Filters








18,730 Hits in 6.0 sec

Predicting Vehicle Behaviors Over An Extended Horizon Using Behavior Interaction Network [article]

Wenchao Ding and Jing Chen and Shaojie Shen
2019 arXiv   pre-print
In this paper, we uncover that clues to vehicle behaviors over an extended horizon can be found in vehicle interaction, which makes it possible to anticipate the likelihood of a certain behavior, even  ...  We adopt a recurrent neural network (RNN) for observation encoding, and based on that, we propose a novel vehicle behavior interaction network (VBIN) to capture the vehicle interaction from the hidden  ...  The interaction-aware models all improve the average prediction time, which verifies the conjecture that interaction can help to predict behaviors over an extended horizon.  ... 
arXiv:1903.00848v2 fatcat:ownit44im5ek7euz4y5x3f3lru

Planning on the fast lane: Learning to interact using attention mechanisms in path integral inverse reinforcement learning [article]

Sascha Rosbach, Xing Li, Simon Großjohann, Silviu Homoceanu, Stefan Roth
2020 arXiv   pre-print
Furthermore, the temporal attention mechanism learns persistent interaction with other vehicles over an extended planning horizon.  ...  In this work, we are concerned with the sequential reward prediction over an extended time horizon.  ...  In this work, we focus on situation-dependent reward predictions using inverse reinforcement learning (IRL) that enables persistent behavior over an extended time horizon.  ... 
arXiv:2007.05798v2 fatcat:fjploythabgknjp6x3gnl2ewkq

Imitating Driver Behavior with Generative Adversarial Networks [article]

Alex Kuefler, Jeremy Morton, Tim Wheeler, Mykel Kochenderfer
2017 arXiv   pre-print
Our model both reproduces emergent behavior of human drivers, such as lane change rate, while maintaining realistic control over long time horizons.  ...  The ability to accurately predict and simulate human driving behavior is critical for the development of intelligent transportation systems.  ...  Although behavioral cloning still outperforms Generative Adversarial Imitation Learning on short (∼2 s) horizons, its greedy behavior prevents it from achieving realistic driving over an extended period  ... 
arXiv:1701.06699v1 fatcat:elhdfvqu75bjjbtqav4eqhwqse

Relational Recurrent Neural Networks For Vehicle Trajectory Prediction

Kaouther Messaoud, Itheri Yahiaoui, Anne Verroust-Blondet, Fawzi Nashashibi
2019 2019 IEEE Intelligent Transportation Systems Conference (ITSC)  
It outputs an estimate of future trajectories over 5s time horizon for longitudinal and lateral prediction RMSE of about 3.34m and 0.48m, respectively.  ...  RRNNs) to tackle the vehicle motion prediction problem.  ...  This enhance the network ability to learn generalized behavior over different drivers and different driving conditions.  ... 
doi:10.1109/itsc.2019.8916887 dblp:conf/itsc/MessaoudYVN19 fatcat:t37h24wjsjg4njvv63lap75gcm

Hierarchical Adaptable and Transferable Networks (HATN) for Driving Behavior Prediction [article]

Letian Wang, Yeping Hu, Liting Sun, Wei Zhan, Masayoshi Tomizuka, Changliu Liu
2021 arXiv   pre-print
Besides, our model is also able to capture variations of driving behaviors among individuals and scenarios by an online adaptation module.  ...  our method outperformed other methods in terms of prediction accuracy and transferability.  ...  Nevertheless, behavior prediction accuracy usually decays exponentially as horizon extends.  ... 
arXiv:2111.00788v3 fatcat:fuoe3txfkfhdrow3nmpz2ejtoq

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

Rohan Chandra, Uttaran Bhattacharya, Aniket Bera, Dinesh Manocha
2019 arXiv   pre-print
In addition, we model horizon-based interactions which are used to implicitly model the driving behavior of each road-agent.  ...  We model the interactions between different road-agents using a novel LSTM-CNN hybrid network for trajectory prediction.  ...  We use an LSTM-CNN hybrid network to model two kinds of weighted interactions between road agents: horizon-based and heterogeneousbased.  ... 
arXiv:1812.04767v3 fatcat:dwyjs4tv45aplmshndvscroqfu

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

Rohan Chandra, Uttaran Bhattacharya, Aniket Bera, Dinesh Manocha
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
In addition, we model horizon-based interactions which are used to implicitly model the driving behavior of each road agent.  ...  We model the interactions between different road agents using a novel LSTM-CNN hybrid network for trajectory prediction.  ...  We use an LSTM-CNN hybrid network to model two kinds of weighted interactions between road agents: horizon-based and heterogeneousbased.  ... 
doi:10.1109/cvpr.2019.00868 dblp:conf/cvpr/ChandraBBM19 fatcat:ycj2ak6arfb27ctc3lyewsmrli

Beyond RMSE: Do machine-learned models of road user interaction produce human-like behavior? [article]

Aravinda Ramakrishnan Srinivasan, Yi-Shin Lin, Morris Antonello, Anthony Knittel, Mohamed Hasan, Majd Hawasly, John Redford, Subramanian Ramamoorthy, Matteo Leonetti, Jac Billington, Richard Romano, Gustav Markkula
2022 arXiv   pre-print
Autonomous vehicles use a variety of sensors and machine-learned models to predict the behavior of surrounding road users.  ...  2) Lane change by an on-highway vehicle to accommodate an on-ramp vehicle 3) Lane changes by vehicles on the highway to avoid lead vehicle conflicts.  ...  This can be either the average error over the entire prediction horizon or average of momentary error at different time points in the prediction horizon.  ... 
arXiv:2206.11110v1 fatcat:tchg5jx24zh5ngfnvbdyzzjk4y

Online Parameter Estimation for Human Driver Behavior Prediction [article]

Raunak Bhattacharyya, Ransalu Senanayake, Kyle Brown, Mykel Kochenderfer
2020 arXiv   pre-print
Driver models are invaluable for planning in autonomous vehicles as well as validating their safety in simulation.  ...  We solve the online parameter estimation problem using particle filtering, and benchmark performance against rule-based and black-box driver models on two real world driving data sets.  ...  The authors thank Jeremy Morton and Louis Dressel for useful discussions.  ... 
arXiv:2005.02597v1 fatcat:x4ahb3wx3vcejdqaahyx4xr7ky

BITS: Bi-level Imitation for Traffic Simulation [article]

Danfei Xu, Yuxiao Chen, Boris Ivanovic, Marco Pavone
2022 arXiv   pre-print
The method also incorporates a planning module to obtain stable long-horizon behaviors.  ...  Simulation is the key to scaling up validation and verification for robotic systems such as autonomous vehicles.  ...  them through an agent's dynamics model (e.g., extended unicycle dynamics [62] for vehicles).  ... 
arXiv:2208.12403v1 fatcat:6zlsuxavtnhvpoxup76a4rjwea

Artificial Intelligence for Vehicle Behavior Anticipation: Hybrid Approach Based on Maneuver Classification and Trajectory Prediction

Abdelmoudjib Benterki, Moussa Boukhnifer, Vincent Judalet, Choubeila Maaoui
2020 IEEE Access  
This paper proposes a hybrid approach to integrate maneuver classification using neural networks and trajectory prediction using Long Short-term Memory (LSTM) networks to get the future positions of adjacent  ...  Understanding the behaviors of surrounding vehicles is essential for improving safety and mobility of autonomous vehicles.  ...  The probability of trajectory prediction of each model was interacted, mixed, combined, and updated to predict the long-term vehicle trajectory using an interactive multiple model trajectory prediction  ... 
doi:10.1109/access.2020.2982170 fatcat:acds353bpbfzxaf76zhsfmcj7u

Transferable and Adaptable Driving Behavior Prediction [article]

Letian Wang, Yeping Hu, Liting Sun, Wei Zhan, Masayoshi Tomizuka, Changliu Liu
2022 arXiv   pre-print
Besides, our model is able to capture variations of driving behaviors among individuals and scenarios by an online adaptation module.  ...  We demonstrate our algorithms in the task of trajectory prediction for real traffic data at intersections and roundabouts from the INTERACTION dataset.  ...  Nevertheless, behavior prediction error usually grows exponentially as the horizon extends.  ... 
arXiv:2202.05140v2 fatcat:gl4ogvq2vraqzhjygbuaqrdgt4

Multi-Agent Imitation Learning for Driving Simulation [article]

Raunak P. Bhattacharyya, Derek J. Phillips, Blake Wulfe, Jeremy Morton, Alex Kuefler, Mykel J. Kochenderfer
2018 arXiv   pre-print
Simulation is an appealing option for validating the safety of autonomous vehicles.  ...  This difference makes such models unsuitable for the simulation of driving scenes, where multiple agents must interact realistically over long time horizons.  ...  We thank Jayesh Gupta for useful discussions.  ... 
arXiv:1803.01044v1 fatcat:c7x7bbxcejh7ddtxdfockkojcq

Pedestrian Behavior Prediction for Automated Driving: Requirements, Metrics, and Relevant Features [article]

Michael Herman, Jörg Wagner, Vishnu Prabhakaran, Nicolas Möser, Hanna Ziesche, Waleed Ahmed, Lutz Bürkle, Ernst Kloppenburg, Claudius Gläser
2021 arXiv   pre-print
Based on human driving behavior we then derive appropriate reaction patterns of an automated vehicle and determine requirements for the prediction of pedestrians.  ...  We furthermore conduct an ablation study to evaluate the importance of different contextual cues and compare these results to ones obtained using established performance metrics for pedestrian prediction  ...  Often, Recurrent Neural Networks (RNN) are used for encoding trajectories of interacting agents and decoding future behavior [15] , [16] .  ... 
arXiv:2012.08418v2 fatcat:x5lk22k5tnauphp5vsriktgyoa

Graph Convolution Networks for Probabilistic Modeling of Driving Acceleration [article]

Jianyu Su, Peter A. Beling, Rui Guo, Kyungtae Han
2020 arXiv   pre-print
prediction horizon.  ...  Results from simulation studies using comprehensive performance metrics support the conclusion that our proposed networks outperform state-of-the-art methods in generating realistic trajectories over a  ...  This work can be extended to prediction in two dimensions, which is an important problem in autonomous driving.  ... 
arXiv:1911.09837v3 fatcat:mbm22hl3tndnxg3fgkuwob4lhe
« Previous Showing results 1 — 15 out of 18,730 results