Filters








57 Hits in 5.3 sec

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.  ...  In this paper we apply NeuroEvolution of Augmenting Topologies (NEAT), a well known neuroevolution approach, to evolve competitive non-player characters for a racing game.  ...  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

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)  
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  ...  We focus on a neuroevolution learning framework based on NEAT and compare three different methods of transfer learning: (i) transfer of the learned behaviors; (ii) transfer of the learning process; (iii  ...  [6] , [5] applied neuroevolution to evolve a driving and an overtaking behavior, in a sophisticated 3D car racing simulator (TORCS).  ... 
doi:10.1109/cig.2011.6032011 dblp:conf/cig/CardamoneCLL11 fatcat:vyprpbm6qfbhnbodkyswqmv5wi

Learning to overtake in TORCS using simple reinforcement learning

Daniele Loiacono, Alessandro Prete, Pier Luca Lanzi, Luigi Cardamone
2010 IEEE Congress on Evolutionary Computation  
In this paper, we applied the Behavior Analysis and Training (BAT) methodology to define a behavior-based architecture for the NPCs in The Open Racing Car Simulator (TORCS), a well-known open source racing  ...  In addition, we also show that the same approach can be successfully applied to adapt a previously learned behavior to a dynamically changing game situation.  ...  THE OPEN RACING CAR SIMULATOR The Open Racing Car Simulator (TORCS) [16] is a stateof-the-art open source car racing simulator which provides a sophisticated physics engine, full 3D visualization, several  ... 
doi:10.1109/cec.2010.5586191 dblp:conf/cec/LoiaconoPLC10 fatcat:yreeg7nkdzhsnpxjf4phc6laey

Neuroevolution in Games: State of the Art and Open Challenges [article]

Sebastian Risi, Julian Togelius
2015 arXiv   pre-print
This paper surveys research on applying neuroevolution (NE) to games.  ...  The article also highlights important open research challenges in the field.  ...  NEAT has been applied successfully to a variety of different game domains, from controlling a simulated car in The Open Car Racing Simulator (TORCS) [7] or a team of robots in the NERO game [89] to  ... 
arXiv:1410.7326v3 fatcat:yqynswodpnbgzdf52mlisix3hu

Neuroevolution in Games: State of the Art and Open Challenges

Sebastian Risi, Julian Togelius
2017 IEEE Transactions on Computational Intelligence and AI in Games  
This paper surveys research on applying neuroevolution (NE) to games.  ...  The article also highlights important open research challenges in the field.  ...  ACKNOWLEDGEMENTS We thank the numerous colleagues who have graciously read and commented on versions of this paper, including Kenneth O. Stanley, Julian Miller, Matt Taylor, Mark J.  ... 
doi:10.1109/tciaig.2015.2494596 fatcat:uenp54gg2vffdolr5awox2ayx4

Comparative Study of NeuroEvolution Algorithms in Reinforcement Learning for Self-Driving Cars

Ahmed AbuZekry
2019 European Journal of Engineering Science and Technology  
Different neuroevolution algorithms are used to train deep neural networks to predict the steering angle of a car in a simulated environment.  ...  Neuroevolution algorithms are compared to the Double Deep Q-Learning (DDQN) algorithm. Based on the experimental results, the neuroevolution algorithms show better performance than DDQN algorithm.  ...  These methods were used to train a neural network to drive in The Open Racing Car Simulator (TORCS) [Wymann et al., 2000] .  ... 
doi:10.33422/ejest.2019.09.38 fatcat:6jzxx4f5jjd5tdchrxgdyy534m

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.  ...  driving style using a simple 2D car simulator [4] , [5] and The Open Racing Car Simulator [6] .  ... 
doi:10.1109/cig.2009.5286480 dblp:conf/cig/CardamoneLL09 fatcat:jewvxvdxzrf5plvtsrrujzevjm

Evolving a real-world vehicle warning system

Nate Kohl, Kenneth Stanley, Risto Miikkulainen, Michael Samples, Rini Sherony
2006 Proceedings of the 8th annual conference on Genetic and evolutionary computation - GECCO '06  
First, NEAT was evaluated in a complex, dynamic simulation with other cars, where it outperformed three hand-coded strawman warning policies and generated warning levels comparable with those of an open-road  ...  This paper describes three advances toward evolving neural networks with NEAT (NeuroEvolution of Augmenting Topologies) to warn about such crashes in real-world environments.  ...  BACKGROUND This section describes the simulator (the Robot Auto Racing Simulator, or RARS), the learning algorithm (NeuroEvolution of Augmenting Topologies, or NEAT), and the experimental method by which  ... 
doi:10.1145/1143997.1144273 dblp:conf/gecco/KohlSMSS06 fatcat:252reh2brjecvjnxma42lhdt4e

Controller for TORCS created by imitation

Jorge Munoz, German Gutierrez, Araceli Sanchis
2009 2009 IEEE Symposium on Computational Intelligence and Games  
The first controller is the winner of the WCCI 2008 Simulated Car Racing Competition, and the second one is a hand coded controller that performs a complete lap in all tracks.  ...  This paper is an initial approach to create a controller for the game TORCS by learning how another controller or humans play the game.  ...  The game is TORCS (The Open Racing Car Simulator 1 , see Figure 1 ), an open car racing simulator that is being used in several competitions of AI.  ... 
doi:10.1109/cig.2009.5286464 dblp:conf/cig/MunozGS09 fatcat:z3xvgiq73bgd3ixqfu67xgachu

Robust optimization through neuroevolution

Paolo Pagliuca, Stefano Nolfi, Ziqiang Zeng
2019 PLoS ONE  
It allows to improve performance on an extended version of the double-pole balancing problem, outperform the best available human-designed controllers on a car racing problem, and generate effective solutions  ...  need to adapt to variations).  ...  TORCS (torcs.sourceforge.net) is a state-of-the-art open source car-racing simulator.  ... 
doi:10.1371/journal.pone.0213193 pmid:30822316 pmcid:PMC6396973 fatcat:tonyx76v6rgzxkmb73jpq2fa3i

Evolving a fuzzy controller for a Car Racing Competition

Diego Perez, Gustavo Recio, Yago Saez
2009 2009 IEEE Symposium on Computational Intelligence and Games  
During the last two years, one of the most popular competitions held on Computational Intelligence in Games conferences is the Car Racing Competition.  ...  These contests motivate and encourage researchers to participate on them, and to apply their work areas to specific games.  ...  ACKNOWLEDGMENTS This work was supported in part by the Spanish MCyT project MSTAR, Ref: TIN2008-06491-C04-03.  ... 
doi:10.1109/cig.2009.5286467 dblp:conf/cig/PerezRSI09 fatcat:gyiy2i2ikjhjtobymxqrygcyym

Generative NeuroEvolution for Deep Learning [article]

Phillip Verbancsics, Josh Harguess
2013 arXiv   pre-print
An important goal for the machine learning (ML) community is to create approaches that can learn solutions with human-level capability.  ...  Thus NeuroEvolution combined with other ML methods provides an intriguing area of research that can replicate the processes in nature.  ...  Lanzi, “On-line neuroevolution applied to the open racing car simulator,” in Proceedings of the 2009 IEEE Congress on Evolutionary Computation (IEEE CEC 2009), (Piscataway, NJ, USA), IEEE Press  ... 
arXiv:1312.5355v1 fatcat:fxqzpaxfhnauxhdomk2q5p4ye4

Comparing crossover operators in Neuro-Evolution with crowd simulations

Sunrise Wang, James Gain, Geoff Nitschke
2014 2014 IEEE Congress on Evolutionary Computation (CEC)  
This paper attempts to address this by comparing the performance of a set of crossover operators with a range of probabilities in three simulations: Car Racing, Mouse Bridge Crossing, and a War-Robot Battle  ...  One possible solution to this problem is to use Neuro-Evolution (NE) to evolve the agent models so that the agents behave realistically and the emergent crowd behaviour is controllable.  ...  the car-racing simulation).  ... 
doi:10.1109/cec.2014.6900483 dblp:conf/cec/WangGN14 fatcat:cr7co3ahxzah5dljryc2ci6sfu

Self-Adapting Goals Allow Transfer of Predictive Models to New Tasks [article]

Kai Olav Ellefsen, Jim Torresen
2019 arXiv   pre-print
We demonstrate that this allows the predictive model to transfer to new scenarios where goals are different, and that the adaptive goals can even adjust agent behavior on-line, changing its strategy to  ...  One reason this is difficult is the challenge of learning accurate models of an environment.  ...  This has been done successfully for simulated car racing [7] , for learning to aim and shoot in a video game [11] and for a health-gathering VizDoom scenario [1] .  ... 
arXiv:1904.02435v2 fatcat:mhkyehmllzecfmhxbmmqmvjbve

Interactive evolution for the procedural generation of tracks in a high-end racing game

Luigi Cardamone, Daniele Loiacono, Pier Luca Lanzi
2011 Proceedings of the 13th annual conference on Genetic and evolutionary computation - GECCO '11  
We present a framework for the procedural generation of tracks for a high-end car racing game (TORCS) using interactive evolution.  ...  The framework maintains multiple populations and allow users to work both on their own population (in single-user mode) or to collaborate with other users on a shared population.  ...  TORCS The Open Racing Car Simulator (TORCS) [1] is a stateof-the-art open-source car racing simulator which provides a sophisticated physics engine, full 3D visualization, several tracks, car models,  ... 
doi:10.1145/2001576.2001631 dblp:conf/gecco/CardamoneLL11 fatcat:yxy6hjjmtffhfntilkpvi7uowm
« Previous Showing results 1 — 15 out of 57 results