Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Needs of Autonomous Vehicles for Safe Operation on Existing Highways
Autonomous vehicles (AVs) will gradually supersede human-driven vehicles (HDVs) in the future. Unlike human drivers who perceive their surroundings with their eyes, AV senses the ambient environments based on sensor fusion. More specifically, light detection and ranging (Lidar) sensors, video cameras, and radars are combined to help AV understand their surroundings. Current highway geometric design elements are mainly based on human driver-related parameters, such as perception and reaction time (PRT). However, AV will mix with HDV on existing highway infrastructures during the transition period. This study focuses on the sight distance aspect of highway geometric design. It is unsafe if an AV cannot effectively detect the objects within its required sight distance using the fused sensing system. Therefore, determining the needed sensor configurations (e.g. height, effective range, and field of view) is necessary for safe AV operation. This study used a simulation approach to determine the needed sensor configurations. First, the required stopping, decision, and passing sight distances for AV were determined considering that AV has a much shorter PRT than HDV. Second, the automated driving toolbox of MATLAB was used to construct different scenarios, each involving the road, obstacles, and actors. The road models were created following current design guides. A virtual AV equipped with a Lidar, cameras, radars, and an impeding agent served as the main actors on each road model. Third, many simulations were conducted with different sensor configurations to determine the Lidar configuration that achieves 100% detection for the safe operation of AV on existing highways.
Needs of Autonomous Vehicles for Safe Operation on Existing Highways
Autonomous vehicles (AVs) will gradually supersede human-driven vehicles (HDVs) in the future. Unlike human drivers who perceive their surroundings with their eyes, AV senses the ambient environments based on sensor fusion. More specifically, light detection and ranging (Lidar) sensors, video cameras, and radars are combined to help AV understand their surroundings. Current highway geometric design elements are mainly based on human driver-related parameters, such as perception and reaction time (PRT). However, AV will mix with HDV on existing highway infrastructures during the transition period. This study focuses on the sight distance aspect of highway geometric design. It is unsafe if an AV cannot effectively detect the objects within its required sight distance using the fused sensing system. Therefore, determining the needed sensor configurations (e.g. height, effective range, and field of view) is necessary for safe AV operation. This study used a simulation approach to determine the needed sensor configurations. First, the required stopping, decision, and passing sight distances for AV were determined considering that AV has a much shorter PRT than HDV. Second, the automated driving toolbox of MATLAB was used to construct different scenarios, each involving the road, obstacles, and actors. The road models were created following current design guides. A virtual AV equipped with a Lidar, cameras, radars, and an impeding agent served as the main actors on each road model. Third, many simulations were conducted with different sensor configurations to determine the Lidar configuration that achieves 100% detection for the safe operation of AV on existing highways.
Needs of Autonomous Vehicles for Safe Operation on Existing Highways
Lecture Notes in Civil Engineering
Gupta, Rishi (Herausgeber:in) / Sun, Min (Herausgeber:in) / Brzev, Svetlana (Herausgeber:in) / Alam, M. Shahria (Herausgeber:in) / Ng, Kelvin Tsun Wai (Herausgeber:in) / Li, Jianbing (Herausgeber:in) / El Damatty, Ashraf (Herausgeber:in) / Lim, Clark (Herausgeber:in) / Easa, Said M. (Autor:in) / Ma, Yang (Autor:in)
Canadian Society of Civil Engineering Annual Conference ; 2022 ; Whistler, BC, BC, Canada
Proceedings of the Canadian Society of Civil Engineering Annual Conference 2022 ; Kapitel: 42 ; 631-642
06.02.2024
12 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Reinforcing and widening existing highways
| Engineering Index Backfile | 1940
British Library Online Contents | 2008
British Library Online Contents | 2012