Semantic 5G Vehicle Communication for Safe Cooperative Autonomous Driving

用于安全协同自动驾驶的语义 5G 车辆通信

• 批准号: EP/Y027787/1
• 负责人: Jianhua He
• 金额: $ 25.55万
• 依托单位: University of Essex
• 依托单位国家: 英国
• 项目类别: Fellowship
• 财政年份: 2023
• 资助国家: 英国
• 起止时间: 2023 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FY027787%2F1
• 关键词: Semantic; 5G; Vehicle; Communication; Safe

Road travel remains one of the most hazardous activities people regularly undertake. More than 1 million people lost their lives on the roads globally due to road accidents every year. Enabled by the autonomous vehicles and vehicle to everything (V2X) communications, cooperative autonomous driving (CAD) is one of the most promising driving applications to reduce road fatalities. Through sharing sensing data and driving intentions, CAD can benefit from earlier and more comprehensive perception of driving environment and hazards, cooperate on trajectory planning and manoeuvre for safe and efficient driving. While CAD holds great potentials for improving safety, its ultra-reliability and high data rate requirements present critical challenges to 5G V2X networks in distributed and ad hoc environments. In this project SECOM original application oriented sematic 5G V2X with efficient architecture and enabling schemes is proposed to tackle the challenges. It enables transmission of application relevant, meaningful and efficiently represented data to the intended receivers with use of semantic context and cross-layer design. It can significantly improve the efficiency of 5G V2X communication and improve the CAD driving safety. SECOM is innovative and will rely on the host and the Fellow's complementary and multi-disciplinary expertise in 5G V2X, semantic segmentation and sensing, autonomous driving and system optimization to tackle the key challenges. The Fellow will receive intensive training in multidisciplinary technical fields and transferable skills from the host and associated partners. Through cutting-edge research, this project will produce original semantic 5G V2X schemes and reliable solutions for CAD, make scientific breakthroughs, improve road safety, enhance the Fellow's career prospects, strengthen knowledge transfer and boost EU research and innovation capacity.

公路旅行仍然是人们经常从事的最危险的活动之一。全球每年有100多万人在道路交通事故中丧生。在自动驾驶车辆和车辆对一切(V2X)通信的支持下，协作式自动驾驶(CAD)是最有前途的减少道路死亡的驾驶应用之一。通过共享传感数据和驾驶意图，CAD可以更早和更全面地感知驾驶环境和危险，合作进行轨迹规划和操纵，以实现安全和高效的驾驶。虽然CAD在提高安全性方面具有巨大的潜力，但其超可靠性和高数据速率要求对分布式和自组织环境中的5G V2X网络提出了严峻的挑战。在本项目中，SECOM提出了面向应用的语义5G V2X，具有高效的架构和使能方案来应对挑战。通过使用语义上下文和跨层设计，它能够将与应用相关的、有意义的和有效表示的数据传输到预期的接收者。它可以显著提高5G V2X通信的效率，提高CAD驾驶安全。赛科姆具有创新性，将依靠东道主和研究员在5G V2X、语义分割和传感、自动驾驶和系统优化方面的互补和多学科专业知识来应对关键挑战。该研究员将从东道主和相关合作伙伴那里接受多学科技术领域和可转让技能的强化培训。该项目将通过前沿研究，为CAD产生原创的语义5G V2X方案和可靠的解决方案，取得科学突破，改善道路安全，增强研究员的职业前景，加强知识转移，提升欧盟的研究和创新能力。