AUC Researchers Explore the Ethics, Trust and Future of Self-Driving Cars

As autonomous vehicles move closer to be part of everyday life, researchers at the American University in Cairo (AUC) are investigating one of the technology’s concerning challenges: how autonomous cars can make safe, ethical decisions while earning the trust of people who use them.

The research, led by Amr El Mougy, associate professor in AUC’s Department of Computer Science and Engineering, was presented during the University’s Faculty at the Forefront media roundtable. The discussion highlighted the advances in autonomous vehicle technology, the ethical dilemma facing the artificial intelligence systems. The team works on research project titled Ethical, Trustworthy, Autonomous: The Vehicles of Tomorrow, which focuses on how autonomous vehicle perceive the environment, detect the surroundings, make decisions and respond to real-life situations.

"Whenever I give a seminar about autonomous driving, I usually ask the audience who would be willing to ride in an autonomous vehicle," El Mougy said. "There's always at least a couple of people who say they would never get into one, and their response is almost always related to trust." However, this number is gradually decreasing and more people started to build trust.

Trust was a recurring theme during the event. A fellow professor emphasized on the importance of the human factor to achieve accountability. The discussion referenced back to 2018 fatal incident involving an autonomous Uber test vehicle in the United States. It sparked a global debate due to the difficulty to determine liability and responsivity.

She added that technological advancement alone is not enough to guarantee public acceptance and trust. She added that Egypt is actively and intensively working on strengthening the safety and reliability of autonomous vehicles, particularly as the technology moves from research environments toward real-world testing.

The project also explores the role of reinforcement learning, a branch of artificial intelligence that enables systems to learn through rewards and consequences. Researchers are investigating how this technology can help autonomous vehicles make safer and more ethical decisions under varying conditions.

“If sensors are the eyes and ears of autonomous driving, algorithms are the brain,” El Mougy said.

Beyond vehicle performance, the research examines how passengers respond when autonomous systems make mistakes or encounter uncertainty. The team is also studying the feasibility of introducing a “black box” system for autonomous vehicles, similar to those used in aviation, to provide transparency and accountability by recording how critical decisions are made.

The research explains how autonomous systems interpret human behavior. One area of focus is pedestrian intention prediction, which aims to help vehicles anticipate the actions of people nearby. Researchers use pedestrian joint detection and tracking to analyze movement patterns, estimate walking direction, velocity, position, head orientation, and body posture. The system also calculates statistical measurements based on the distances between detected body joints to improve predictions of pedestrian behavior.

Despite advances in sensing and artificial intelligence, El Mougy noted that autonomous systems still face significant challenges. Researchers must account for numerous uncertainties, including changing weather conditions, terrain variations, sensor inaccuracies, and the unpredictable movement of vehicles and pedestrians. Such factors can lead to object misclassification and inaccurate trajectory predictions, making safe navigation and path planning more difficult.

To address these challenges, El Mougy and his team are developing mathematical models that help autonomous systems make better decisions under uncertain conditions. The research also incorporates reinforcement learning, a form of artificial intelligence inspired by the way humans and animals learn from rewards and consequences.

The multidisciplinary group includes seven engineer-research assistants working alongside El Mougy. Team members specialize in areas ranging from mathematical modeling and hypothesis development to mobile application design, cybersecurity, and autonomous vehicle implementation.

Following the roundtable, attendees toured AUC's Connected Autonomous Vehicles Laboratory (CAVLab), a research platform established by El Mougy in 2023 to advance research in connected and autonomous transportation systems. The team currently operates an autonomous golf cart while they plan to have three connected vehicles in the near future.

During the demonstration, attendees observed a golf cart being operated remotely through a mobile application developed by the research team. The application sends signals to the vehicle over an internet connection, allowing users to control its movement in real time. The team believes that and autonomous vehicle technologies are the future of transportation systems.

Attendees were also invited to interact directly with the technology, taking turns driving the golf cart and riding as passengers while researchers explained how the system processes information and makes decisions.

The vehicles are equipped with a range of advanced sensing technologies, including RGB cameras, stereo cameras, LiDAR systems, radars, thermal cameras and GPS/INS units. Together, these technologies create what researchers describe as a multi-modal perception system, enabling vehicles to collect and interpret information from multiple sources simultaneously.

The research received $300,000 in funding in 2025 through the African Engineering and Technology Network (Afretec), a partnership among nine universities aimed at accelerating digital innovation and technological research across Africa.

While fully autonomous vehicles are still under development, El Mougy further explained that future autonomous vehicle is expected to be introduced in Egypt with a gradual rollout strategy, starting with highways and controlled roads that can be a potential early testing and deployment zone. He gives examples of roads such as the Ain Sokhna Highway, with its wide multi-lane structure and relatively predictable traffic flow.

For El Mougy and his team, the challenge is not simply building smarter vehicles. It is ensuring that these vehicles can make decisions that are safe, transparent and worthy of public confidence.