One of the key challenges for autonomous vehicles is the ability to accurately predict the motion of other objects in the surrounding environment, such as pedestrians or other vehicles. In this contribution, a novel motion forecasting approach for autonomous vehicles is developed, inspired by the work of Gilles et al. [1]. We predict multiple heatmaps with a neural-network-based model for every traffic participant in the vicinity of the autonomous vehicle; with one heatmap per timestep. The heatmaps are used as input to a novel sampling algorithm that extracts coordinates corresponding to the most likely future positions. We experiment with different encoders and decoders, as well as a comparison of two loss functions. Additionally, a new grid-scaling technique is introduced, showing further improved performance. Overall, our approach improves state-of-the-art miss rate performance for the function-relevant prediction interval of 3 seconds while being competitive in longer prediction intervals (up to eight seconds). The evaluation is done on the public 2022 Waymo motion challenge.
The WayHome: Long-Term Motion Prediction on Dynamically Scaled Grids
2023-09-24
845359 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Long-term Vehicle Motion Prediction
British Library Conference Proceedings | 2009
|Statistical long-term motion prediction
TIBKAT | 2016
|Long-term vehicle motion prediction
IEEE | 2009
|LTN: Long-Term Network for Long-Term Motion Prediction
IEEE | 2021
|