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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Autonomous driving at intersections is one of the most complicated andaccident-prone traffic scenarios, especially with mixed traffic participantssuch as vehicles, bicycles and pedestrians. The driving policy should make safedecisions to handle the dynamic traffic conditions and meet the requirements ofon-board computation. However, most of the current researches focuses onsimplified intersections considering only the surrounding vehicles andidealized traffic lights. This paper improves the integrated decision andcontrol framework and develops a learning-based algorithm to deal with complexintersections with mixed traffic flows, which can not only take account ofrealistic characteristics of traffic lights, but also learn a safe policy underdifferent safety constraints. We first consider different velocity models forgreen and red lights in the training process and use a finite state machine tohandle different modes of light transformation. Then we design different typesof distance constraints for vehicles, traffic lights, pedestrians, bicyclesrespectively and formulize the constrained optimal control problems (OCPs) tobe optimized. Finally, reinforcement learning (RL) with value and policynetworks is adopted to solve the series of OCPs. In order to verify the safetyand efficiency of the proposed method, we design a multi-lane intersection withthe existence of large-scale mixed traffic participants and set practicaltraffic light phases. The simulation results indicate that the trained decisionand control policy can well balance safety and tracking performance. Comparedwith model predictive control (MPC), the computational time is three orders ofmagnitude lower.

Integrated Decision and Control at Multi-Lane Intersections with Mixed Traffic Flow