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

Active Traffic Management strategies are often adopted in real-time toaddress such sudden flow breakdowns. When queuing is imminent, SpeedHarmonization (SH), which adjusts speeds in upstream traffic to mitigatetraffic showckwaves downstream, can be applied. However, because SH depends ondriver awareness and compliance, it may not always be effective in mitigatingcongestion. The use of multiagent reinforcement learning for collaborativelearning, is a promising solution to this challenge. By incorporating thistechnique in the control algorithms of connected and autonomous vehicle (CAV),it may be possible to train the CAVs to make joint decisions that can mitigatehighway bottleneck congestion without human driver compliance to altered speedlimits. In this regard, we present an RL-based multi-agent CAV control model tooperate in mixed traffic (both CAVs and human-driven vehicles (HDVs)). Theresults suggest that even at CAV percent share of corridor traffic as low as10%, CAVs can significantly mitigate bottlenecks in highway traffic. Anotherobjective was to assess the efficacy of the RL-based controller vis-\`a-visthat of the rule-based controller. In addressing this objective, we dulyrecognize that one of the main challenges of RL-based CAV controllers is thevariety and complexity of inputs that exist in the real world, such as theinformation provided to the CAV by other connected entities and sensedinformation. These translate as dynamic length inputs which are difficult toprocess and learn from. For this reason, we propose the use of GraphicalConvolution Networks (GCN), a specific RL technique, to preserve informationnetwork topology and corresponding dynamic length inputs. We then use this,combined with Deep Deterministic Policy Gradient (DDPG), to carry outmulti-agent training for congestion mitigation using the CAV controllers.

Leveraging the Capabilities of Connected and Autonomous Vehicles and Multi-Agent Reinforcement Learning to Mitigate Highway Bottleneck Congestion