English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This paper uses supervised learning, random search and deep reinforcementlearning (DRL) methods to control large signalized intersection networks. Thetraffic model is Cellular Automaton rule 184, which has been shown to be aparameter-free representation of traffic flow, and is the most efficientimplementation of the Kinematic Wave model with triangular fundamental diagram.We are interested in the steady-state performance of the system, both spatiallyand temporally: we consider a homogeneous grid network inscribed on a torus,which makes the network boundary-free, and drivers choose random routes. As abenchmark we use the longest-queue-first (LQF) greedy algorithm. We find that:(i) a policy trained with supervised learning with only two examplesoutperforms LQF, (ii) random search is able to generate near-optimal policies,(iii) the prevailing average network occupancy during training is the majordeterminant of the effectiveness of DRL policies. When trained under free-flowconditions one obtains DRL policies that are optimal for all trafficconditions, but this performance deteriorates as the occupancy during trainingincreases. For occupancies > 75% during training, DRL policies perform verypoorly for all traffic conditions, which means that DRL methods cannot learnunder highly congested conditions. We conjecture that DRL's inability to learnunder congestion might be explained by a property of urban networks found here,whereby even a very bad policy produces an intersection throughput higher thandownstream capacity. This means that the actual throughput tends to beindependent of the policy. Our findings imply that it is advisable for currentDRL methods in the literature to discard any congested data when training, andthat doing this will improve their performance under all traffic conditions.

Large-scale traffic signal control using machine learning: some traffic flow considerations