CCRI: MEDIUM: Collaborative Research: Community Platforms for Safe, Agile, and Coordinated Autonomy

CCRI：中：协作研究：安全、敏捷和协调自治的社区平台

• 批准号: 1925587
• 负责人: Rahul Mangharam
• 金额: $ 58.05万
• 依托单位: University of Pennsylvania
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1925587&HistoricalAwards=false
• 关键词: CCRI; MEDIUM; Collaborative; Research; Community

F1/10 is a shared, open-sourced infrastructure for the development and validation of new approaches to autonomous perception, planning, control and coordination. This community platform will facilitate autonomous system research, education and bench-marking through the creation of a new class of high-performance autonomous racing cars, that are 1/10th the size of a real (Formula 1) car and can reach a top speed of 50mph. The goal is to enable a wide range of experimental research and education in Safe, Secure, Coordinated and Efficient Autonomy. Much of today's research on Autonomous vehicles (AVs) is limited to experimentation on expensive commercial vehicles that require large teams with diverse skills and power-hungry platforms. Testing the limits of safety and performance on such vehicles is costly and hazardous. It is also outside the reach of most academic departments and research groups. Furthermore, little research has been devoted to the development of safety benchmarks and infrastructure for certifying autonomous systems, guaranteeing the safety of data-driven decision making, power-efficient perception and control for efficient autonomy, active coordinated safety for large vehicle fleets, and cyber-physical security of autonomous vehicles. All of these are fundamental requirements on the road to achieving the social benefits of autonomous vehicles, and autonomous systems more generally. F1/10 provides rapid prototyping hardware, software and algorithmic platforms, with full documentation and community support for research and development of future autonomous systems. Autonomous vehicles are emerging as integral components of the US economy, and in the future, they may be at the heart of the national economy's competitiveness in transportation and logistics. Enabling such a transformation requires building reliably safe and efficient autonomous vehicles and the contributions of a very diverse community of researchers from engineering, human factors and computer science. The project's intellectual merit resides in the design and deployment of scaled autonomous cars that are affordable, safe to experiment with, easy to customize, can be shared widely, and that can support a wide range of necessary research in autonomous vehicles. This project's broader goal is to stimulate and accelerate cross-disciplinary research in autonomous vehicle safety and efficiency, at a time when such work is urgently needed. This project will enable researchers in autonomous hardware, safety, security, and traffic to produce high-confidence autonomous systems.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

F1/10是一个共享的开源基础设施，用于开发和验证自主感知、规划、控制和协调的新方法。这个社区平台将通过创建一种新的高性能自动驾驶赛车汽车来促进自动驾驶系统的研究，教育和基准测试，这些赛车的尺寸是真实的（F1）汽车的十分之一，最高速度可达50英里/小时。其目标是在安全，可靠，协调和有效的自治方面进行广泛的实验研究和教育。如今，大部分关于自动驾驶汽车（AV）的研究都局限于昂贵的商用车上的实验，这些商用车需要拥有不同技能和耗电平台的大型团队。测试这些车辆的安全和性能极限是昂贵和危险的。这也是大多数学术部门和研究小组所无法企及的。此外，很少有研究致力于开发用于认证自动驾驶系统的安全基准和基础设施，保证数据驱动决策的安全性，高效自主的节能感知和控制，大型车队的主动协调安全以及自动驾驶车辆的网络物理安全。所有这些都是实现自动驾驶汽车和更广泛的自动驾驶系统社会效益的基本要求。F1/10提供快速原型硬件、软件和算法平台，并为未来自主系统的研发提供完整的文档和社区支持。自动驾驶汽车正在成为美国经济不可或缺的组成部分，在未来，它们可能是国民经济在运输和物流方面竞争力的核心。实现这种转变需要建造可靠、安全、高效的自动驾驶汽车，以及来自工程、人为因素和计算机科学的各种研究人员的贡献。该项目的智力价值在于设计和部署规模化的自动驾驶汽车，这些汽车价格实惠，实验安全，易于定制，可以广泛共享，并且可以支持自动驾驶汽车的广泛必要研究。该项目更广泛的目标是刺激和加速自动驾驶汽车安全性和效率的跨学科研究，目前迫切需要这样的工作。该项目将使自主硬件、安全、安保和交通领域的研究人员能够生产出高置信度的自主系统。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

You Don't Know When I Will Arrive: Unpredictable Controller Synthesis for Temporal Logic Tasks

你不知道我什么时候到来：时态逻辑任务的不可预测控制器综合

• DOI: 10.1016/j.ifacol.2023.10.1519
• 发表时间: 2023
• 期刊: IFAC-PapersOnLine
• 作者: Chen, Yu;Yang, Shuo;Mangharam, Rahul;Yin, Xiang
• 通讯作者: Yin, Xiang

Deriving Spatial Policies for Overtaking Maneuvers with Autonomous Vehicles

导出自动驾驶车辆超车的空间策略

• DOI: 10.1109/comsnets53615.2022.9668548
• 发表时间: 2022
• 期刊: IEEE International Conference on COMmunication Systems & NETworkS
• 作者: Bhargav, Jayanth;Betz, Johannes;Zehng, Hongrui;Mangharam, Rahul
• 通讯作者: Mangharam, Rahul

Teaching Autonomous Systems at 1/10th-scale: Design of the F1/10 Racecar, Simulators and Curriculum

以 1/10 比例教授自主系统：F1/10 赛车、模拟器和课程的设计

• DOI: 10.1145/3328778.3366796
• 发表时间: 2020
• 期刊: Proceedings of the 51st ACM Technical Symposium on Computer Science Education
• 作者: Agnihotri, Abhijeet;O'Kelly, Matthew;Mangharam, Rahul;Abbas, Houssam
• 通讯作者: Abbas, Houssam

Stress Testing Autonomous Racing Overtake Maneuvers with RRT

使用 RRT 对自动赛车超车动作进行压力测试

• DOI: 10.1109/iv51971.2022.9827237
• 发表时间: 2022
• 期刊: IEEE Intelligent Vehicles Symposium (IV
• 作者: Bak, Stanley;Betz, Johannes;Chawla, Abhinav;Zheng, Hongrui;Mangharam, Rahul
• 通讯作者: Mangharam, Rahul

A Benchmark Comparison of Imitation Learning-based Control Policies for Autonomous Racing

基于模仿学习的自动赛车控制策略的基准比较

• DOI: 10.1109/iv55152.2023.10186780
• 发表时间: 2023
• 期刊: IEEE Intelligent Vehicles Symposium
• 作者: Sun, Xiatao;Zhou, Mingyan;Zhuang, Zhijun;Yang, Shuo;Betz, Johannes;Mangharam, Rahul
• 通讯作者: Mangharam, Rahul