UNIVERSITY PARK, Pa. — Bolstering their efforts to transform the undergraduate laboratory curriculum, researchers in the Penn State Department of Mechanical Engineering (ME) recently published their innovative methods to engage students and enhance their learning in the American Society of Engineering Education’s Advances in Engineering Education.
“People often think the purpose of labs is to reinforce the knowledge students have learned in their lectures,” said Siu Ling “Pansy” Leung, director of undergraduate laboratories and the paper’s lead author. “However, literature suggests that doesn’t necessarily improve the students’ understanding on the materials.”
With this in mind, the ME Lab Renovation Committee sought to redesign the entire laboratory experience to find new ways to fortify the students’ engineering knowledge and additionally equip them with problem solving skills.
“We want to train students for their future careers,” Leung said. “Instead of giving them procedure-orientated instructions, we give them tools and material to figure out how to solve the problem.”
The work to introduce new course lessons was well underway before the COVID-19 pandemic. However, the circumstances of an all-virtual curriculum in the spring of 2020 led the team to accelerate their plans.
“We took this situation as an opportunity to pilot one of the experiments in our new problem-based laboratory course virtually,” Leung said.
The first of several modules on varying engineering topics, a novel lesson plan introduced by the team aims to reinforce statistics knowledge imparted in a junior level required course. The new challenge was that the students needed to apply that information to solve a problem using machine learning.
As the hands-on exercise, the overall lesson plan charged students to develop an algorithm that allowed their smartphones to predict human movement.
“Our students encounter machine learning every day,” Leung said. “A smartphone recognizes your speech, a thermostat changes to the temperature you like.”
The students were provided with a large data set of measurements of different activities, like walking and running. Using that information, they were asked to use their prior statistical knowledge to extract machine learning features from these data and develop a mathematical model that could accurately predict human movement. With that algorithm set, they tested and validated their work through their personal smartphones.
Leung reports the students exceeded her expectations, with some students reporting 100% accuracy in their models. However, she was most impressed that students took the initiative to return to the project after its completion.
“If someone’s algorithm was able to predict the motion 20% of the time, some said, ‘OK, how can I improve this?’” she said. “That is the beauty of these experiments. Students have the ability and the interest to go back and improve their own work.”
All of the new laboratory courses, including this module, will be rolled out fully by the fall of 2022 and encompass modern engineering challenges.
“For instance, in the fluid mechanics module, we will be tying it to renewable energy solutions in a wind turbine,” Leung said. “All of these lessons aim to teach about a topic that our students see in the world right now that engineers are trying to make better and better.”
Together with the new construction of the ME Knowledge Lab, Karen Thole, distinguished professor and mechanical engineering and department head, is confident the education that undergraduates receive will prepare them well as future engineers.
“A thriving 21st-century learning environment has to embrace agility, create flexibility, adapt to technology and support virtual team collaborations,” Thole said. “The success of this program inspires us to keep making innovations.”
The ME Knowledge Lab Committee who contributed to this project included: Glenn Professor of Engineering Education Eric Marsh; Professors Joseph Sommer and Sean Brennan; Associate Professor Stephen Lynch; Wormley Early Career Professor Tak-Sing Wong; Assistant Professors Daniel Cortes, Jean-Michel Mongeau and Brian Foley; Brianne Hargrove, doctoral student in mechanical engineering; and Andrea Gregg, director of online pedagogy.