Kernel Trajectory Maps for Multi-Modal Probabilistic Motion Prediction

Weiming Zhi, Lionel Ott, Fabio Ramos
Proceedings of the Conference on Robot Learning, PMLR 100:1405-1414, 2020.

Abstract

Understanding the dynamics of an environment, such as the movement of humans and vehicles, is crucial for agents to achieve long-term autonomy in urban environments. This requires the development of methods to capture the multimodal and probabilistic nature of motion patterns. We present kernel trajectory maps (KTM) to capture the trajectories of movement in an environment. KTMs leverage the expressiveness of kernels from non-parametric modelling by projecting input trajectories onto a set of representative trajectories, to condition on a sequence of observed waypoint coordinates, and predict a multi-modal distribution over possible future trajectories. The output is a mixture of continuous stochastic processes, where each realisation is a continuous functional trajectory, which can be queried at arbitrarily fine time steps.

Cite this Paper

BibTeX
@InProceedings{pmlr-v100-zhi20a, title = {Kernel Trajectory Maps for Multi-Modal Probabilistic Motion Prediction}, author = {Zhi, Weiming and Ott, Lionel and Ramos, Fabio}, booktitle = {Proceedings of the Conference on Robot Learning}, pages = {1405--1414}, year = {2020}, editor = {Kaelbling, Leslie Pack and Kragic, Danica and Sugiura, Komei}, volume = {100}, series = {Proceedings of Machine Learning Research}, month = {30 Oct--01 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v100/zhi20a/zhi20a.pdf}, url = {https://proceedings.mlr.press/v100/zhi20a.html}, abstract = {Understanding the dynamics of an environment, such as the movement of humans and vehicles, is crucial for agents to achieve long-term autonomy in urban environments. This requires the development of methods to capture the multimodal and probabilistic nature of motion patterns. We present kernel trajectory maps (KTM) to capture the trajectories of movement in an environment. KTMs leverage the expressiveness of kernels from non-parametric modelling by projecting input trajectories onto a set of representative trajectories, to condition on a sequence of observed waypoint coordinates, and predict a multi-modal distribution over possible future trajectories. The output is a mixture of continuous stochastic processes, where each realisation is a continuous functional trajectory, which can be queried at arbitrarily fine time steps.} }
Endnote
%0 Conference Paper %T Kernel Trajectory Maps for Multi-Modal Probabilistic Motion Prediction %A Weiming Zhi %A Lionel Ott %A Fabio Ramos %B Proceedings of the Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2020 %E Leslie Pack Kaelbling %E Danica Kragic %E Komei Sugiura %F pmlr-v100-zhi20a %I PMLR %P 1405--1414 %U https://proceedings.mlr.press/v100/zhi20a.html %V 100 %X Understanding the dynamics of an environment, such as the movement of humans and vehicles, is crucial for agents to achieve long-term autonomy in urban environments. This requires the development of methods to capture the multimodal and probabilistic nature of motion patterns. We present kernel trajectory maps (KTM) to capture the trajectories of movement in an environment. KTMs leverage the expressiveness of kernels from non-parametric modelling by projecting input trajectories onto a set of representative trajectories, to condition on a sequence of observed waypoint coordinates, and predict a multi-modal distribution over possible future trajectories. The output is a mixture of continuous stochastic processes, where each realisation is a continuous functional trajectory, which can be queried at arbitrarily fine time steps.
APA
Zhi, W., Ott, L. & Ramos, F.. (2020). Kernel Trajectory Maps for Multi-Modal Probabilistic Motion Prediction. Proceedings of the Conference on Robot Learning, in Proceedings of Machine Learning Research 100:1405-1414 Available from https://proceedings.mlr.press/v100/zhi20a.html.

Related Material