67 Hits in 2.4 sec

Learning to overtake in TORCS using simple reinforcement learning

Daniele Loiacono, Alessandro Prete, Pier Luca Lanzi, Luigi Cardamone
2010 IEEE Congress on Evolutionary Computation  
We applied simple reinforcement learning, namely Q-learning, to learn both these overtaking behaviors.  ...  We tested our approach in several overtaking situations and compared the learned behaviors against one of the best NPC provided with TORCS.  ...  Then, we applied Q-learning [13] , a simple but widely used reinforcement learning approach, to develop a sophisticated overtaking behavior which we integrated in our behavior-based architecture.  ... 
doi:10.1109/cec.2010.5586191 dblp:conf/cec/LoiaconoPLC10 fatcat:yreeg7nkdzhsnpxjf4phc6laey

Deep Reinforcement Learning for Autonomous Driving [article]

Sen Wang, Daoyuan Jia, Xinshuo Weng
2019 arXiv   pre-print
Reinforcement learning has steadily improved and outperform human in lots of traditional games since the resurgence of deep neural network.  ...  In order to fit DDPG algorithm to TORCS, we design our network architecture for both actor and critic inside DDPG paradigm.  ...  Here we only discuss recent advances in autonomous driving by using reinforcement learning or deep learning techniques.  ... 
arXiv:1811.11329v3 fatcat:ff57x3jd7zbpfazjhvenshzaka

Learning Driving Behaviors for Automated Cars in Unstructured Environments [chapter]

Meha Kaushik, K. Madhava Krishna
2019 Lecture Notes in Computer Science  
We use DDPG with intelligent choice of reward function and exploration policy to learn various driving behaviors(Lanekeeping, Overtaking, Blocking, Defensive, Opportunistic) for a simulated car in unstructured  ...  The core of Reinforcement learning lies in learning from experiences. The performance of the agent is hugely impacted by the training conditions, reward functions and exploration policies.  ...  They have used simple Q-Learning for the same. [18] also uses Q-Learning to learn overtaking behaviors. Most of the previous work is based out of Deep Q-Networks and Q-Learning.  ... 
doi:10.1007/978-3-030-11021-5_36 fatcat:wdobavfvzjht7nz7b2fbc37tbm

Transfer of driving behaviors across different racing games

Luigi Cardamone, Antonio Caiazzo, Daniele Loiacono, Pier Luca Lanzi
2011 2011 IEEE Conference on Computational Intelligence and Games (CIG'11)  
learned in TORCS.  ...  In this paper, we investigate how to transfer driving behaviors from The Open Racing Car Simulator (TORCS) to VDrift, which are two well known open-source racing games featuring rather different physics  ...  The overtaking issue was consider also in [18] using a modular fuzzy architecture and in [15] using Reinforcement Learning.  ... 
doi:10.1109/cig.2011.6032011 dblp:conf/cig/CardamoneCLL11 fatcat:vyprpbm6qfbhnbodkyswqmv5wi

Exploring applications of deep reinforcement learning for real-world autonomous driving systems [article]

Victor Talpaert, Ibrahim Sobh, B Ravi Kiran, Patrick Mannion, Senthil Yogamani, Ahmad El-Sallab, Patrick Perez
2019 arXiv   pre-print
We first provide an overview of the tasks in autonomous driving systems, reinforcement learning algorithms and applications of DRL to AD systems.  ...  Deep Reinforcement Learning (DRL) has become increasingly powerful in recent years, with notable achievements such as Deepmind's AlphaGo.  ...  Multiple-goal RL for overtaking: In (Ngai and Yung, 2011) a multiple-goal reinforcement learning (MGRL) framework is used to solve the vehicle overtaking problem.  ... 
arXiv:1901.01536v3 fatcat:y3gck5rznjglvim4gem5dvb2ue

MADRaS : Multi Agent Driving Simulator [article]

Anirban Santara, Sohan Rudra, Sree Aditya Buridi, Meha Kaushik, Abhishek Naik, Bharat Kaul, Balaraman Ravindran
2020 arXiv   pre-print
MADRaS uses multiprocessing to run each agent as a parallel process for efficiency and integrates well with popular reinforcement learning libraries like RLLib.  ...  MADRaS can be used to create driving tasks whose complexities can be tuned along eight axes in well-defined steps. This makes it particularly suited for curriculum and continual learning.  ...  Prajapat of ETH Zurich for his useful tips on the implementation of inter-vehicular communication in MADRaS.  ... 
arXiv:2010.00993v1 fatcat:fg5gth3jn5g5rgaa5go3yk6bhq

Evolving competitive car controllers for racing games with neuroevolution

Luigi Cardamone, Daniele Loiacono, Pier Luca Lanzi
2009 Proceedings of the 11th Annual conference on Genetic and evolutionary computation - GECCO '09  
In particular, we focused on The Open Car Racing Simulator (TORCS), an open source car racing simulator, already used as a platform for several scientific competitions dedicated to games.  ...  We suggest that a competitive controller should have two basic skills: it should be able to drive fast and reliably on a wide range of tracks and it should be able to effectively overtake the opponents  ...  In an early work, Pyeatt and Howe [12] applied reinforcement learning to learn racing behaviors in RARS, an open source car racing simulator.  ... 
doi:10.1145/1569901.1570060 dblp:conf/gecco/CardamoneLL09 fatcat:5tcemhhp6rd3nixviz67wiayn4

Overtaking opponents with blocking strategies using fuzzy logic

Enrique Onieva, Luigi Cardamone, Daniele Loiacono, Pier Luca Lanzi
2010 Proceedings of the 2010 IEEE Conference on Computational Intelligence and Games  
The behavior we developed has been integrated in an existing fuzzy-based architecture for driving simulated cars and tested using The Open Car Racing Simulator (TORCS).  ...  In car racing, blocking refers to maneuvers that can prevent, disturb or possibly block an overtaking action by an incoming car.  ...  [14] applied reinforcement learning to learn two overtaking behaviors: (i) overtaking on a straight exploiting the drag effect of the opponent; (ii) overtaking close to a turn using braking delay.  ... 
doi:10.1109/itw.2010.5593364 dblp:conf/cig/OnievaCLL10 fatcat:n7v3cbcc3rgyrfsf4x6vmxwex4

Closing the gap towards end-to-end autonomous vehicle system [article]

Yonatan Glassner, Liran Gispan, Ariel Ayash, Tal Furman Shohet
2019 arXiv   pre-print
In this paper, we amend the basic E2E architecture to address these shortcomings, while retaining the power of end-to-end learning.  ...  We analyze the effect of each concept and present driving performance in a highway scenario in the TORCS simulator. Video is available in this link:  ...  Conditional value at risk (CVaR) is a risk measure used in finance and lately adopted for the reinforcement learning setting [21] .  ... 
arXiv:1901.00114v2 fatcat:pi3h4d4mp5auxmtiace7zvjpgq

Gaze Training by Modulated Dropout Improves Imitation Learning [article]

Yuying Chen, Congcong Liu, Lei Tai, Ming Liu, Bertram E. Shi
2019 arXiv   pre-print
Prediction error in steering commands is reduced by 23.5% compared to uniform dropout.  ...  Imitation learning by behavioral cloning is a prevalent method that has achieved some success in vision-based autonomous driving.  ...  [15] learned driving policy by reinforcement learning in the TORCS simulator and transfered the learned policy to real-world driving data through virtual to real image translation network.  ... 
arXiv:1904.08377v2 fatcat:yleqwqx2nfbjbab5ela26p67ne

Learn to Make Decision with Small Data for Autonomous Driving: Deep Gaussian Process and Feedback Control

Wenqi Fang, Shitian Zhang, Hui Huang, Shaobo Dang, Zhejun Huang, Wenfei Li, Zheng Wang, Tianfu Sun, Huiyun Li
2020 Journal of Advanced Transportation  
Compared with the amount of training data in deep reinforcement learning, our method uses only 0.34% of its size and obtains similar simulation results.  ...  It may be useful for real road tests in the future.  ...  Acknowledgments e authors would like to thank Junta Wu's, who is the author of AMDDPG, help for training the DDPG network, supplying the data, and providing many helpful clarifications and suggestions.  ... 
doi:10.1155/2020/8495264 fatcat:iqr37tuxsne2tps35xazvpt5du

Survey of Deep Reinforcement Learning for Motion Planning of Autonomous Vehicles [article]

Szilárd Aradi
2020 arXiv   pre-print
A wide range of techniques in Machine Learning itself have been developed, and this article describes one of these fields, Deep Reinforcement Learning (DRL).  ...  Academic research in the field of autonomous vehicles has reached high popularity in recent years related to several topics as sensor technologies, V2X communications, safety, security, decision making  ...  ACKNOWLEDGMENT The research reported in this paper was supported by the Higher Education Excellence Program of the Ministry of Human Capacities in the frame of Artificial Intelligence research area of  ... 
arXiv:2001.11231v1 fatcat:l6l2ptyyxjc3dhseza5bsleoje

Learning drivers for TORCS through imitation using supervised methods

Luigi Cardamone, Daniele Loiacono, Pier Luca Lanzi
2009 2009 IEEE Symposium on Computational Intelligence and Games  
In this paper, we apply imitation learning to develop drivers for The Open Racing Car Simulator (TORCS).  ...  Our approach can be classified as a direct method in that it applies supervised learning to learn car racing behaviors from the data collected from other drivers.  ...  Note that, while in [14] , [9] , [10] , [6] NEAT is applied to learn a controller, in this case, NEAT is used to learn a predictive model from the logs collected during several runs of TORCS. return  ... 
doi:10.1109/cig.2009.5286480 dblp:conf/cig/CardamoneLL09 fatcat:jewvxvdxzrf5plvtsrrujzevjm

DeepDriving: Learning Affordance for Direct Perception in Autonomous Driving [article]

Chenyi Chen, Ari Seff, Alain Kornhauser, Jianxiong Xiao
2015 arXiv   pre-print
To demonstrate this, we train a deep Convolutional Neural Network using recording from 12 hours of human driving in a video game and show that our model can work well to drive a car in a very diverse set  ...  Our representation provides a set of compact yet complete descriptions of the scene to enable a simple controller to drive autonomously.  ...  This work is partially supported by gift funds from Google, Intel Corporation and Project X grant to the Princeton Vision Group, and a hardware donation from NVIDIA Corporation.  ... 
arXiv:1505.00256v3 fatcat:yhog6uyrgrdbdkkxjvti3kkqkq

A Method of Personalized Driving Decision for Smart Car Based on Deep Reinforcement Learning

Xinpeng Wang, Chaozhong Wu, Jie Xue, Zhijun Chen
2020 Information  
Finally, compare the proposed method with classic Deep Reinforcement Learning algorithms.  ...  To date, automatic driving technology has become a hotspot in academia. It is necessary to provide a personalization of automatic driving decision for each passenger.  ...  All authors have read and agreed to the published version of the manuscript. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/info11060295 fatcat:6h7r4ibvqbagnee2vqzpx3wnii
« Previous Showing results 1 — 15 out of 67 results