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

Car-following models have made significant contributions to our understandingof longitudinal driving behavior. However, they often exhibit limited accuracyand flexibility, as they cannot fully capture the complexity inherent incar-following processes, or may falter in unseen scenarios due to theirreliance on confined driving skills present in training data. It is worthnoting that each car-following model possesses its own strengths and weaknessesdepending on specific driving scenarios. Therefore, we proposeEnsembleFollower, a hierarchical planning framework for achieving advancedhuman-like car-following. The EnsembleFollower framework involves a high-levelReinforcement Learning-based agent responsible for judiciously managingmultiple low-level car-following models according to the current state, eitherby selecting an appropriate low-level model to perform an action or byallocating different weights across all low-level components. Moreover, wepropose a jerk-constrained kinematic model for more convincing car-followingsimulations. We evaluate the proposed method based on real-world driving datafrom the HighD dataset. The experimental results illustrate thatEnsembleFollower yields improved accuracy of human-like behavior and achieveseffectiveness in combining hybrid models, demonstrating that our proposedframework can handle diverse car-following conditions by leveraging thestrengths of various low-level models.

EnsembleFollower: A Hybrid Car-Following Framework Based On Reinforcement Learning and Hierarchical Planning