Physics-Guided Multi-Modal Motion Prediction with Interaction-Aware GRU
Technologies · 15.07.2026 · 14 (7) · 433
DOI: 10.3390/technologies14070433 ↗Kurz gefasst
Fachartikel in der Zeitschrift Technologies, online erschienen am 15.07.2026. Verfasst von Umut Özkan und 4 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Committee of Science of the Ministry of Science and Higher Education of the Republic of Kazakhstan.
Abstract (Originalsprache)
In the Argoverse 2 experiments reported here, the simplest Constant Turn Rate and Acceleration (CTRA) decoder was stable but missed many interaction-driven turns and merges, while residual decoders without enough control improved early displacement but increased final-horizon error. This paper therefore studies a compact decoder in which each of the six futures is represented as a CTRA anchor plus an autoregressive position residual. The residual gated recurrent unit (GRU) is initialized from fused target-history, top-k neighbor, and lane-polyline context, and its contribution is scaled by a mode-specific gate and learned exponential decay. On the 10k/2k sanity ablations, CTRA-only decoding reached minFDE6=7.189 m, while autoregressive residuals with a larger correction GRU reduced it to 4.157 m; removing the gate increased it again to 4.946 m. On the full Argoverse 2 validation split, the final configuration achieves a minimum average displacement error of minADE6=1.21 m and a minimum final displacement error of minFDE6=2.78 m. The reported diagnostics show that the compact model generates a useful six-mode set, but still needs better probability ranking for top-1 selection.
Förderung
- Committee of Science of the Ministry of Science and Higher Education of the Republic of Kazakhstan
Zitieren
APA
Özkan, U., Shayea, I., Rzayeva, L., Batkuldin, A., & Nyssanov, N. (2026). Physics-Guided Multi-Modal Motion Prediction with Interaction-Aware GRU. Technologies, 14(7), 433. https://doi.org/10.3390/technologies14070433
Vancouver
Özkan U, Shayea I, Rzayeva L, Batkuldin A, Nyssanov N. Physics-Guided Multi-Modal Motion Prediction with Interaction-Aware GRU. Technologies. 2026;14(7):433. doi:10.3390/technologies14070433
BibTeX
@article{zkan2026,
author = {Özkan, Umut and Shayea, Ibraheem and Rzayeva, Leila and Batkuldin, Alisher and Nyssanov, Nursultan},
title = {Physics-Guided Multi-Modal Motion Prediction with Interaction-Aware GRU},
journal = {Technologies},
year = {2026},
volume = {14},
number = {7},
pages = {433},
doi = {10.3390/technologies14070433}
}
EUVIMED