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

Achieving a proper balance between planning quality, safety and efficiency isa major challenge for autonomous driving. Optimisation-based motion plannersare capable of producing safe, smooth and comfortable plans, but often at thecost of runtime efficiency. On the other hand, naively deploying trajectoriesproduced by efficient-to-run deep imitation learning approaches might riskcompromising safety. In this paper, we present PILOT -- a planning frameworkthat comprises an imitation neural network followed by an efficient optimiserthat actively rectifies the network's plan, guaranteeing fulfilment of safetyand comfort requirements. The objective of the efficient optimiser is the sameas the objective of an expensive-to-run optimisation-based planning system thatthe neural network is trained offline to imitate. This efficient optimiserprovides a key layer of online protection from learning failures or deficiencyin out-of-distribution situations that might compromise safety or comfort.Using a state-of-the-art, runtime-intensive optimisation-based method as theexpert, we demonstrate in simulated autonomous driving experiments in CARLAthat PILOT achieves a seven-fold reduction in runtime when compared to theexpert it imitates without sacrificing planning quality.

PILOT: Efficient Planning by Imitation Learning and Optimisation for Safe Autonomous Driving