Fuell: Artificial Intelligence and camera-based driver drowsiness detection and warning system

Fuell：人工智能和基于摄像头的驾驶员困倦检测和警告系统

• 批准号: 99674
• 金额: $ 24.25万
• 依托单位: FUELL LIMITED
• 依托单位国家: 英国
• 项目类别: Collaborative R&D
• 财政年份: 2021
• 资助国家: 英国
• 起止时间: 2021 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=99674
• 关键词: Fuell; Artificial; Intelligence; camera; based

**_Fuell -- Artificial Intelligence and camera-based driver drowsiness detection and warning system_**Driver fatigue is a global killer, costing billions in lost time, efficiency, insurance costs and lives.Fuell's objective is to develop a fatigue management system based on physiological detection using an in vehicle camera or a driver's smartphone camera that detects the onset of drowsiness and warns drivers to take action.This project will build on an existing Fuell algorithm-based system that uses Heart Rate Variability (HRV) derived from the Photoplethysmography (PPG) sensor on a Garmin smartwatch to continuously measure fatigue and warn the driver up to 20 minutes in advance of a likely sleep event._This project will use a camera, such as a standard driver-facing in-cab camera or a smart phone camera, to detect vital signs, matching the accuracy of the current wearable-based Fuell system. Critically, this will make it more readily commercial by reducing the cost per driver of the system._In this project, we will take existing rPPG knowledge and adapt Fuell algorithms to create a system that will deliver the detection of drowsiness at a commercially exploitable cost.The Fuell fatigue management system will ultimately apply to any sector where operator fatigue can have an impact on safety, efficiency and health, not just drivers.Worker fatigue is a global problem and the Fuell fatigue management system will be applicable world-wide.

** *驾驶疲劳是一个全球性的杀手，造成了数十亿美元的时间损失、效率损失、保险成本损失和生命损失。Fuell的目标是开发一种基于生理检测的疲劳管理系统，该系统使用车载摄像头或驾驶员的智能手机摄像头来检测睡意的开始，并警告驾驶员采取行动。该项目将建立在现有的基于fuwell算法的系统基础上，该系统使用Garmin智能手表上的Photoplethysmography （PPG）传感器产生的心率变异性（HRV）来持续测量疲劳，并提前20分钟警告驾驶员可能发生的睡眠事件。该项目将使用一个摄像头，如标准的面向驾驶员的驾驶室摄像头或智能手机摄像头，来检测生命体征，与当前基于可穿戴设备的Fuell系统的精度相匹配。至关重要的是，这将通过降低系统的每个驾驶员的成本，使其更容易商业化。在这个项目中，我们将利用现有的rPPG知识和Fuell算法来创建一个系统，该系统将以商业上可开发的成本提供困倦检测。fuwell疲劳管理系统最终将适用于任何操作员疲劳会影响安全、效率和健康的行业，而不仅仅是司机。工人疲劳是一个全球性的问题，燃料疲劳管理系统将在世界范围内适用。