Multidisciplinary Senior Design Team Showcases Remarkable Innovation at RTX Autonomous Vehicle Competition


October 10, 2024

Members of the CS-MAE team preparing for the competition at Virginia Tech

The buzz of drones and the hum of innovation filled the air at the Virginia Tech Drone Park as student teams from various East Coast universities participated in the 2024 Autonomous Vehicle Competition (AVC) University Innovation Showcase this past spring. Representing GW Engineering was a multidisciplinary senior design capstone team from the Departments of Computer Science and Mechanical and Aerospace Engineering (CS-MAE), whose ingenuity at the competition earned them the Innovation and Engineering on the Fly Award!

Hosted by RTX, formerly Raytheon, AVC’s mission is to bolster undergraduate experiential learning programs at participating universities by teaching students valuable technical, project management, and problem-solving skills. The company hosts multiple competitions by region, with this section including George Mason University, Howard University, and the University of South Florida, in addition to GW and Virginia Tech.

To participate, students were challenged to research, design, develop, integrate, and test unmanned vehicle system hardware and software components. For AVC 2024, each team was tasked with constructing two autonomous vehicles, one for aerial operations and one for ground maneuvers. The objective of the air vehicles was to detect and track a target, which was the ground vehicle. The CS-MAE team split into two teams to tackle these tasks.

During development, recent mechanical engineering graduate Ryan Rafati said that they learned countless industry-applicable project management skills from budget allocation to submitting material orders. The testing of their unmanned aerial vehicle (UAV) posed significant hurdles due to airspace restrictions over Washington, D.C.

“The District of Columbia essentially has a no-fly zone around a 15-mile radius. This severely hindered our ability to test our UAV. Our team then turned to flying at the Virginia Science and Technology campus and the University of Maryland Flight Cage to test our drone,” said Rafati.

Beginning in mid-February, the team traveled on weekends to GW’s VSTC campus, a 60-90 minute trip each way via metro and rideshares. Karl Simon, a recent computer science graduate, noted Virginia provided a place to legally fly their vehicle, better GPS signals and more space to fly (and crash). Later, they moved testing to UMD’s Fearless Flight Factory. Simon says this football field-sized outdoor netted enclosure mimicked the competition conditions the closest, which was crucial in the final phases of development and testing.

“This Capstone project taught our team many important lessons. One of which was time management and organization, as working with a cross-disciplinary team made it hard to find times when all ten of us could meet and see each other’s progress. However, closer to our second semester, we were able to dedicate more time and work closely with one another,” Rafati said.

The competition’s first challenge highlighted the benefits of this interdisciplinary team as it required the team’s UAV to scan the ground for the opposing schools’ motionless unmanned ground vehicles (UGVs). Once detected, the UAV had to navigate towards the UGV and deliver a water blast of 20ml to activate a moisture sensor on the vehicle. This water blast could be seen as similar to tasks such as firefighting or search and rescue. MAE students utilized their expertise to design the deployment system that delivered the water blast, while the CS students developed the algorithm for the UAV to seek out UGVs.

Additional challenges involved the UGV in motion, driving along both known and unknown paths, to simulate a moving target. Simon said the UGVs had aruco markers on top, which are designed to be quickly located by computer vision systems, making it easier for the UAV to identify the ground vehicle.

At Virginia Tech, the UGV faced unexpectedly tall grass, causing the team’s vehicle to sink and struggle to move. The CS-MAE team worked overnight before the competition to adopt a new design that reduces the overall weight and increases the ground clearance of the chassis of the vehicle. Rafati said in their revised design, they rebuilt the frame of the UGV using cardboard in place of 3D-printed material to lighten the load and added new tires to overcome the tall grass.

“These students did an incredible job, overcoming challenges both before and during the competition. They worked well together and showed off impressive creativity and deep technical skills,” said Timothy Wood, an associate professor of computer science who advised the team.

By successfully navigating these obstacles, the GW Engineering team demonstrated remarkable adaptability, which earned them the Innovation and Engineering on the Fly Award. The judges praised the team on the award certificate, stating, “From redesigning a system in hours to accommodate unexpected field conditions to making changes to your code and vehicles during the competition itself to optimize performance, you epitomized the spirit of hard work and innovation that the competition is centered on.”

“This whole experience, from large development challenges throughout the year to last-minute troubleshooting during the competition, taught me the value of teamwork, the importance of planning and testing, and gave me a sense for what real-world engineering looks like in industry. There are client constraints, strict timelines, budgets, and countless technical problems along the way that don’t make engineering projects easy,” said Simon. “However, having people on the team with different skill sets makes problem-solving so much more efficient and effective. This, combined with having a team that shared a deeply rooted motivation and passion for their work, is what led us to success.”

Professor Wood would like to thank Steve Shooter for arranging GW Engineering’s involvement in the competition with RTX and establishing the cross-department teams; Mitch Narrins for mentoring the MAE students; Xiadong Qu and Josh Shapiro for assisting the CS teams; and Jarick Cammarato for serving as the team’s mentor from RTX. Next year, the faculty mentors are excited to involve electrical and computer engineering students, expanding their multidisciplinary approach and allowing more students to experience its benefits.