Off the Beaten Sidewalk: Pedestrian Prediction in Shared Spaces for Autonomous Vehicles

  title={Off the Beaten Sidewalk: Pedestrian Prediction in Shared Spaces for Autonomous Vehicles},
  author={Cyrus Anderson and Ram Vasudevan and Matthew Johnson-Roberson},
  journal={IEEE Robotics and Automation Letters},
Pedestrians and drivers interact closely in a wide range of environments. Autonomous vehicles (AVs) correspondingly face the need to predict pedestrians’ future trajectories in these same environments. Traditional model-based prediction methods have been limited to making predictions in highly structured scenes with signalized intersections, marked crosswalks, or curbs. Deep learning methods have instead leveraged datasets to learn predictive features that generalize across scenes, at the cost… 

Figures and Tables from this paper

Pedestrian Trajectory Prediction using Context-Augmented Transformer Networks

This work is introducing a framework based on the transformer networks that were shown recently to be more efficient and outperformed RNNs in many sequential-based tasks and has outperformed the compared baseline approaches in both short-term and long-term prediction horizons.

Pedestrian Trajectory Prediction for Real-Time Autonomous Systems via Context-Augmented Transformer Networks

A framework based on the transformer networks that were demonstrated recently to be more efficient and outperformed RNNs in many sequential-based tasks is introduced, relying on a fusion of sensor modalities, namely the past positional information, agent interactions information and scene physical semantics information as an input to the framework.

Multimodal Hybrid Pedestrian: A Hybrid Automaton Model of Urban Pedestrian Behavior for Automated Driving Applications

A hybrid automaton model of multimodal pedestrian behavior called Multimodal Hybrid Pedestrian (MHP) is developed that is applicable to a wide variety of urban scenarios—midblocks, intersections, one-way, and two-way streets, etc., and the probabilistic predictions from the model can be utilized for AV motion planning.

Guest Editorial: Introduction to the Special Issue on Long-Term Human Motion Prediction

The recent trend shows that the community aims to develop end-to-end approaches that make use of prediction as the gluing unit between perception and planning/control units, as well as novel deep learning architectures allow better interleaving of the aforementioned units.

Sub-Goal Social Force Model for Collective Pedestrian Motion Under Vehicle Influence

The proposedSG-SFM introduced a new design of vehicle influence on pedestrian motion, which was smoothly combined with the influence of surrounding pedestrians using the sub-goal concept to describe the collective pedestrian motion under vehicle influence.

Behavior Prediction of Traffic Actors for Intelligent Vehicle Using Artificial Intelligence Techniques: A Review

The findings show that using sophisticated input representation that includes traffic rules and road geometry, artificial intelligence-based solutions applied to behavior prediction of traffic actors for intelligent vehicles have shown promising success, particularly in complex driving scenarios.



Social LSTM: Human Trajectory Prediction in Crowded Spaces

This work proposes an LSTM model which can learn general human movement and predict their future trajectories and outperforms state-of-the-art methods on some of these datasets.

TrafficPredict: Trajectory Prediction for Heterogeneous Traffic-Agents

The proposed long short-term memory-based (LSTM-based) realtime traffic prediction algorithm, TrafficPredict, uses an instance layer to learn instances' movements and interactions and has a category layer to learning the similarities of instances belonging to the same type to refine the prediction.

Discrete Residual Flow for Probabilistic Pedestrian Behavior Prediction

This work proposes the discrete residual flow network (DRF-Net), a convolutional neural network for human motion prediction that captures the uncertainty inherent in long-range motion forecasting and effectively captures multimodal posteriors over future human motion by predicting and updating a discretized distribution over spatial locations.

Multi-Agent Tensor Fusion for Contextual Trajectory Prediction

  • Tianyang ZhaoYifei Xu Y. Wu
  • Computer Science
    2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2019
This work encodes multiple agents' past trajectories and the scene context into a Multi-Agent Tensor, then applies convolutional fusion to capture multiagent interactions while retaining the spatial structure of agents and thescene context.

TraPHic: Trajectory Prediction in Dense and Heterogeneous Traffic Using Weighted Interactions

A new algorithm for predicting the near-term trajectories of road agents in dense traffic videos using a novel LSTM-CNN hybrid network for trajectory prediction that outperform state-of-the-art methods on dense traffic datasets by 30%.

SS-LSTM: A Hierarchical LSTM Model for Pedestrian Trajectory Prediction

A novel hierarchical LSTM-based network is proposed to consider both the influence of social neighbourhood and scene layouts in pedestrian trajectory prediction, showing that the method outperforms other methods and that using circular shape neighbourhood improves the prediction accuracy.

Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networks

A graph-based generative adversarial network that generates realistic, multimodal trajectory predictions by better modelling the social interactions of pedestrians in a scene and achieves state-of-the-art performance comparing it to several baselines on existing trajectory forecasting benchmarks.

Learning Generative Socially Aware Models of Pedestrian Motion

A novel Bayesian generative model is proposed, which generalizes a popular class of deterministic multi-agent motion models and is capable of encoding traffic agents’ interactions in a latent probabilistic space and can be exploited for inference, prediction, and trained for learning of agent interaction dynamics.

Analysis and Prediction of Pedestrian Crosswalk Behavior during Automated Vehicle Interactions

A hybrid systems model is developed that uses pedestrians’ gap acceptance behavior and constant velocity dynamics for long-term pedestrian trajectory prediction when interacting with AVs and compared measures of pedestrian crossing behaviors in the immersive virtual environment to that in the real world, and found similarities between the two.

Top-view Trajectories: A Pedestrian Dataset of Vehicle-Crowd Interaction from Controlled Experiments and Crowded Campus

This study proposed two pedestrian trajectory datasets, CITR dataset and DUT dataset, so that the pedestrian motion models can be further calibrated and verified, especially when vehicle influence on pedestrians plays an important role.