Safe Operation of Connected Vehicles in Complex and Unforeseen Environments

Hovakimyan, Naira; Kim, Hunmin; Wan, Wenbin; Tao, Chuyuan

Permalink

https://hdl.handle.net/2142/115966 Copy

Description

Title

Safe Operation of Connected Vehicles in Complex and Unforeseen Environments

Author(s)

• Hovakimyan, Naira
• Kim, Hunmin
• Wan, Wenbin
• Tao, Chuyuan

Issue Date

2022-08

Keyword(s)

• Vehicle Operations
• Proactive Control
• Safety
• Path Planning

Abstract

Autonomous vehicles (AVs) have a great potential to transform the way we live and work, significantly reducing traffic accidents and harmful emissions on the one hand and enhancing travel efficiency and fuel economy on the other. Nevertheless, the safe and efficient control of AVs is still challenging because AVs operate in dynamic environments with unforeseen challenges. This project aimed to advance the state-of-the-art by designing a proactive/reactive adaptation and learning architecture for connected vehicles, unifying techniques in spatiotemporal data fusion, machine learning, and robust adaptive control. By leveraging data shared over a cloud network available to all entities, vehicles proactively adapted to new environments on the proactive level, thus coping with large-scale environmental changes. On the reactive level, control-barrier-function-based robust adaptive control with machine learning improved the performance around nominal models, providing performance and control certificates. The proposed research shaped a robust foundation for autonomous driving on cloud-connected highways of the future.

Publisher

Illinois Center for Transportation

Series/Report Name or Number

I-ACT-21-06

Type of Resource

text

Language

en

Copyright and License Information

No restrictions. This document is available through the National Technical Information Service, Springfield, VA 22161.

Owning Collections