Intel International Science and Engineering Fair (ISEF)

The IEEE Computer Society continued its longstanding participation in Intel’s ISEF,  the world’s largest precollege science fair.  Young scientists showcased their innovative projects, networked with fellow students and professionals, and competed for more than $3 million in awards and scholarships.

A team of Computer Society volunteer judges, led by chief judge Lowell Johnson, evaluated computer science and engineering projects over the course of two days.  The judges selected three individual and two team projects for Computer Society recognition.  Jessica Alexis Richeri, 17 from Centennial High School, Corona, CA won the $1,000 IEEE Computer Society top individual prize for his timely project “Autonomous Robotic Vehicle, Saving Lives, Preventing Accidents One at a Time". This project integrated hardware with algorithms for automated driving of a vehicle using visual sensory input.  She demonstrated integration of several image pattern recognition algorithms to detect vehicles and the lane markings on a road, one of which she devised herself using a database to store images of the rear ends of a large number of vehicle types.  Richeri demonstrated the algorithms in a field trial in a human-driven car on local streets.  She also demonstrated the integration of the lane marking detection algorithm together with a steering algorithm in a small robotic wheeled vehicle she built, then tested on her high school's sports track.  Richeri also received the IEEE Foundation Presidents' Scholarship Award of $10,000, payable over four years for undergraduate study in engineering or a related field.

The Computer Society’s Team First Award was awarded to Scotty Allan Chung, Yajit Kumar Jair, and Carlos E del-Castillo-Negrete, all from Oak Ridge High School, TN for their project "A Genetic Algorithm Approach to Minimizing Beam Loss in High Power Particle Accelerators".  This project improved beam loss by an automated control strategy to adjust the pairs of quadrupole magnets that focus beams of charged particles in an accelerator by using genetic algorithms. The team characterized the control solution space, putting it into a form to which they could apply genetic algorithms.  They tested their control strategy on a simulator provided to them by their mentors at ORNL and refined their selection process.  Finally, when they had come up with a promising control strategy, they tried it on the actual particle accelerator.  They were able to match or exceed the performance of human operators in minimizing beam loss, and more importantly, they demonstrated the feasibility of their approach for generating such strategies.