Engineers at Hong Kong University of Science and Technology developed a cognitive encoding framework for autonomous vehicles (AVs) to think like human drivers.
The new technology helps AVs make smarter, safer decisions in complex traffic situations, improving road safety for all users.
It reduces overall traffic risk by 26% and cuts danger to vulnerable road users like pedestrians and cyclists by over 50%.
Cars using this system experienced an 8% drop in risk and finished routes nearly 14% faster, enhancing efficiency.
The system mimics human-like thinking by considering multiple factors simultaneously, known as 'social sensitivity.'
It assesses individual risks, prioritizes safety for the most vulnerable, and predicts how actions may impact others.
The technology underwent testing in 2,000 traffic simulations, showing improved safety and enhanced efficiency in driving.
It is adaptable to local driving rules and cultural norms, making it suitable for global implementation.
Collaboration with top universities in China and the U.S. aims to enhance the system's intelligence and adaptability further.
The next phase involves building a diverse database of driving behavior from different countries.
With this progress, self-driving cars are becoming more responsible and thoughtful on the road, akin to human drivers.