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

Accurate trajectory prediction of vehicles is essential for reliableautonomous driving. To maintain consistent performance as a vehicle drivingaround different cities, it is crucial to adapt to changing trafficcircumstances and achieve lifelong trajectory prediction model. To realize it,catastrophic forgetting is a main problem to be addressed. In this paper, adivergence measurement method based on conditional Kullback-Leibler divergenceis proposed first to evaluate spatiotemporal dependency difference among varieddriving circumstances. Then based on generative replay, a novel lifelongvehicle trajectory prediction framework is developed. The framework consists ofa conditional generation model and a vehicle trajectory prediction model. Theconditional generation model is a generative adversarial network conditioned onposition configuration of vehicles. After learning and merging trajectorydistribution of vehicles across different cities, the generation model replaystrajectories with prior samplings as inputs, which alleviates catastrophicforgetting. The vehicle trajectory prediction model is trained by the replayedtrajectories and achieves consistent prediction performance on visited cities.A lifelong experiment setup is established on four open datasets including fivetasks. Spatiotemporal dependency divergence is calculated for different tasks.Even though these divergence, the proposed framework exhibits lifelong learningability and achieves consistent performance on all tasks.

Lifelong Vehicle Trajectory Prediction Framework Based on Generative Replay