The variability of wind conditions and turbulence produced by hilly topography, tall buildings, and trees posed a significant challenge to obtaining reproducible performance of the KEMMS components. The best results are obtained when the kite type and kite size are selected to match the current weather conditions. Generally delta-Conyne and Rokkaku (hexagonal) kites are reliable, steady lifters in moderate winds, and flexible parafoils are effective over a wide wind range when the winds are steady (not gusty).
The proposed plan to implement motor-based flight control on stunt kites met with limited success. Stunt kites are designed to be highly responsive to the pilot’s control motions, which proved difficult to reproduce using small gear motors and stepper motors. Wind variability and jerky motion of the kite often led to rapid changes in line tension that overloaded the motors. Continued development of motor systems and implementation with more inherently stable kite designs are planned for future work. Three-dimensional movement of KEMMS is currently possible through moving the anchor point horizontally (e.g. the pilot walking) and by winding/unwinding the kite tether line or use of an Evan’s loop for vertical motion.
Measurement of weather data, position (altitude and GPS), and air quality were achieved using custom-built, microprocessor-based (Arduino) dataloggers and commercially available devices (Kestrel 5500 weather meter and PocketLab Air). The sensors are typically more stable when mounted from the kite line rather than being attached directly to the kite. This enables launching the kite first and later adding or lifting a sensor platform. Three methods were used: (1) stabilizing the sensor platform using a “Picavet” suspension system attached to two points on the kite line, (2) stabilizing the sensor platform aerodynamically using an “Aeropod” supplied by the NASA collaborators, and (3) attaching a pulley to the kite line enabling an “Evan’s” loop to quickly raise the sensor platform to any height (below the pulley). Measured GPS locations were mapped on Google Earth showing the paths traversed by the kite and sensors during field testing. Measurements with a weather meter enabled observing dependence of both wind speed and temperature on altitude. Measured air quality metrics included particulate matter (PM 10 and PM 2.5) concentration, ozone concentration, and carbon dioxide concentration; more detailed air quality measurements and other metrics will be considered in future projects.
There is an active community of kite hobbyists who are promotors of Kite Aerial Photography (KAP). A crucial aspect of an effective KAP rig is minimizing the movement of the camera to stabilize images even when the kite is constantly moving around in small or large oscillations. KAP can produce stunning images of the local environment and matching images and video to data collection could be an effective way to make more impactful statements about the data. In addition, Structure from Motion (SfM) techniques can convert a series of still images or video into rendered three-dimensional structures. In this project KAP and SfM were explored for obtaining visual records of field tests, for diagnostics of the sensor system, and for environmental evaluation. A consumer-grade camera drone was purchased to compare image quality and understand better the benefits and limitations of drones in comparison to KEMMS.
This Phase I project included the initial stages of development of a novel KEMMS air sampling system. This system enables using a kite to air-lift sampling tubing. Then a vacuum pump at ground-level draws air through the tubing to air quality equipment on the ground. To our knowledge, such an air sampling system has not been implemented previously using kites, drones, or balloons. The primary advantage of this air sampling system is that it enables using heavy, expensive, and highly accurate monitoring equipment on the ground. The primary disadvantage is the weight of the tubing and suction required to achieve sufficient flowrates over longer distances. In Phase I, 100 feet of ¼” Teflon tubing was tested to show that greater than 15 liters per minute flowrate can be achieved with 4 PSI of vacuum. This same tubing was lifted using an aerodynamically stabilized mount and an “Evan’s” loop. An Evan’s loop uses a pulley mounted on the kite line and a lifting line passing through the pulley to raise and lower the sampling tubing while the kite remains at approximately the same altitude. In several tests, the 100’ sampling tube was suspended full-length from the Evan’s loop for over an hour. Scaling up to 200-300’ sample lines should be straightforward in Phase II. Field testing in summer 2019 using a vacuum pump, sampling tubing, and a Picarro cavity ringdown spectrometer enabled measuring composition of the sampled air and demonstrated of technical feasibility of this system.
Summary/Accomplishments (Outputs/Outcomes):
Phase I has focused on field testing initial prototypes of the various KEMMS sub-systems. This research enables more accurate specifications for the integrated system in future projects.
Kite performance: more than ten different kites were evaluated as lifting kites during field testing in Phase I. Smaller kites such as delta-Conyne, Rokkaku and parafoil kites are effective for light sensors and cameras with sufficient wind. A Cody kite was constructed with pockets for inserting datalogging sensors directly onto the kite. Larger parafoil kites, such as a 60 ft2 parafoil that we used on multiple occasions, are required for the heavier air sampling system or for low-wind (less than 10 MPH) testing. Stunt kites proved to be difficult to control using simple motor systems.
To evaluated the aerodynamics and forces involved in kite systems, a physical model consisting of pulleys weights and strings was built in the lab to replicate the various forces that act on a kite and a mathematical kite pitch (varying tilt of the nose of the kite) model was implemented in Excel; these models help students improve their understanding of force balances and relationships between dominant parameters: angle of attack, lift and drag coefficients, and geometry. In the field delta-Conyne kites were modified with tails attached at varying locations to observe how asymmetry of aerodynamic forces affect kite flight. A kite with clear fabric and “tufts” attached to the back surface enables visualizing airflow across wings; in most cases separated flow was observed on the back surface of the kite. Two modified, steerable parafoils were obtained from collaborators; a radio control transmitter operated motors that could open and close vents on the upper surface of the kites. Field observations showed that these kites achieved limited maneuverability during opening and closing of the vents.
Environmental Sensing: Several stabilized sensing platforms were tested: a NASA aerodynamically-stabilized Aeropod, a Picavet stabilized by multiple tether points, and an Evans loop air sampler. Initial environmental measurements focused on GPS and weather with lightweight, inexpensive sensors lifted by the kite. Field data enabled plotting the kite locations on GoogleEarth and comparing GoogleEarth images to camera images in the field. The weather sensor enabled observing variations in wind speed and temperature with altitude. Air quality metric variation with altitude is uncertain based on limited field tests completed to date but the field tests have shown the ability to synchronize GPS, weather, and air quality measurements with lightweight sensors.
Imaging: A “Picavet” camera mount was built to stabilize images from aerial cameras. A consumer-grade drone was purchased for comparison and development of image processing. Images will be used for documenting field conditions and for Structure from Motion analysis. In the future projects, wireless and gimballed cameras will be used for improved flexibility in imaging.
Air sampling: A prototype for the air sampling system built during Phase I included 100 ft of Teflon tubing, a pressure gauge, a flow meter, a vacuum pump and aerodynamically-stabilized support on an “Evan’s” loop lifting system. Flowrates were measured at 15 L/min and ~3.9 psi. Field tests with a Picarro air quality monitors at the end of summer 2019 demonstrated technical feasibility of this system.
Accurate data pertaining to the spatial variability of air quality is necessary for good decision-making about methods for mitigating air pollution and methods for improving air quality. The proposed KEMMS platform provides a flexible method for measuring the spatial distribution of air quality metrics that could be deployed downwind of potential point sources of emissions such power plants and waste treatment plants, distributed sources of emissions such as municipalities and agricultural operations, or natural phenomena such as forest fires. The KEMMS system is envisioned as a low-cost user-friendly technique that could be used by private citizens, municipalities, companies, or organizations to obtain a more accurate estimate of local air quality issues.
Conclusions:
Monitoring the concentration of environmental pollutants is critical for effective decision-making about how to improve air quality. The use of Unmanned Aerial Vehicles (UAV) such as drones is attractive to provide detailed data about the spatial variation of air quality metrics; however, UAVs have flight times limited by battery life, public acceptance of UAVs is challenging, and there are increasingly stringent restrictions on the safe operating zones for UAVs. This project explored an alternative kite-based system for aerial monitoring of air quality. Kites have the potential to be lower cost than UAVs, require less energy to operate, and may have operational advantages such as flying at higher wind speeds and in areas inaccessible to UAVs. The Phase I field tests demonstrate the feasibility of using kites to lift multiple types of sensors for data logging and imaging combined with GPS. The novel air sampling system has been successfully demonstrated in the laboratory and in field conditions.
Journal Articles:
No journal articles submitted with this report: View all 3 publications for this projectSupplemental Keywords:
Methane, natural gas, carbon monoxide, nitrous oxides, drones, asthma, smog, VOC, volatile organic carbon, volcanos, smokeThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.