DiversityGAN: Diversity-Aware Vehicle Motion Prediction via Latent Semantic Sampling

Huang, Xin; McGill, Stephen G.; DeCastro, Jonathan A.; Fletcher, Luke; Leonard, John J.; Williams, Brian; Rosman, Guy

doi:10.48550/arxiv.1911.12736

Cited by 6 publications

(8 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example [21] defines prediction in this setting as trajectory forecasting, a class of imitation learning with non-interaction. Such approaches usually incorporate probabilistic latent variable priors that capture uncertainty as a set of non-interpretable random variables, or leverage Generative Adversarial Networks (GAN) [27]. However, their major drawback is the reliance on sequential sampling, which accumulates errors in each step.…”

Section: Related Workmentioning

confidence: 99%

Self-Supervised Action-Space Prediction for Automated Driving

Janjoš¹,

Dolgov²,

Zöllner³

2021

Preprint

View full text Add to dashboard Cite

Making informed driving decisions requires reliable prediction of other vehicles' trajectories. In this paper, we present a novel learned multi-modal trajectory prediction architecture for automated driving. It achieves kinematically feasible predictions by casting the learning problem into the space of accelerations and steering angles -by performing action-space prediction, we can leverage valuable model knowledge. Additionally, the dimensionality of the action manifold is lower than that of the state manifold, whose intrinsically correlated states are more difficult to capture in a learned manner. For the purpose of action-space prediction, we present the simple Feed-Forward Action-Space Prediction (FFW-ASP) architecture. Then, we build on this notion and introduce the novel Self-Supervised Action-Space Prediction (SSP-ASP) architecture that outputs future environment context features in addition to trajectories. A key element in the self-supervised architecture is that, based on an observed action history and past context features, future context features are predicted prior to future trajectories. The proposed methods are evaluated on real-world datasets containing urban intersections and roundabouts, and show accurate predictions, outperforming state-of-the-art for kinematically feasible predictions in several prediction metrics.

show abstract

Section: Related Workmentioning

confidence: 99%

Self-Supervised Action-Space Prediction for Automated Driving

Janjoš¹,

Dolgov²,

Zöllner³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Recent developments in learning-based methods of trajectory prediction such as RNNs [9], [10] allow for improved prediction in crowded environments, outperforming parametric based methods such as SFM [11]. These methods have been applied to multimodal prediction by learning semantically meaningful latent representations in conditional variational autoencoders [12], [13] and GANs [14], or through clustering modal paths in output distributions [15]. However, these methods can still fail to outperform even simple baselines such as constant velocity models in many situations [16].…”

Section: Arxiv:200612906v1 [Cscv] 23 Jun 2020mentioning

confidence: 99%

“…This method, as well as a similar GAN based approach proposed by [18], also make use of overhead contextual scene inputs, which are often difficult to capture in autonomous driving systems. Prior work [2], [5], [14] using GANs for trajectory prediction has followed the assumption from GAN application to image synthesis that we cannot efficiently evaluate the output distribution, but can sample from it, requiring multiple iterations to identify the true multimodal distribution. However, our problem's output distribution is much lower-dimensional than image synthesis, and has been modelled previously by GMMs as in [12], [15], [19], allowing a distribution to be generated from a single iteration.…”

Section: Arxiv:200612906v1 [Cscv] 23 Jun 2020mentioning

confidence: 99%

Probabilistic Crowd GAN: Multimodal Pedestrian Trajectory Prediction Using a Graph Vehicle-Pedestrian Attention Network

Eiffert

Shan

et al. 2020

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Understanding and predicting the intention of pedestrians is essential to enable autonomous vehicles and mobile robots to navigate crowds. This problem becomes increasingly complex when we consider the uncertainty and multimodality of pedestrian motion, as well as the implicit interactions between members of a crowd, including any response to a vehicle. Our approach, Probabilistic Crowd GAN, extends recent work in trajectory prediction, combining Recurrent Neural Networks (RNNs) with Mixture Density Networks (MDNs) to output probabilistic multimodal predictions, from which likely modal paths are found and used for adversarial training. We also propose the use of Graph Vehicle-Pedestrian Attention Network (GVAT), which models social interactions and allows input of a shared vehicle feature, showing that inclusion of this module leads to improved trajectory prediction both with and without the presence of a vehicle. Through evaluation on various datasets, we demonstrate improvements on the existing state of the art methods for trajectory prediction and illustrate how the true multimodal and uncertain nature of crowd interactions can be directly modelled.

show abstract

“…Predictions, however, are inherently uncertain, so it is desirable to represent uncertainty in predictions of possible future states and reason about this uncertainty while planning. This desire is motivating ongoing work in the behavior prediction community to go beyond single mean average precision (MAP) prediction and develop methods for generating probabilistic predictions [1]- [4]. In the most general sense, this involves learning joint distributions for the future states of all the agents conditioned on their past trajectories and other context specific variables (e.g.…”

Section: Introductionmentioning

confidence: 99%

Fast Risk Assessment for Autonomous Vehicles Using Learned Models of Agent Futures

Wang¹,

Huang²,

Jasour³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

This paper presents fast non-sampling based methods to assess the risk of trajectories for autonomous vehicles when probabilistic predictions of other agents' futures are generated by deep neural networks (DNNs). The presented methods address a wide range of representations for uncertain predictions including both Gaussian and non-Gaussian mixture models for predictions of both agent positions and controls. We show that the problem of risk assessment when Gaussian mixture models (GMMs) of agent positions are learned can be solved rapidly to arbitrary levels of accuracy with existing numerical methods. To address the problem of risk assessment for non-Gaussian mixture models of agent position, we propose finding upper bounds on risk using Chebyshev's Inequality and sums-of-squares (SOS) programming; they are both of interest as the former is much faster while the latter can be arbitrarily tight. These approaches only require statistical moments of agent positions to determine upper bounds on risk. To perform risk assessment when models are learned for agent controls as opposed to positions, we develop TreeRing, an algorithm analogous to tree search over the ring of polynomials that can be used to exactly propagate moments of control distributions into position distributions through nonlinear dynamics. The presented methods are demonstrated on realistic predictions from DNNs trained on the Argoverse and CARLA datasets and are shown to be effective for rapidly assessing the probability of low probability events.

show abstract

DiversityGAN: Diversity-Aware Vehicle Motion Prediction via Latent Semantic Sampling

Cited by 6 publications

References 42 publications

Self-Supervised Action-Space Prediction for Automated Driving

Self-Supervised Action-Space Prediction for Automated Driving

Probabilistic Crowd GAN: Multimodal Pedestrian Trajectory Prediction Using a Graph Vehicle-Pedestrian Attention Network

Fast Risk Assessment for Autonomous Vehicles Using Learned Models of Agent Futures

Contact Info

Product

Resources

About