The rapid growth of the ride-sharing industry has brought many benefits but also introduced significant safety challenges. To address this, we developed TransformerYOLO, a new AI system that enhances passenger safety by combining two powerful technologies: YOLO, known for its speed in detecting objects, and transformers, recognized for their ability to understand complex interactions. Our goal is to create a system that not only quickly spots potential threats but also understands what’s happening inside a vehicle. Traditional object detection models often struggle in cars due to factors like changing light and limited space. Our design prioritizes privacy by focusing on detecting objects and actions rather than identifying individuals. The system also includes a real-time alert mechanism to immediately detect and respond to threats. Through this research, we aim to set a new standard for safety in the ride-sharing sector, improving user trust and industry practices. TransformerYOLO’s ability to assess and understand the in-vehicle environment enhances the safety experience for both passengers and drivers. Our system’s adaptability and privacy-conscious design make it a promising solution for widespread use, paving the way for safer and more reliable urban transportation.
A Multi-Modal Transformer Optimized Object Identification Model for Passenger Safety in Ridesharing Vehicles
2024-11-23
373591 byte
Conference paper
Electronic Resource
English
A general equilibrium model for multi-passenger ridesharing systems with stable matching
ArXiv | 2023
|Passenger‐perception dynamic ridesharing service based on parallel technology
Wiley | 2023
|A System for Multi-Passenger Urban Ridesharing Recommendations with Ordered Multiple Stops
BASE | 2020
|