Penn Engineering

The School of Engineering and Applied Science, established in 1852, is composed of six academic departments and numerous interdisciplinary centers, institutes, and laboratories. At Penn Engineering, we are preparing the next generation of innovative engineers, entrepreneurs and leaders. Our unique culture of cooperation and teamwork, emphasis on research, and dedicated faculty advisors who teach as well as mentor, provide the ideal environment for the intellectual growth and development of well-rounded global citizens.

Search results

Now showing 1 - 10 of 132
  • Publication
    Learning Environmental Models With Multi-Robot Teams Using A Dynamical Systems Approach
    (2022-01-01) Salam, Tahiya
    Robots monitoring complex, spatiotemporal phenomena require rich, meaningful representations of the environment. This thesis presents methods for representing the environment as a dynamical system with machine learning techniques. Specifically, we formulate machine learning methods that lend to data-driven modeling of the phenomena. The data-driven modeling explicitly leverages theoretical foundations of dynamical systems theory. Dynamical systems theory offers mathematical and physically interpretable intuitions about the environmental representation. The contributions presented include distributed algorithms, online adaptation, uncertainty quantification, and feature extraction to allow for the actualization of these techniques on-board robots. The environmental representations guide robot behavior in developing strategies such as optimal sensing and energy-efficient navigation. The methods and procedures provided in this thesis were verified across complex, spatiotemporal environments and on experimental robots.
  • Publication
    Verifying the Safety of Autonomous Systems with Neural Network Controllers
    (2020-12-01) Ivanov, Radoslav; Carpenter, Taylor J.; Weimer, James; Alur, Rajeev; Pappas, George; Lee, Insup
    This paper addresses the problem of verifying the safety of autonomous systems with neural network (NN) controllers. We focus on NNs with sigmoid/tanh activations and use the fact that the sigmoid/tanh is the solution to a quadratic differential equation. This allows us to convert the NN into an equivalent hybrid system and cast the problem as a hybrid system verification problem, which can be solved by existing tools. Furthermore, we improve the scalability of the proposed method by approximating the sigmoid with a Taylor series with worst-case error bounds. Finally, we provide an evaluation over four benchmarks, including comparisons with alternative approaches based on mixed integer linear programming as well as on star sets.
  • Publication
    Cellular Agriculture
    (2022-04-19) Kim, Christina; Kishun, Amanda; Lubna, Fahmida
    Cellular agriculture is a field of biotechnology focused on the production of animal products using cells grown in vitro . Traditional meat production consumes vast amounts of water, arable land, and feed crops, as well as driving deforestation, emitting large amounts of greenhouse gases, and creating large potential reservoirs for zoonotic diseases. As the global demand for meat increases, continuing to scale up the industry for slaughtered meat could have disastrous consequences for the environment. Growing cells in bioreactors creates the potential to drastically decrease land requirements, feed requirements, and other environmental impacts. For example, hindgut fermentation of feed, the main source of methane emissions from cattle farming, can be eliminated entirely by supplying the cells with pure glucose. This report proposes a process to produce 35 million pounds per year of a cultured ground beef product. The process starts with a starter colony of bovine muscle satellite cells, which are proliferated, differentiated to bovine muscle fiber, and then dewetted, mixed with plant-based fat, and extruded to the final product. Bubble column bioreactors are used for the seed train, final proliferation, and differentiation steps in order to adequately oxygenate large process volumes without threatening cell viability. The process shows profitability at a price of $100 per pound of product. The plant has a return on investment of 217%, an investor’s rate of return of 223%, and a cumulative net present value of about $2 billion over the plant’s lifespan.
  • Publication
    CO2 Sequestration by Allam Cycle
    (2021-04-20) Chaturvedi, Raghav; Kennedy, Eric; Metew, Sarron
    Natural gas powerplants account for 40% of the electricity generation in the United States[1] and 617 million tons of CO2 emissions a year[2]. The largest powerplants with carbon capture technology utilize a post-combustion absorption technology that must treat a large volume of flue gas and compress CO2 to pipeline specifications from near-ambient pressure. The Allam cycle, patented in 2013 by Rodney Allam, uses oxy-combustion and a supercritical CO2 stream as the working fluid to produce high-purity liquid pipeline CO2. While it was developed commercially at a 50-megawatt thermal (MWt) plant in 2018, the economics for a larger, 300 MW plant had not been documented. This project shows that under the current US tax code, the Allam cycle is less economical than the traditional natural gas combined cycle (NGCC) and NGCC with CDR. However, due to the over 99% capture rate, compared to 90% in post-combustion capture, the breakeven credit to traditional NGCC of $112/tonne for the Allam cycle is lower than the NGCC with CDR breakeven credit of $121/tonne. Similarly, for a desired IRR of 15%, the CO2 credit required for the Allam cycle is $163/tonne compared to $188/tonne for the NGCC with CDR. The Allam cycle provides increasingly better financial returns than the NGCC with CDR as the tax credit for sequestration rises. The results of this analysis were produced by first simulating both powerplants in Aspen Plus, and then conducting a discounted cash flow analysis for various scenarios.
  • Publication
    Calibrated Prediction with Covariate Shift via Unsupervised Domain Adaptation
    (2020-03-01) Park, Sangdon; Bastani, Osbert; Weimer, James; Lee, Insup
    Reliable uncertainty estimates are an important tool for helping autonomous agents or human decision makers understand and leverage predictive models. However, existing approaches to estimating uncertainty largely ignore the possibility of covariate shift—i.e., where the real-world data distribution may differ from the training distribution. As a consequence, existing algorithms can overestimate certainty, possibly yielding a false sense of confidence in the predictive model. We propose an algorithm for calibrating predictions that accounts for the possibility of covariate shift, given labeled examples from the training distribution and unlabeled examples from the real-world distribution. Our algorithm uses importance weighting to correct for the shift from the training to the real-world distribution. However, importance weighting relies on the training and real-world distributions to be sufficiently close. Building on ideas from domain adaptation, we additionally learn a feature map that tries to equalize these two distributions. In an empirical evaluation, we show that our proposed approach outperforms existing approaches to calibrated prediction when there is covariate shift.
  • Publication
    Repeated Jumping with the REBOund: Self-Righting Jumping Robot Leveraging Bistable Origami-Inspired Design
    (2022-03-01) Sun, Yuchen; Wang, Joanna; Sung, Cynthia R.
    Repeated jumping is crucial to the mobility of jumping robots. In this paper, we extend upon the REBOund jumping robot design, an origami-inspired jumping robot that uses the Reconfigurable Expanding Bistable Origami (REBO) pattern as its body. The robot design takes advantage of the pattern's bistability to jump with controllable timing. For jump repeatedly, we also add self-righting legs that utilize a single motor actuation mechanism. We describe a dynamic model that captures the compression of the REBO pattern and the REBOund self-righting process and compared it to the physical robot. Our experiments show that the REBOund is able to successfully self-right and jump repeatedly over tens of jumps. Supplemental video:
  • Publication
    Exploring with Sticky Mittens: Reinforcement Learning with Expert Interventions via Option Templates
    (2022-11-17) Dutta, Souradeep; Sridhar, Kaustubh; Bastani, Osbert; Dobriban, Edgar; Weimer, James; Parish-Morris, Julia
    Long horizon robot learning tasks with sparse rewards pose a significant challenge for current reinforcement learning algorithms. A key feature enabling humans to learn challenging control tasks is that they often receive expert intervention that enables them to understand the high-level structure of the task before mastering low-level control actions. We propose a framework for leveraging expert intervention to solve long-horizon reinforcement learning tasks. We consider option templates, which are specifications encoding a potential option that can be trained using reinforcement learning. We formulate expert intervention as allowing the agent to execute option templates before learning an implementation. This enables them to use an option, before committing costly resources to learning it. We evaluate our approach on three challenging reinforcement learning problems, showing that it outperforms state-of-the-art approaches by two orders of magnitude.