A team from the University of Houston received a grant to continue its work on using AI and digital twin technology to better evaluate bridges in Texas. Photo via uh.edu

A University of Houston professor has received a grant from the Texas Department of Transportation (TxDOT) to improve the efficiency and effectiveness of how bridges are inspected in the state.

The $505,286 grant will support the project of Vedhus Hoskere, assistant professor in the Civil and Environmental Engineering Department, over three years. The project, “Development of Digital Twins for Texas Bridges,” will look at how to use drones, cameras, sensors and AI to support Texas' bridge maintenance programs.

“To put this data in context, we create a 3D digital representation of these bridges, called digital twins,” Hoskere said in a statement. “Then, we use artificial intelligence methods to help us find and quantify problems to be concerned about. We’re particularly interested in any structural problems that we can identify - these digital twins help us monitor changes over time and keep a close eye on the bridge. The digital twins can be tremendously useful for the planning and management of our aging bridge infrastructure so that limited taxpayer resources are properly utilized.”

The project began in September and will continue through August 2026. Hoskere is joined on the project by Craig Glennie, the Hugh Roy and Lillie Cranz Cullen Distinguished Chair at Cullen College and director of the National Center for Airborne Laser Mapping, as the project’s co-principal investigator.

According to Hoskere, the project will have implications for Texas's 55,000 bridges (more than twice as many as any other state in the country), which need to be inspected every two years.

Outside of Texas, Hoskere says the project will have international impact on digital twin research. Hoskere chairs a sub-task group of the International Association for Bridge and Structural Engineering (IABSE).

“Our international efforts align closely with this project’s goals and the insights gained globally will enhance our work in Texas while our research at UH contributes to advancing bridge digitization worldwide,” he said. “We have been researching developing digital twins for inspections and management of various infrastructure assets over the past 8 years. This project provides us an opportunity to leverage our expertise to help TxDOT achieve their goals while also advancing the science and practice of better developing these digital twins.”

Last year another UH team earned a $750,000 grant from the National Science Foundation for a practical, Texas-focused project that uses AI. The team was backed by the NSF's Convergence Accelerator for its project to help food-insecure Texans and eliminate inefficiencies within the food charity system.

UH Professor Vedhus Hoskere received a three-year, $505,286 grant from TxDOT for a bridge digitization project. Photo via uh.edu

The company wants to make Texas "the home of self-driving trucks." PRNewsfoto/Kodiak Robotics

Self-driving semi trucks are now hauling cargo in and out of Houston

Autopilot

The Interstate 45 freight corridor between Houston and Dallas now serves as a testing ground for self-driving cargo trucks.

Silicon Valley startup Kodiak Robotics Inc. recently began sending its autonomous 18-wheel trucks on trips between Texas' two largest metro areas, co-founder and CEO Don Burnette says. The trucks are carrying paid cargo, but Kodiak won't identify the customer or customers. The company also won't say how many trips the trucks are making each day.

The Texas initiative represents Kodiak's first foray into commercial deliveries. Wired.com notes that pretty much every player in the autonomous truck sector has conducted tests in Texas or is carrying commercial loads in the Lone Star State, which boasts more than 2,300 miles of interstate highways.

For its part, Kodiak aims to make Texas "the home of self-driving trucks."

According to a 2016 report from the Texas Department of Transportation, nearly half of all truck freight in Texas goes through the I-45 corridor's 11 counties. In some spots, trucks make up more than one-fourth of the traffic in the corridor, which runs 276 miles from Galveston to where I-45 intersects with Interstate 20 in Dallas County, the TxDOT report says.

"The importance of the I‐45 freight corridor to the movement of goods extends beyond Texas because much of the freight originating or passing through the corridor is destined to other domestic and international markets," the report says.

For now, Burnette says, two people are aboard each Kodiak truck traveling between Houston and Dallas — a safety driver and a safety engineer.

"Dallas will be our home base for testing and operations for the foreseeable future," Burnette says. "Kodiak plans to continue refining and testing its trucks until the last truck-involved accident happens on public roads."

Kodiak's Dallas office, which opened in March, employs eight people. The company plans to relocate soon to new office space in the Dallas suburb of Lancaster, Burnette says.

At this time, Kodiak doesn't plan to hire any workers in Houston, he says.

From its base in the Dallas area, Kodiak envisions expanding its service to routes throughout Texas, but it's focusing solely on the Houston-to-Dallas route for the time being, Burnette says.

Kodiak picked Texas for its truck tests, in part, because of the "warm welcome" extended by Gov. Greg Abbott, TxDOT, the Texas Department of Public Safety, the Texas A&M Transportation Institute, and other segments of the public sector, he says.

In addition, Burnette says, Kodiak chose Texas "because of its great people, freight-rich economy, reasonable regulatory structure, and robust infrastructure."

In 2017, Texas enacted laws enabling driverless vehicles, including long-haul trucks, to operate on the state's roads.

"Texas is a leader in the testing and implementation of connected and automated vehicles, and Kodiak's willingness to partner with academia and public agencies to ensure safe deployment of new technology will add significant value to our transportation system," Christopher Poe, assistant director of the Texas A&M Transportation Institute, says in a release.

Burnette, co-founder of Otto Trucking LLC, a self-driving truck startup purchased in 2016 by Uber Technologies Inc., and fellow entrepreneur Paz Eshel established Kodiak in 2018 to "redefine" long-haul trucking through self-driving technology.

Kodiak says its autonomous technology is designed to ease pressures facing the trucking industry, including a shortage of drivers and high turnover among drivers, while improving highway safety, fostering business efficiency, reducing traffic congestion, and cutting down on harmful emissions.

"Long-haul trucking is primed for autonomous technology," Kodiak says in a post on Medium. "Highway driving is more structured and predictable than urban driving. This means there are fewer decisions for drivers to make and [it's] a better fit for autonomous vehicles."

"As hard as it is to navigate city streets, autonomous vehicles are much closer to being able to drive on more structured interstate highways, which have no jaywalking pedestrians, no aggressive cyclists, and no runaway pets," Kodiak adds. "That's why we've focused on building technology specifically for long-haul trucks driving on highway routes, often referred to as the 'middle mile.'"

If you have an opinion about trains, here's your chance. Rendering courtesy of Texas Central

TxDOT asks Houstonians for input on rail projects

All aboard

The Texas Department of Transportation is updating a document called the Texas Rail Plan and is seeking input from the public.

The update is designed to reflect the latest rail project priorities and fulfill eligibility requirements for federal funding. Federal requirements say that states' rail plans must be updated every four years to establish policy, priorities, and implementation strategies for freight and passenger rail in the state.

The Texas Rail Plan includes a list of current and future rail projects, which are also depicted on a map. The plan keeps inventory of all rail lines; analyzes rail service goals and contributions to the economy; catalogs and assesses potential infrastructure projects; and examines finance strategies.

The update project began in summer 2018. Meetings of stakeholders were held, and now there's an opportunity for public input. (Stakeholders include citizens, neighboring states, public agencies, and the private rail industry.)

There'll be another round of meetings in the spring, and then the updated plan will be released in summer 2019.

TxDOT hosted a public meeting on December 11, when it presented the update and asked for public questions and input. After that meeting, they extended their deadline for comments to March 1, 2019.

As for now, the public can review and provide input on the plan via this website which explains the history of the Rail Plan and some of the reasons why an update is being done.

There's a survey and online form to submit public comments until January 8, 2019.

If you feel equipped to answer 10 questions such as, "What could be done in Texas to improve freight rail access, promote economic development, and enhance the state's competitiveness in national markets and the global marketplace?" — then this survey should be right up your alley. (Although it should be noted that the hardest questions are first and they get easier as you go along. Also, it's multiple choice.)

These options provide an opportunity for the public to comment on all rail-related issues in Texas, both freight and passenger, as well as existing and future projects and programs.

------

This story originally appeared on CultureMap.

Ad Placement 300x100
Ad Placement 300x600

CultureMap Emails are Awesome

Houston doctor wins NIH grant to test virtual reality for ICU delirium

Virtual healing

Think of it like a reverse version of The Matrix. A person wakes up in a hospital bed and gets plugged into a virtual reality game world in order to heal.

While it may sound far-fetched, Dr. Hina Faisal, a Houston Methodist critical care specialist in the Department of Surgery, was recently awarded a $242,000 grant from the National Institute of Health to test the effects of VR games on patients coming out of major surgery in the intensive care unit (ICU).

The five-year study will focus on older patients using mental stimulation techniques to reduce incidences of delirium. The award comes courtesy of the National Institute on Aging K76 Paul B. Beeson Emerging Leaders Career Development Award in Aging.

“As the population of older adults continues to grow, the need for effective, scalable interventions to prevent postoperative complications like delirium is more important than ever,” Faisal said in a news release.

ICU delirium is a serious condition that can lead to major complications and even death. Roughly 87 percent of patients who undergo major surgery involving intubation will experience some form of delirium coming out of anesthesia. Causes can range from infection to drug reactions. While many cases are mild, prolonged ICU delirium may prevent a patient from following medical advice or even cause them to hurt themselves.

Using VR games to treat delirium is a rapidly emerging and exciting branch of medicine. Studies show that VR games can help promote mental activity, memory and cognitive function. However, the full benefits are currently unknown as studies have been hampered by small patient populations.

Faisal believes that half of all ICU delirium cases are preventable through VR treatment. Currently, a general lack of knowledge and resources has been holding back the advancement of the treatment.

Hopefully, the work of Faisal in one of the busiest medical cities in the world can alleviate that problem as she spends the next half-decade plugging patients into games to aid in their healing.

Houston scientists develop breakthrough AI-driven process to design, decode genetic circuits

biotech breakthrough

Researchers at Rice University have developed an innovative process that uses artificial intelligence to better understand complex genetic circuits.

A study, published in the journal Nature, shows how the new technique, known as “Combining Long- and Short-range Sequencing to Investigate Genetic Complexity,” or CLASSIC, can generate and test millions of DNA designs at the same time, which, according to Rice.

The work was led by Rice’s Caleb Bashor, deputy director for the Rice Synthetic Biology Institute and member of the Ken Kennedy Institute. Bashor has been working with Kshitij Rai and Ronan O’Connell, co-first authors on the study, on the CLASSIC for over four years, according to a news release.

“Our work is the first demonstration that you can use AI for designing these circuits,” Bashor said in the release.

Genetic circuits program cells to perform specific functions. Finding the circuit that matches a desired function or performance "can be like looking for a needle in a haystack," Bashor explained. This work looked to find a solution to this long-standing challenge in synthetic biology.

First, the team developed a library of proof-of-concept genetic circuits. It then pooled the circuits and inserted them into human cells. Next, they used long-read and short-read DNA sequencing to create "a master map" that linked each circuit to how it performed.

The data was then used to train AI and machine learning models to analyze circuits and make accurate predictions for how untested circuits might perform.

“We end up with measurements for a lot of the possible designs but not all of them, and that is where building the (machine learning) model comes in,” O’Connell explained in the release. “We use the data to train a model that can understand this landscape and predict things we were not able to generate data on.”

Ultimately, the researchers believe the circuit characterization and AI-driven understanding can speed up synthetic biology, lead to faster development of biotechnology and potentially support more cell-based therapy breakthroughs by shedding new light on how gene circuits behave, according to Rice.

“We think AI/ML-driven design is the future of synthetic biology,” Bashor added in the release. “As we collect more data using CLASSIC, we can train more complex models to make predictions for how to design even more sophisticated and useful cellular biotechnology.”

The team at Rice also worked with Pankaj Mehta’s group in the department of physics at Boston University and Todd Treangen’s group in Rice’s computer science department. Research was supported by the National Institutes of Health, Office of Naval Research, the Robert J. Kleberg Jr. and Helen C. Kleberg Foundation, the American Heart Association, National Library of Medicine, the National Science Foundation, Rice’s Ken Kennedy Institute and the Rice Institute of Synthetic Biology.

James Collins, a biomedical engineer at MIT who helped establish synthetic biology as a field, added that CLASSIC is a new, defining milestone.

“Twenty-five years ago, those early circuits showed that we could program living cells, but they were built one at a time, each requiring months of tuning,” said Collins, who was one of the inventors of the toggle switch. “Bashor and colleagues have now delivered a transformative leap: CLASSIC brings high-throughput engineering to gene circuit design, allowing exploration of combinatorial spaces that were previously out of reach. Their platform doesn’t just accelerate the design-build-test-learn cycle; it redefines its scale, marking a new era of data-driven synthetic biology.”