Long-term vehicle motion prediction

Citations of this article
Mendeley users who have this article in their library.
Get full text


Future driver assistance systems will have to cope with complex traffic<br />situations, especially in the road crossing scenario. To detect potentially<br />hazardous situations as early as possible, it is therefore desirable<br />to know the position and motion of the ego-vehicle and vehicles around<br />it for several seconds in advance. For this purpose, we propose in<br />this study a long-term prediction approach based on a combined trajectory<br />classification and particle filter framework. As a measure for the<br />similarity between trajectories, we introduce the quaternion-based<br />rotationally invariant longest common subsequence {(QRLCS)} metric.<br />The trajectories are classified by a radial basis function {(RBF)}<br />classifier with an architecture that is able to process trajectories<br />of arbitrary non-uniform length. The particle filter framework simultaneously<br />tracks and assesses a large number of motion hypotheses ({\textasciitilde}102),<br />where the class-specific probabilities estimated by the {RBF} classifier<br />are used as a-priori probabilities for the hypotheses of the particle<br />filter. The hypotheses are clustered with a mean-shift technique<br />and are assigned a likelihood value. Motion prediction is performed<br />based on the cluster centre with the highest likelihood. While traditional<br />motion prediction based on curve radius and acceleration is inaccurate<br />especially during turning manoeuvres, we show that our approach achieves<br />a reasonable motion prediction even for long prediction intervals<br />of 3 s for these complex motion patterns.




Hermes, C., Wohler, C., Schenk, K., & Kummert, F. (2009). Long-term vehicle motion prediction. In IEEE Intelligent Vehicles Symposium, Proceedings (pp. 652–657). https://doi.org/10.1109/IVS.2009.5164354

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free