“Active Sensing” News, ESS Journal, Press Release ​

1. Nationwide News through Spectrum Network Link

News on AI Data-Driven Active Sensing

2. ESS Journal Publication Link

In collaboration with Masahiro Ono at Hui Su at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (80NM0018D0004), and Masashi Minamide at the University of Tokyo, the paper “A Sampling-Based Path Planning Algorithm for Improving Observations in Tropical Cyclones” was accepted for publication by Earth and Space Science (ESS) Journal. https://doi.org/10.1029/2020EA001498

Earth and Space Science Journal, January 2022

3. NCA&T Press Release

As another hurricane season begins, a computational science and engineering faculty member and two Ph.D. students at North Carolina Agricultural and Technical State University aim to improve tropical forecasting and modeling by deploying drones to collect data from the eye of hurricanes.

Hyoshin Park, Ph.D., and Ph.D. candidates Larkin Folsom and Justice Darko will explore ways to maximize the data collection and possible flight paths in future storms in this project supported by NASA’s Jet Propulsion Laboratory (JPL). During the summer of 2019, Park visited JPL and formulated the problem of efficient hurricane data collection using drones in collaboration with Masahiro Ono, Ph.D., a research technologist at the laboratory.

“Our department’s innovative, artificial intelligence-based model explores the search space to improve the hurricane track and intensity,”

said Park, an assistant professor in the College of Engineering

In 2014, the National Oceanic and Atmospheric Administration (NOAA) released two unmanned aerial vehicles (UAVs) into the eye of Hurricane Edouard, at the time the first category 3 or stronger storm to form in the Atlantic Ocean since Hurricane Sandy. The UAVs were the first-ever deployed into the eye of a tropical system and collected data inside the eye and the storm’s outer vortex, enduring for 68 minutes before plunging into the ocean.

The team will use early measurements from NOAA when the hurricane is forming to determine the storm’s wind-field and other important data. The crew will then know the ideal location to drop up to 10 drones, which will communicate with each other and to their base on the plane, effectively collecting as much data as possible.

For years, hurricane hunters have launched dropsondes into the center of the storm, near its eyewall, to determine the boundaries and structures of a hurricane. The device contains a GPS receiver, along with pressure, temperature and humidity (PTH) sensors to capture atmospheric profiles and thermodynamic data.

This method, while critical to yield important data for forecasting, requires aircraft and crew to drop to unsafe altitudes for a better view of the drop zone. This maneuver takes more time and fuel and puts the crew at risk. Additionally, the process requires numerous dropsondes to gather sufficient data about the hurricane and its forecast.

The team will repurpose the methodology from Park and Folsom’s 2018 proposal funded by the National Science Foundation’s Division of Information and Intelligent Systems Robust Intelligence program. The study focused on maximizing data collection for a Mars mission exploration of an unexplored environment with no flow of humans or traffic.

“This research is important because it will help improve hurricane forecasts, saving local economies money by more precisely constraining the forecast track and intensity. Locations that will not be impacted can keep their economies open, and those that will be impacted can more effectively prepare.”

said Folsom

Additionally, the research will help improve public trust in tropical forecasting. The A&T team members are also joined by Hui Sui, Ph.D., principal investigator and JPL engineering and science directorate Stratosphere and Upper Troposphere, Masahiro Ono, Ph.D., JPL research technologist from Robotic Surface Mobility and Masashi Minamide, Ph.D., an assistant professor at the University of Tokyo. 

EAST GREENSBORO, N.C. (June 1, 2020) –

written by Alexander Saunders

A Semi-autonomous Vehicle Patent Granted

System and Method for Predicting Effects of Forward Glance Durations on Latent Hazard Detections, USSN 16/244,758 invention is allowed by USPTO.Hyoshin “John” Park, Ph.D., an assistant professor in the Department of Computational and Data Science at North Carolina Agricultural and Technical State University, has been awarded his second patent, “System and Method for Predicting Effects of Forward Glance Durations on Latent Hazard Detection.” Recently, Park received his first patent for a cost-effective solution to reduce traffic congestion.

The patented system addresses two issues that have not previously been combined: detecting road hazards while simultaneously determining if the driver has lost focus on the road. This approach is particularly relevant to society’s safe transition to an autonomous future.

One possible commercial application is a safety system that uses a camera to monitor the driver’s eye motions along with a series of cameras or sensors that detect external hazards. A chime, beep, or flash of light will alert the driver both to refocus on the road and when it detects a hazard.

Park acknowledges that drivers frequently encounter distractions while on the road.

“Performance of in-vehicle, secondary tasks require a driver to alternate their glances between the inside of the vehicle and the forward roadway,” he said. “While previous research has shown that thresholds of off-road and durations are critical to latent hazard detection, there has been no research to predict the probability of hazard detection in a time series considering all possible forward glance durations.”

The patented system is the result of a study of 45 subjects testing eight computer-based scenarios in which they glanced at objects or other distractions both inside and outside of the vehicle during a driving simulation. When the tests concluded, Park used machine learning to measure whether the hazard detection likelihoods are correctly predicted per the different scenarios tested.

“A microbenchmark approach based on Hidden Markov models is explored to infer the transition probability of hazard detection that changes dynamically between the detection and non-detection stages. The model is cross-validated and demonstrated to be accurate and robust,” said Park.

This technology can be an important contribution to the automotive industry as it increasingly employs mixed-autonomous safety features.

Written by Alexander Saunders

End of content

No more pages to load