11,679 Hits in 6.1 sec

An Empirical Evaluation of Deep Learning on Highway Driving [article]

Brody Huval, Tao Wang, Sameep Tandon, Jeff Kiske, Will Song, Joel Pazhayampallil, Mykhaylo Andriluka, Pranav Rajpurkar, Toki Migimatsu, Royce Cheng-Yue, Fernando Mujica, Adam Coates (+1 others)
2015 arXiv   pre-print
In this paper, we presented a number of empirical evaluations of recent deep learning advances.  ...  Numerous groups have applied a variety of deep learning techniques to computer vision problems in highway perception scenarios.  ...  In addition, the authors would like to thank the author of Overfeat, Pierre Sermanet, for their helpful suggestions on image detection.  ... 
arXiv:1504.01716v3 fatcat:hcxgimqmb5a4jf236guqcmnzrq

Imitation Learning for Vision-based Lane Keeping Assistance [article]

Christopher Innocenti, Henrik Lindén, Ghazaleh Panahandeh, Lennart Svensson, Nasser Mohammadiha
2017 arXiv   pre-print
Evaluation is based on two proposed performance metrics measuring how well the vehicle is positioned in a lane and the smoothness of the driven trajectory.  ...  The policy is successfully learned via imitation learning using real-world data collected from human drivers and is evaluated in closed-loop simulated environments, demonstrating good driving behaviour  ...  Test Scenarios Scenarios used for empirical evaluation in the Unity game engine included driving on a six-lane highway with smooth curvature and a two-lane country road modelled as a much more winding  ... 
arXiv:1709.03853v1 fatcat:bbmn3jdpnne4jory4aqaqbt64m

Deep Surrogate Q-Learning for Autonomous Driving [article]

Maria Kalweit, Gabriel Kalweit, Moritz Werling, Joschka Boedecker
2022 arXiv   pre-print
We further propose an efficient implementation based on a permutation-equivariant deep neural network architecture of the Q-function to estimate action-values for a variable number of vehicles in sensor  ...  Challenging problems of deep reinforcement learning systems with regard to the application on real systems are their adaptivity to changing environments and their efficiency w.r.t. computational resources  ...  Third, we evaluate the performance in the open-source traffic simulator SUMO and furthermore train on the open real highD dataset containing top-down recordings of German highways, which is an important  ... 
arXiv:2010.11278v2 fatcat:6miucbedxnbypcscxhtvroxriu

Imitating Driver Behavior with Generative Adversarial Networks [article]

Alex Kuefler, Jeremy Morton, Tim Wheeler, Mykel Kochenderfer
2017 arXiv   pre-print
highway simulations.  ...  The ability to accurately predict and simulate human driving behavior is critical for the development of intelligent transportation systems.  ...  of highway driving that outperforms the state of the art on several metrics.  ... 
arXiv:1701.06699v1 fatcat:elhdfvqu75bjjbtqav4eqhwqse

VeMo: Enabling Transparent Vehicular Mobility Modeling at Individual Levels with Full Penetration [article]

Yu Yang, Xiaoyang Xie, Zhihan Fang, Fan Zhang, Yang Wang, Desheng Zhang
2018 arXiv   pre-print
More importantly, we evaluate VeMo with (i) a large-scale ETC system with tracking devices at 773 highway entrances and exits capturing more than 2 million vehicles every day; (ii) a fleet consisting of  ...  Understanding and predicting real-time vehicle mobility patterns on highways are essential to address traffic congestion and respond to the emergency.  ...  This work is partially supported by the by Rutgers Research Council, Rutgers Global Center, China 973 Program (2015CB352400), National Natural Science Foundation of China (41401470).  ... 
arXiv:1812.02780v3 fatcat:mz3sh4as2jhwvo4bnfphvrsope

Faster Training of Very Deep Networks Via p-Norm Gates [article]

Trang Pham, Truyen Tran, Dinh Phung, Svetha Venkatesh
2016 arXiv   pre-print
However, there is limited work in analysing the role of gating in the learning process.  ...  Gates are employed in many recent state-of-the-art recurrent models such as LSTM and GRU, and feedforward models such as Residual Nets and Highway Networks.  ...  of F1-score, Learning curves on training sets.  ... 
arXiv:1608.03639v1 fatcat:lhw2ms77mrfmfd2rgx4ygidjta

Automated Lane Change Strategy using Proximal Policy Optimization-based Deep Reinforcement Learning [article]

Fei Ye, Xuxin Cheng, Pin Wang, Ching-Yao Chan, Jiucai Zhang
2020 arXiv   pre-print
Machine learning-based methods offer an alternative approach, as Deep reinforcement learning (DRL) has shown promising success in many application domains including robotic manipulation, navigation, and  ...  In this study, we propose an automated lane change strategy using proximal policy optimization-based deep reinforcement learning, which shows great advantages in learning efficiency while still maintaining  ...  An early work of applying deep RL to lane change can be found in [11] , where the Q-masking technique is proposed to act as a mask on the output Q-values in a deep Q-learning framework to obtain a high-level  ... 
arXiv:2002.02667v2 fatcat:7beyqzig2remtnuczthjwosi7q

Can ADAS Distract Driver's Attention? An RGB-D Camera and Deep Learning-Based Analysis

Luca Ulrich, Francesca Nonis, Enrico Vezzetti, Sandro Moos, Giandomenico Caruso, Yuan Shi, Federica Marcolin
2021 Applied Sciences  
Subsequently, these images have been analyzed using a deep learning-based approach, i.e., a convolutional neural network (CNN) specifically trained to perform facial expression recognition (FER).  ...  In the present study, an experiment involving auditory and haptic ADAS has been conducted involving 11 participants, whose attention has been monitored during their driving experience.  ...  Student concentration evaluation index in an E-learning context using facial emotion analysis.  ... 
doi:10.3390/app112411587 fatcat:ietm3k3hpjb2hmagtwgsq7achq

Attention Monitoring and Hazard Assessment with Bio-Sensing and Vision: Empirical Analysis Utilizing CNNs on the KITTI Dataset [article]

Siddharth, Mohan M. Trivedi
2019 arXiv   pre-print
We collect user data on twelve subjects and show how in the absence of very large-scale datasets, we can still use pre-trained deep learning convolution networks to extract meaningful features from both  ...  The other challenge in evaluating multi-modal sensory applications is the absence of very large scale EEG data because of the various limitations of using EEG in the real world.  ...  In this manner, we computed 90 features based on face-localized points from a particular trial. 2) Deep Learning-based features: For the extraction of deep learning-based features, we used the VGG-Faces  ... 
arXiv:1905.00503v2 fatcat:fmi4jp6dpreofauc7uoent6lw4

Learning Invariant Representations for Reinforcement Learning without Reconstruction [article]

Amy Zhang, Rowan McAllister, Roberto Calandra, Yarin Gal, Sergey Levine
2021 arXiv   pre-print
We also test a first-person highway driving task where our method learns invariance to clouds, weather, and time of day.  ...  We study how representation learning can accelerate reinforcement learning from rich observations, such as images, without relying either on domain knowledge or pixel-reconstruction.  ...  Acknowledgements: We thank Audrey Durand for an insightful weekend of discussion that led to one of the results in this paper.  ... 
arXiv:2006.10742v2 fatcat:qggextd3afbb3b5v3775fblz4q

Offline Reinforcement Learning for Autonomous Driving with Safety and Exploration Enhancement [article]

Tianyu Shi, Dong Chen, Kaian Chen, Zhaojian Li
2021 arXiv   pre-print
Reinforcement learning (RL) is a powerful data-driven control method that has been largely explored in autonomous driving tasks.  ...  Experimental results in highway and parking traffic scenarios show that our approach outperforms the conventional RL method, as well as state-of-the-art offline RL algorithms.  ...  (a) Returns in highway. (b) Returns in parking. Figure 2 : 2 Figure 2: Comparison on evaluation returns between the proposed approach and state-of-the-art benchmarks.  ... 
arXiv:2110.07067v2 fatcat:56r2zae3mjhpvmnyzrt6syltmq

Guest editorial: special issue on reinforcement learning for real life

Yuxi Li, Alborz Geramifard, Lihong Li, Csaba Szepesvari, Tao Wang
2021 Machine Learning  
After a rigorous reviewing process, we accepted 11 articles, each of which was assessed by at least three reviewers, with one, mostly two, or three rounds of revisions.  ...  The main goals of the special issue are to: (1) identify key research problems that are critical for the success of real-world applications; (2) report progress on addressing these critical issues; and  ...  the performance with scenarios of highway driving, user preference inference in social networks, and water management.  ... 
doi:10.1007/s10994-021-06041-3 fatcat:ew3uhfhhevdd5khjeq2umhby7a

Corner Case Generation and Analysis for Safety Assessment of Autonomous Vehicles [article]

Haowei Sun, Shuo Feng, Xintao Yan, Henry X. Liu
2021 arXiv   pre-print
To address the challenge brought by high dimensionality, the driving environment is formulated based on Markov Decision Process, and the deep reinforcement learning techniques are applied to learn the  ...  Simulation results of a highway driving environment show that the proposed methods can effectively generate and identify the valuable corner cases.  ...  of Michigan for funding the research.  ... 
arXiv:2102.03483v1 fatcat:iaewq3rtwrbshjhgcq3fl3a6ku

Review of Learning-based Longitudinal Motion Planning for Autonomous Vehicles: Research Gaps between Self-driving and Traffic Congestion [article]

Hao Zhou, Jorge Laval, Anye Zhou, Yu Wang, Wenchao Wu, Zhu Qing, Srinivas Peeta
2021 arXiv   pre-print
This paper reviews the current state of the art in mMP, with an exclusive focus on its impact on traffic congestion.  ...  We find that: i) the AV industry has been mostly focusing on the long tail problem related to safety and overlooked the impact on traffic congestion, ii) the current public self-driving datasets have not  ...  (42) introduced the development of AVs and basics of deep learning methods, as well as summarized recent research on theories and applications of deep learning for AVs.  ... 
arXiv:1910.06070v2 fatcat:cki7rmrukzd45gzxwj3fowcd2i

Integration and Fusion of Geologic Hazard Data under Deep Learning and Big Data Analysis Technology

Feng He, Chunxue Liu, Hongjiang Liu, Zhihan Lv
2021 Complexity  
To discuss the analysis and evaluation of highway landslides, the application of data mining methods combined with deep learning frameworks in geologic hazard evaluation and monitoring is explored preliminarily  ...  Finally, an entropy weight gray clustering evaluation method based on data mining analysis is proposed, and the performances of several methods are verified.  ...  Acknowledgments is work was supported by the Scientific Research Fund Project of Yunnan Provincial Department of Education "Earthquake relief asymmetric information game dynamics model in complicated landforms  ... 
doi:10.1155/2021/2871770 fatcat:jaqldqvw4favrkqpvpcholygei
« Previous Showing results 1 — 15 out of 11,679 results