US EditionPublished • loading... • Updated
Brain Activity of Passengers Might Aid Self-Driving Vehicles
The study found integrating passenger brain data improved safety, comfort, and learning speed of self-driving cars using a deep reinforcement learning algorithm.
- In a study published in Cyborg and Bionic Systems, researchers built a system that combines passenger brain data with self-driving systems to improve safety.
- Amid concerns that current systems can struggle in fast-changing situations, researchers turned to functional Near-Infrared Spectroscopy as it can provide cognitive inputs tied to risk perception and emotions to enhance autonomous driving safety.
- Using noninvasive fNIRS to track passenger brain signals, the team fed data into a deep reinforcement learning–based algorithm that shifted driving modes when unease was detected, outperforming traditional software.
- Zhang said the team acknowledged limits and called for further validation, while future research will aim to validate the algorithm in more complex driving scenarios.
- Given those limits, the driving scenarios tested were simple and study participants came from a narrow demographic, while the Union of Concerned Scientists offers more self-driving safety context.
Insights by Ground AI
18 Articles
18 Articles
Coverage Details
Total News Sources18
Leaning Left0Leaning Right5Center8Last UpdatedBias Distribution62% Center
Bias Distribution
- 62% of the sources are Center
62% Center
C 62%
R 38%
Factuality
To view factuality data please Upgrade to Premium
