United States Department of Transportation
i
Smart Connected and Automated Vehicle Fleet Management: Developing Regional Dispatch Decision Support for Congestion Mitigation
-
2023-08-01
-
[PDF-10.97 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:2/1/2020 – 2/28/2023
-
Corporate Publisher:
-
Abstract:With the arrival of new technologies like connected and self-driving autonomous vehicles (AVs), the workload of regional dispatchers will increase. To this end, the CADS (Congestion Alerting Decision Support) tool was developed to support strategic transportation planning (on the order of weeks to months) and tactical transportation planning (on the order of hours to days). It allows dispatchers and other decision makers the ability to run what-if simulations to determine how and when to allocate future resources for different scenarios. Such a tool can be used operationally, like in the days before a hurricane to improve planning, but also in training to provide a host of practice scenarios for new planners, dispatchers and other emergency planning personnel. One current limitation of CADS is its inability to connect congestion metrics to predictions of safety, which could be very useful to dispatchers, especially as AVs increase in numbers. Research determined that when operating in mixed traffic as a following vehicle in a platooning scenario, AVs had longer response times, which could lead to an increase in the number and severity of conflicts, and thus problems with potential overall safety. If CADS could track metrics like AV platoon length, heterogeneity of platoon vehicles on a highway and braking response times, it is possible that CADS could highlight risk profiles for areas of congestion that involve AVs. However, additional research showed that incorporating commonly-used models likely do well in predicting where AVs will be at a given time, but struggle with predicting acceleration and deceleration, indicating more work is needed before such simulations can realistically be used for risk projections.
-
Format:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
