研究目的
To propose a deep learning assisted VLC unmanned vehicles system that can automatically capture and identify different traffic signs to reduce the incidence of traffic accidents.
研究成果
The proposed deep learning assisted VLC unmanned vehicle system is feasible and accurate for identifying traffic signs and controlling vehicle status, as verified by theoretical analysis and a demo system.
研究不足
The paper does not explicitly mention the limitations of the research.
The methodology includes the design of a deep learning assisted VLC unmanned vehicles system. The system captures traffic signs using an in-built camera, identifies them using a deep learning algorithm implemented in a Micro Control Unit (MCU), and controls the vehicle's motion based on the identified signs. The demo system consists of three modules: a traffic sign detection module, a VLC message sending module, and a VLC message receiving module, implemented using Raspberry pi 3Bs as MCUs.
独家科研数据包,助您复现前沿成果,加速创新突破
获取完整内容