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

Deep reinforcement learning (DRL) has a great potential for solving complexdecision-making problems in autonomous driving, especially in mixed-trafficscenarios where autonomous vehicles and human-driven vehicles (HDVs) drivetogether. Safety is a key during both the learning and deploying reinforcementlearning (RL) algorithms process. In this paper, we formulate the on-rampmerging as a Markov Decision Process (MDP) problem and solve it with anoff-policy RL algorithm, i.e., Soft Actor-Critic for Discrete Action Settings(SAC-Discrete). In addition, a motion predictive safety controller including amotion predictor and an action substitution module, is proposed to ensuredriving safety during both training and testing. The motion predictor estimatesthe trajectories of the ego vehicle and surrounding vehicles from kinematicmodels, and predicts potential collisions. The action substitution moduleupdates the actions based on safety distance and replaces risky actions, beforesending them to the low-level controller. We train, evaluate and test ourapproach on a gym-like highway simulation with three different levels oftraffic modes. The simulation results show that even in harder trafficdensities, our proposed method still significantly reduces collision rate whilemaintaining high efficiency, outperforming several state-of-the-art baselinesin the considered on-ramp merging scenarios. The video demo of the evaluationprocess can be found at: https://www.youtube.com/watch?v=7FvjbAM4oFw

Autonomous Highway Merging in Mixed Traffic Using Reinforcement Learning and Motion Predictive Safety Controller