Gauging hazards in the air

Lincoln Laboratory staff help students design sensors for measuring air quality

A new focus for the MIT Civil and Environmental Engineering (CEE) Department provided an opportunity for Lincoln Laboratory technical staff members to share their expertise in sensor systems with undergraduates confronting their collaborative senior project. In the spring 2014 course Sensor Networks and the Smart Campus (officially Senior Civil and Environmental Design, Course 1.013), students were charged with developing a system for monitoring air quality; they designed and built a network of sensors to measure levels of gases and particulates around the MIT campus. Helping the CEE students, who are accustomed to designing structures, navigate the complexities of developing sensors that rely on electronics and embedded computing were Lincoln Laboratory engineers who volunteered as mentors for this course held at the MIT Lincoln Laboratory Beaver Works Center.

Mentors for the CEE classSeen at the final presentation of the CEE capstone project were Lincoln Laboratory mentors for the spring course, left to right, Theo Tsiligkaridis, Danelle Shah, Loren Wood, and Jen King Jao. Absent from the photo is Ken Mawhinney.

"We were one of the first classes taught in the Beaver Works space," says Associate Professor Colette Heald, who directed the class along with Associate Professor Jesse Kroll and lecturer Eben Cross. "Through our early discussion with Bob Shin [director of the Beaver Works Center and head of Lincoln Laboratory’s Intelligence, Surveillance, and Reconnaissance (ISR) and Tactical Systems Division] and others at Lincoln Laboratory, we learned of the mentoring model the Lab had employed in other capstone classes at MIT."

The CEE department's objective of preparing engineers who understand the impact of environmental issues on designs for sustainable infrastructure led to this first in a series of senior-level courses aimed at assessing the environment through multimodal networked sensors, or "pervasive sensing." The assigned topic for the 2014 design course required of all senior CEE majors was air quality. Eight students in the fall term's elective pre-capstone course, Sensor Network Design, came up with the prototype sensor node that would be implemented into a network during the spring course. Loren Wood and Jesse Linnell of the Homeland Protection and Air Traffic Control Division worked with these eight students on developing a model that could be inexpensively fabricated with commercially available components, such as a laser particle sensor and a microprocessor.

Photo of student assembling a sensor nodeIn the Beaver Works Center, the CEE students assembled the sensors; here students are working on the wiring of the nodes. Photo credit: Eben Cross.

In the spring term, the 20 CEE seniors built 24 replicas of the prototype, and these sensor nodes were deployed to both indoor and outdoor locations around campus to take calibrated measurements of ozone, carbon monoxide, nitric oxide, nitrogen dioxide, and particulates such as dust and pollen. Their Lincoln Laboratory mentors met with them weekly. "We helped with the overall system design and the development of the nodes,”" says Wood, who along with Danelle Shah, Kenneth Mawhinney, and Theo Tsiligkaridis of the ISR and Tactical Systems Division, and Jen King Jao of the Air and Missile Defense Technology Division, assisted at the spring course.

"There were a few specific areas—electronics, signal processing, data handling, and network architecture—where one-on-one discussions with the Lincoln Lab mentors led to better practice and understanding by the students," says Cross, who added that Suzanne Vettese, a technician in the Laboratory's Fabrication Engineering Group and not formally a mentor, volunteered time to teach students how to crimp power connections and solder the printed circuit boards. In addition, guest lecturers at the course, Paul Breimyer and Adam Norige of the Laboratory's Homeland Protection and Air Traffic Control Division, shared their experience in developing networked systems and intuitive visualizations for sensor data.

"Our role was to ask questions of the students, force them to go through their thought processes," says Shah. Cross credits this questioning with helping students "to gain a better sense of the real-world expectations for a sensor network project."

Photo of a CLAIRITY sensorThe sensor for the CLAIRITY network is built primarily from commercially available parts and is housed in a weather-resistant case 3D printed at the Beaver Works Center.

Student teams were each assigned to a particular aspect of the network—for example, hardware procurement and fabrication, sensor calibration, software code development. "The students did a great job collaborating with each other; I think this was essential for the success of the project," says Tsiligkaridis. "My involvement consisted of helping the students troubleshoot and calibrate the sensors, giving a few pointers on technical matters," he adds.

The network, named CLAIRITY, continuously transmitted data via a wireless connection to a central computer for storage in a database that would populate the data display at the CLAIRITY website. This web portal relays near-real-time results of the monitors' measurements. The CLAIRITY reports have revealed that the air around the campus contains gas and particulate levels within the Environmental Protection Agency's acceptable ranges. Shah notes that the Laboratory mentors saw opportunities for exploring more implications of the air quality data, but "there was just not enough time to both collect the data and exploit them. One of the things the Lab does well is making sense of data; we hope to offer this help in future collaborations."  

Another thing the mentors offered the students was a visual aid to their final presentation of the project on 6 May in Building 46. A 3D-printed map of the MIT campus was transported to campus, and its computer interface was configured to "light up" a node site on demand and then project the CLAIRITY reports of that location's air quality. The map, called LuminoCity, had been constructed for the Laboratory's PrintLab innovation challenge by a team who used the Lincoln Laboratory Technology Office Innovation Lab's 3D printing equipment.

The 3D-printed map of MITThe 3D-printed LuminoCity map of the MIT campus, created by Lincoln Laboratory staff members, was loaned to the CEE students to dramatize their air-quality-monitoring project.

"This [using the map] was a last-minute idea," says Shah, who was contacted by Zach Weber, one of LuminoCity's developers who thought the map could be useful in presenting the students' data. "Zach and I worked feverishly the night before the presentation to design the time-lapse visualizations to display two days' worth of CLAIRITY raw data collected from the nodes."

The interactive map added a touch of drama to a successful project that demonstrated a cost-effective system that planners could employ to gather air quality assessments useful in designing environmentally "smart" cities.

Posted June 2014

top of page