Road Tracking Using Deep Reinforcement Learning for Self-driving Car Applications

Deep reinforcement learning has received wide attentions recently. It combines deep learning with reinforcement learning and shows to be able to solve unprecedented challenging tasks. This paper proposes an efficient approach based on deep reinforcement learning to tackle the road tracking problem arisen from self-driving car applications. We propose a new neural network which collects input states from forward car facing views and produces suitable road tracking actions. The actions are derived from encoding the tracking directions and movements. We perform extensive experiments and demonstrate the efficacy of our approach. In particular, our approach has achieved 93.94% driving accuracy, outperforming the state-of-the-art approaches in literature.

[1]  Taolue Chen,et al.  Deep finger texture learning for verifying people , 2018, IET Biom..

[2]  Fang Liu,et al.  Reinforcement learning-based feature learning for object tracking , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..

[3]  Mucahit Karaduman,et al.  Deep learning based traffic direction sign detection and determining driving style , 2017, 2017 International Conference on Computer Science and Engineering (UBMK).

[4]  Anil A. Bharath,et al.  Deep Reinforcement Learning: A Brief Survey , 2017, IEEE Signal Processing Magazine.

[5]  Zhe Zhang,et al.  Reinforcement-Learning-Based Intelligent Maximum Power Point Tracking Control for Wind Energy Conversion Systems , 2015, IEEE Transactions on Industrial Electronics.

[6]  Aurobinda Routray,et al.  Real-time eye gaze direction classification using convolutional neural network , 2016, 2016 International Conference on Signal Processing and Communications (SPCOM).

[7]  Jin Young Choi,et al.  Action-Decision Networks for Visual Tracking with Deep Reinforcement Learning , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Vladimir Pavlovic,et al.  Reinforcement Learning for Robust and Efficient Real-World Tracking , 2010, 2010 20th International Conference on Pattern Recognition.

[9]  Mauricio Barahona,et al.  On periodic reference tracking using batch-mode reinforcement learning with application to gene regulatory network control , 2013, 52nd IEEE Conference on Decision and Control.

[10]  Marc Peter Deisenroth,et al.  Deep Reinforcement Learning: A Brief Survey , 2017, IEEE Signal Processing Magazine.

[11]  Carl E. Rasmussen,et al.  Reinforcement learning with reference tracking control in continuous state spaces , 2011, IEEE Conference on Decision and Control and European Control Conference.

[12]  O. Grigore.  Reinforcement learning neural network used in a tracking system controller , 2000, Proceedings 9th IEEE International Workshop on Robot and Human Interactive Communication. IEEE RO-MAN 2000 (Cat. No.00TH8499).

[13]  Clay M. Thompson,et al.  Image processing toolbox [for use with Matlab] , 1995 .

[14]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[15]  Jin Young Choi,et al.  Action-Driven Visual Object Tracking With Deep Reinforcement Learning , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[16]  Xin Zhang,et al.  End to End Learning for Self-Driving Cars , 2016, ArXiv.

[17]  Xue Jinlin,et al.  Neurofuzzy velocity tracking control with reinforcement learning , 2009, 2009 9th International Conference on Electronic Measurement & Instruments.

[18]  Deva Ramanan,et al.  Tracking as Online Decision-Making: Learning a Policy from Streaming Videos with Reinforcement Learning , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[19]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[20]  Etienne Perot,et al.  End-to-End Driving in a Realistic Racing Game with Deep Reinforcement Learning , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[21]  Antonio M. López,et al.  The SYNTHIA Dataset: A Large Collection of Synthetic Images for Semantic Segmentation of Urban Scenes , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).