Intention-aware Denoising Diffusion Model for Trajectory Prediction

arXiv:2403.09190v1 Announce Type: new
Abstract: Trajectory prediction is an essential component in autonomous driving, particularly for collision avoidance systems. Considering the inherent uncertainty of the task, numerous studies have utilized generative models to produce multiple plausible future trajectories for each agent. However, most of them suffer from restricted representation ability or unstable training issues. To overcome these limitations, we propose utilizing the diffusion model to generate the distribution of future trajectories. Two cruxes are to be settled to realize such …

abstract agent arxiv autonomous autonomous driving collision cs.ai cs.cv denoising diffusion diffusion model driving future generative generative models however multiple prediction representation studies systems them training trajectory type uncertainty