aiqbench analysis
This research introduces NeMo-map, a neural implicit function framework that models spatio-temporal motion patterns without relying on discrete spatial sampling or expensive offline construction. By directly mapping coordinates to Semi-Wrapped Gaussian Mixture Model parameters, the method achieves superior accuracy in trajectory prediction and smooth generalization across sparse real-world datasets. The approach represents a significant advancement in dynamic map representation for safe robot operation in complex human environments.
NeMo-map: Neural Implicit Flow Fields for Spatio-Temporal Motion Mapping
arXiv:2510.14827v3 Announce Type: replace Abstract: Safe and efficient robot operation in complex human environments can benefit from good models of site-specific motion patterns. Maps of Dynamics (MoDs) provide such models by encoding statistical motion patterns in a map, but existing representations use discrete spatial sampling and typically require costly offline construction. We propose a continuous spatio-temporal MoD representation based on implicit neural functions that directly map coordinates to the parameters of a Semi-Wrapped Gaussian Mixture Model. This removes the need for discretization and imputation for unevenly sampled regions, enabling smooth generalization across both space and time. Evaluated on two public datasets with real-world people tracking data, our method achieves better accuracy of motion representation and smoother velocity distributions in sparse regions while still being computationally efficient, compared to available baselines. The proposed approach demonstrates a powerful and efficient way of modeling complex human motion patterns and high performance in the trajectory prediction downstream task. The code is publicly available at https://github.com/test-bai-cpu/nemo-map.
Abstract:Safe and efficient robot operation in complex human environments can benefit from good models of site-specific motion patterns. Maps of Dynamics (MoDs) provide such models by encoding statistical motion patterns in a map, but existing representations use discrete spatial sampling and typically require costly offline construction. We propose a continuous spatio-temporal MoD representation based on implicit neural functions that directly map coordinates to the parameters of a Semi-Wrapped Gaussian Mixture Model. This removes the need for discretization and imputation for unevenly sampled regions, enabling smooth generalization across both space and time. Evaluated on two public datasets with real-world people tracking data, our method achieves better accuracy of motion representation and smoother velocity distributions in sparse regions while still being computationally efficient, compared to available baselines. The proposed approach demonstrates a powerful and efficient way of modeling complex human motion patterns and high performance in the trajectory prediction downstream task. The code is publicly available at this https URL.
Robotics (cs.RO)
arXiv:2510.14827 [cs.RO]
(or arXiv:2510.14827v3 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2510.14827 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 249 words
Related coverage
Physics-Informed Neural Networks for Depth-Averaged Avalanche Dynamics
arXiv · Sep 30, 2026
LLM-enhanced spatio-temporal learning for grid-level docked bike sharing demand prediction
arXiv · Oct 6, 2026
Mechanical Field Networks: Structured Neural Dynamics for Multivariate Systems
arXiv · Jun 12, 2026
Anticipatory Risk-Guided Reinforcement Learning for Safe Flight Through Dynamic Clutter
arXiv · Jul 28, 2026