Spatiotemporal Learning of Directional Uncertainty in Urban Environments With Kernel Recurrent Mixture Density Networks

Zhi, Weiming; Senanayake, Ransalu; Ott, Lionel; Ramos, Fábio

doi:10.1109/lra.2019.2931262

Cited by 29 publications

(10 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, these methods are effective in short-term predictions. To make long-term predictions, flow-based methods [18] [19] are presented to capture the directional flow in the scene by learning from observed trajectories. To handle complicated scenarios, many learning-based algorithms, including Gaussian mixture regression [20], hidden Markov models [21], Gaussian process [22], random tree searching [23], and dynamic Bayesian networks (DBNs) [24] have been proposed.…”

Section: A Trajectory Prediction Methodsmentioning

confidence: 99%

A Novel Trajectory Generator Based on a Constrained GAN and a Latent Variables Predictor

Yang

Wang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Forecasting pedestrian trajectory is critical for versatile applications, such as autonomous driving and social robot, when they work in human-centric environments. However, it is challenging to predict pedestrians' future trajectories due to the inherent human properties and pedestrians' social interactions. Recent works predict future trajectories by using a generative model, which captures social interactions with pooling-or graph-based strategies and generates multi-modal outputs with latent variables sampled from random Gaussian noise. Nevertheless, they introduce little human knowledge, which is beneficial for improved prediction performance. In this work, we propose to learn informative latent variables from pedestrians' future trajectories. Moreover, we present a distance-direction pooling module, which captures social interactions in a more intuitive manner. Besides, we introduce an additional constraint on generative adversarial network optimization to generate more realistic results. Two benchmarking datasets, ETH [1] and UCY [2], are used to evaluate the proposed method. Comparisons between our method and several state-of-the-art methods demonstrate the superiority of the proposed method in generating more accurate future trajectories. INDEX TERMS trajectory forecasting, generative adversarial network, latent variable predictor, future uncertainty.

show abstract

Section: A Trajectory Prediction Methodsmentioning

confidence: 99%

A Novel Trajectory Generator Based on a Constrained GAN and a Latent Variables Predictor

Yang

Wang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…However, a simple kinematic model is unsuitable for long-term predictions. For long prediction horizons, flow-based methods [12][13] are proposed to learn the directional flow from observed trajectories in the scene. Subsequently, trajectories are generated by recursively sampling the distribution of future motion derived from the learned directional flow.…”

Section: A Trajectory Prediction Methodsmentioning

confidence: 99%

TPPO: A Novel Trajectory Predictor with Pseudo Oracle

Yang¹,

Yan²,

Wang³

et al. 2020

Preprint

View full text Add to dashboard Cite

Forecasting pedestrian trajectories in dynamic scenes remains a critical problem with various applications, such as autonomous driving and socially aware robots. Such forecasting is challenging due to human-human and humanobject interactions and future uncertainties caused by human randomness. Generative model-based methods handle future uncertainties by sampling a latent variable. However, few previous studies carefully explored the generation of the latent variable. In this work, we propose the Trajectory Predictor with Pseudo Oracle (TPPO), which is a generative model-based trajectory predictor. The first pseudo oracle is pedestrians' moving directions, and the second one is the latent variable estimated from observed trajectories. A social attention module is used to aggregate neighbors' interactions on the basis of the correlation between pedestrians' moving directions and their future trajectories. This correlation is inspired by the fact that a pedestrian's future trajectory is often influenced by pedestrians in front. A latent variable predictor is proposed to estimate latent variable distributions from observed and ground-truth trajectories. Moreover, the gap between these two distributions is minimized during training. Therefore, the latent variable predictor can estimate the latent variable from observed trajectories to approximate that estimated from ground-truth trajectories. We compare the performance of TPPO with related methods on several public datasets. Results demonstrate that TPPO outperforms state-ofthe-art methods with low average and final displacement errors. Besides, the ablation study shows that the prediction performance will not dramatically decrease as sampling times decline during tests.

show abstract

“…Thus, Jovan et al [29] use periodic Poisson processes to characterise the behaviour in time of different rooms of a building, which helps to better capture the nature of human activities. Zhi et al [30] use an LSTM network to model a multi-modal probability density function over the possible directions in which an object can move over time. Krajnik et al [20] present a model for introducing time into discrete and continuous spatial representations by modelling longterm, pseudo-periodic variations caused by human activities or natural processes, wrapping time in several dimensions.…”

Section: B Human Motion Modellingmentioning

confidence: 99%

Robotic Exploration for Learning Human Motion Patterns

2022

View full text Add to dashboard Cite

Understanding how people are likely to move is key to efficient and safe robot navigation in human environments. However, mobile robots can only observe a fraction of the environment at a time, while the activity patterns of people may also change at different times. This paper introduces a new methodology for mobile robot exploration to maximise the knowledge of human activity patterns by deciding where and when to collect observations. We introduce an exploration policy driven by the entropy levels in a spatio-temporal map of pedestrian flows, and compare multiple spatio-temporal exploration strategies including both informed and uninformed approaches. The evaluation is performed by simulating mobile robot exploration using real sensory data from three long-term pedestrian datasets. The results show that for certain scenarios the models built with proposed exploration system can better predict the flow patterns than uninformed strategies, allowing the robot to move in a more socially compliant way, and that the exploration ratio is a key factor when it comes to the model prediction accuracy.

show abstract

Spatiotemporal Learning of Directional Uncertainty in Urban Environments With Kernel Recurrent Mixture Density Networks

Cited by 29 publications

References 18 publications

A Novel Trajectory Generator Based on a Constrained GAN and a Latent Variables Predictor

A Novel Trajectory Generator Based on a Constrained GAN and a Latent Variables Predictor

TPPO: A Novel Trajectory Predictor with Pseudo Oracle

Robotic Exploration for Learning Human Motion Patterns

Contact Info

Product

Resources

About