Need a last-minute Halloween costume idea? This neural network has got you covered!

Need a last-minute costume idea for Halloween? How about a cyborg bat? Or a vampire shark? Or a magic sexy hamburger?
These are all costumes generated by a neural network trained by Jacobs School alumna Janelle Shane. Shane, who earned a Ph.D. in electrical engineering at UC San Diego in the lab of Professor Shaya Fainman, works with lasers by day. But her hobby is working with neural networks to create funny data sets.
For this project, she crowdsourced 4500 costume ideas from her blog readers and fed them to a neural network.
The network did not disappoint, generating costume suggestions such as vampire Big Bird, celery blue Frankenstein and strawberry shark.
Soon, Shane's readers were getting into the game and drawing the costumes the neural network suggested.
First up, strawberry shark:

And then: Bearley Quinn (courtesy of Twitter user @vonbees):

But Shane's readers weren't done. Soon they started making some of the costume suggestions a reality.
Twitter user Liz Walsh dressed up as the Dragon of Liberty:

Twitter user @HerbLovesTech and his wife dressed up as Professor Panda and Shark Princess:

And Shane? She took her inspiration from an entry in the costume data base. She will be Ruth Vader Ginsburg (that's a mash up of Supreme Court Justice Ruth Bader Ginsburg and Star Wars villain Darth Vader):

For more neural-network generated Halloween costumes, read Shane's blog post here. And read this news story by writer Rae Paoletta here and this Popular Mechanics story by writer Sophie Weiner.

UC San Diego Mechanical and Aerospace Engineering Professor Miroslav Krstic Receives ASME Rufus Oldenburger Medal

Photo: Krstic (r) with Peter Meckl,
Chair of ASME Dynamic Systems & Control Division. 

UC San Diego mechanical and aerospace engineering professor Miroslav Krstic received the ASME Rufus Oldenburger Medal for lifetime achievements in automatic control at 10th ASME Dynamic Systems& Control Conference in Washington, DC in October 2017. 

Krstic’s acceptance lecture was on control of congested traffic (abstract at end of blog post).

Krstic serves as Sr. Assoc. Vice Chancellor for Research at UC San Diego. He is Director of the Cymer Center for Control Systems and Dynamics and holds the Daniel L. Alspach Endowed Chair in Dynamic Systems and Control.

Krstic is the mechanical and aerospace engineering department’s second recipient of the Oldenburger Medal, following Professor Bob Bitmead in 2014.

The Rufus Oldenburger Medal is a prestigious Society award for lifetime achievements in automatic control. Inaugurated in 1968, the medal recognizes significant contributions and outstanding achievements in the field of automatic control. Such achievements may be, for example, in the areas of education, research, development, innovation, and service to the field and profession. The award was established to honor Rufus Oldenburger for his distinctive achievements in the field and for his service to the Society and the Division. The list of recipients is a true honor role of major contributors to the science and profession of control. 

Abstract: Control of freeway traffic using ramp metering is a “boundary control” problem when modeling is approached using widely adopted coupled hyperbolic PDE models of the Aw-Rascle-Zhang type, which include the velocity and density states, and which incorporate a model of driver reaction time. Unlike the “free traffic” regime, in which ramp metering can affect only the dynamics downstream of the ramp, in the “congested traffic” regime ramp metering can be used to suppress stop-and-go oscillations both downstream and upstream of the ramp - though not both simultaneously. Controlling the traffic upstream of a ramp is harder - and more interesting - because, unlike in free traffic, the control input doesn’t propagate at the speed of the vehicles but at a slower speed, which depends on a weighted difference between the vehicle speed and the traffic density. I will show how PDE backstepping controllers, which have been effective recently in oil drilling and production applications (similarly modeled by coupled hyperbolic PDEs), can help stabilize traffic, even in the absence of distributed measurements of vehicle speed and density, and when driver reaction times are unknown.

Blue LINC hosts Medical Innovators Hall of Fame Series

The Blue LINC Healthcare Incubator, UC San Diego's first biomedical incubator, will kick off its new Medical Innovators Hall of Fame Series with a presentation by Michael Ackermann, former CEO of med-tech startup Oculeve. Oculeve, which developed a tear-simulation device for those with dry-eye disease, was co-founded by Garrett Smith, a Ph.D. candidate in bioengineering at the Jacobs School of Engineering, and eventually acquired by Allergan.

During his talk titled "From University Collaboration to $100M Acquisition: A Tearful Tale of BioDesign," Ackermann will explain how acquisition by a global pharmaceutical giant is helping him achieve his goal of reaching as many patients as possible and will highlight his journey as a BioDesign Fellow at the Stanford Byers Center for BioDesign. Ackermann will discuss why big tech companies have yet to disrupt healthcare and how that translates into big opportunities for entrepreneurs, students, and faculty interested in startups.

The seminar is scheduled for Thursday, Oct. 26 from 6:00- 7:15 p.m. in Fung Auditorium in the Powell-Focht Bioengineering Hall. Register to attend at http://bluelincsd.com/.

A new model for electrochemical kinetics in nanoscale systems

Understanding the speed at which electrochemical reactions occur can provide scientific insight for various processes ranging from biochemical reactions to charge storage in capacitors and batteries. However, to date, many of the theoretical and experimental analyses of electrochemical reaction speed- such as those in the widely used Butler-Volmer formulations are based on classical thermodynamics and adapt 19th century-based Arrhenius theory. In these cases, the charge transfer rate is assumed to constantly increase with applied voltage. While complementary theories consider the influence of the configurational rearrangements in the electrolyte and energy level occupancy, none have related the kinetics to the specific arrangement of the electrons in the material constituting the electrode. The latter aspect is very important for nanoscale materials where the bulk is but a small part of the whole.

Recently, a team of engineers at UC San Diego led by professor of mechanical engineering Prab Bandaru and involving Ph.D. students Hidenori Yamada and Rajaram Narayanan, probed in detail, both theoretically and experimentally, the specific characteristics of a nanostructured material with respect to its effect on charge transfer. They demonstrated that in a one-dimensional nanotube, the electrons are confined to a line, while in two-dimensional graphene, the electrons are confined to a plane. Based on these findings, the researchers expect that the restriction on electron motion hinders charge transfer and electrochemical kinetics. On the other hand, the reduced electron scattering could enhance the kinetics. The team resolved these issues by taking advantage of the specific arrangement of the electrons in the nanostructure. They applied their theories to explain the experimental variation of the electrochemical rate constant of single layer graphene.

(a) Atomic force microscopy image of a section of the single layer graphene (SLG) sample transferred onto a p-Si/SiO₂ substrate. The wrinkles on the sample surface corresponding to the line scan (white line) are displayed in the lower left inset. The Raman spectrum of the transferred SLG is indicated in the top right inset. (b) Schematic of the three-electrode droplet electrochemical cell (actual experimental arrangement shown in the top right inset). The SLG working electrode (WE), Pt wire counter electrode (CE) and a reference (REF) saturated calomel electrode are indicated.

The researchers detailed their findings in a recent issue of the Journal of Physical Chemistry Letters. 

The team discovered that the charge transfer rate may either increase, decrease or remain constant, and that such variation is sensitive to the orientation as well as the relevant dimensionality of the nanostructure. As charge transfer per unit time determines the electrical current that may be obtained from a given electrode, the UC San Diego study provides a firm rationale for the use of nanostructures in charge storage electrodes, with applications encompassing solid state battery-related systems, wearable sensors, etc., where electrical current modulations would impact energy and power delivery.

A plot of the charge transfer related electrochemical rate constant (k) normalized to the k_η=0V as a function of the applied voltage (η), considered with respect to the redox potential. The experimental data is a poor fit with the theoretical fits expected from conventional Butler-Volmer (B-V) kinetics as well as three-dimensional Marcus-Hush-Chidsey (MHC) kinetics, but could be fit well through a dimensionality dependent electrochemical model proposed by a team of engineers at UC San Diego.

Paper: Dimensionality-Dependent Electrochemical Kinetics at the Single-Layer Graphene–Electrolyte Interface, R. Narayanan, H. Yamada, B.C. Marin, A. Zaretski, and P.R. Bandaru, J. Phys. Chem. Lett., 2017, 8 (17), pp 4004–4008.

3D-Printed Space Rocket Startup Funded by New VC Fund Contrary Capital

UC San Diego Jacobs School of Engineering students are the founders of one of the first two university startups to receive funding by a new VC fund called Contrary Capital.

Contrary Capital has a novel take on tracking down university startups to invest in. The details are outlined in a story by Mike Freeman ( @TechDiego on Twitter ) in the San Diego Union Tribune:

This venture cap fund taps university students to find the best on-campus investments.

The Jacobs School startup that received funding is Additive Rocket Corporation (ARC), which 3D prints high-impulse, low-cost, lightweight metal rocket engines for the space industry. Additive Rocket Corp. Founded in 2015 by recent graduates Andy Kieatiwong and Kyle Adriany. According to the ARC website, “space exploration hinges on innovation of propulsion technology.”

The ARC students have participated in a number of entrepreneurism programs on campus, including The Basement and the Qualcomm Institute Innovation Space.

We look forward to tracking ARC’s successes. 

Good luck, and may the [propulsive] force be with you!

Combining soft robotics and space technology

Paul Glick, a Ph.D. student at the Jacobs School, got a unique chance to do hands-on at the Jet Propulsion Laboratory in Pasadena, Calif.
Glick, who works in the lab of mechanical engineering professor and roboticist Michael Tolley, got to design and carry out most of the experiments for an electrostatic gripper for flexible objects build by JPL and UC Berkeley engineers. The team presented their work at the IROS 2017 conference in late September in Vancouver.
Glick is part of the NASA Space Technology Research Fellowship program. He works to bring soft robotics to space technology. Here is a more detailed description of his research. 
Tolley's group will present some of their research at the Oct. 27 Contextual Robotics Forum here on the UC San Diego campus. 
Watch a video of the gripper that Glick ran experiments on in action: