A pair of University of Texas at Arlington researchers are teaming with Lockheed Martin Corp. to optimize how unmanned vehicles use sensors to gather information. Frank Lewis, the Moncrief-O’Donnell Chair and professor of electrical engineering, and Yan Wan, an associate professor of electrical engineering, are helping the company determine how to use sensors on vehicles that are underwater, on the ground and in the air to measure and predict the activities of a target.
Moving targets and environmental factors–such as water, air clarity or visibility–may interfere with sensor signals. Therefore, the team is using a Kalman filter to gather information from multiple sensors, then fuse it together from different angles to estimate a target’s location, velocity, physical properties and other information.
When uncertainty exists in a dynamic system, a Kalman filter is an algorithm that can be used to combine information to make an educated guess about what a system will do next. The algorithm allows operators to adapt to changing missions and environmental conditions to produce an increased effective field of view, improved sensor coverage and more accurate target estimation.
The researchers will use the $50,000 contract with Lockheed to ensure the algorithm will work with the company’s software.
“In distributed control systems, where multiple unmanned vehicles work together and receive commands based on their relative position to each other and the desired outcome or goal of the task, the ability to quickly and accurately predict behaviors is a vital component of a satisfactory outcome for the mission,” Wan said. “We are happy to apply our knowledge of distributed controls in collaboration with Lockheed Martin because it will help them develop better platforms for the technology and it will help us better understand the potential applications of our research.”
Lewis and Wan’s research is an example of data-driven discovery, one of four themes of UTA’s Strategic Plan 2020. It also shows how College of Engineering researchers are collaborating with industry to help with real-world research and development efforts.
“Lockheed Martin employs hundreds of our engineering alumni and has been one of our valued partners for many years,” said Peter Crouch, dean of the College of Engineering. “As a leading research university, we constantly seek collaborations between our faculty and industry to stay abreast of industry needs and trends.”
Lewis is a world-renowned expert in distributed controls and sensors, with more than $10 million in research funding in his career. His key contribution has been to add additional self-learning mechanisms to neural network controllers for dynamical systems, such as aircraft and robots, by making it possible to design and tune the neural adaptive controllers based on the patterns of use of different modes in the system. This tuning process provides greater reliability and stability for neural adaptive control.
Wan has a $442,000 National Science Foundation CAREER award to develop an innovative theoretical framework for cyber-physical systems that enable airborne networking, direct flight-to-flight communication for flexible information sharing, safe maneuvering and coordination of time-critical missions. She also is the team lead on a $998,803 grant from the National Science Foundation that will develop a networked airborne computing platform for multiple unmanned aerial systems.
Source: Press Release
UAV DACH: Beitrag im Original auf https://www.uasvision.com/2019/08/05/university-of-texas-and-lockheed-martin-team-on-unmanned-vehicle-project/, mit freundlicher Genehmigung von UAS Vision automatisch importiert. Der Beitrag gibt nicht unbedingt die Meinung oder Position des UAV DACH e.V. wieder. Das Original ist in englischer Sprache.