Reactive Collision Avoidance using Evolutionary Neural Networks

Hesham M. Eraqi, Youssef Emad Eldin, Mohamed N. Moustafa

2016

Abstract

Collision avoidance systems can play a vital role in reducing the number of accidents and saving human lives. In this paper, we introduce and validate a novel method for vehicles reactive collision avoidance using evolutionary neural networks (ENN). A single front-facing rangefinder sensor is the only input required by our method. The training process and the proposed method analysis and validation are carried out using simulation. Extensive experiments are conducted to analyse the proposed method and evaluate its performance. Firstly, we experiment the ability to learn collision avoidance in a static free track. Secondly, we analyse the effect of the rangefinder sensor resolution on the learning process. Thirdly, we experiment the ability of a vehicle to individually and simultaneously learn collision avoidance. Finally, we test the generality of the proposed method. We used a more realistic and powerful simulation environment (CarMaker), a camera as an alternative input sensor, and lane keeping as an extra feature to learn. The results are encouraging; the proposed method successfully allows vehicles to learn collision avoidance in different scenarios that are unseen during training. It also generalizes well if any of the input sensor, the simulator, or the task to be learned is changed.

References

  1. Ahmadizar, Fardin, Soltanian, Khabat, AkhlaghianTab, Fardin, Tsoulos, Loannis. 2015. Artificial neural network development by means of a novel combination of grammatical evolution and genetic algorithm. In Engineering Applications of Artificial Intelligence 39.
  2. Durand, N., Alliot, J.M., Noailles, J., 1996. Collision avoidance using neural networks learned by genetic algorithms. In Ninth International Conference on Industrial and Engineering Applications of Artificial Intelligence and Expert Systems, Fukuoka.
  3. Fu, W., Hadj-Abdelkader, H., Colle, E., 2013. Reactive collision avoidance using B-spline representation: Application for mobile robot navigation. In Mobile Robots (ECMR), European Conference on, Barcelona.
  4. Liu, H., Stoll, N., Junginger, S., Thurow, K., 2013. Mobile robot for life science automation, In Int. J. Adv. Robot. Syst., vol. 10, pp. 1-14.
  5. Mahajan, Richa, Gaganpreet, Kaur, 2013. Neural networks using genetic algorithms. In International Journal of Computer Applications.
  6. Montana, D., Davis, L., 1989. Training feedforward neural networks using genetic algorithms. In IJCAI'89 Proceedings of the 11th international joint conference on Artificial intelligence.
  7. Schaffer, J. David, Whitley, Darrell, Eshelman, Larry J., 1992. Combinations of genetic algorithms and neural networks: A survey of the state of the art. In IEEE Conference Proceedings.
  8. Sipper, M., Azaria, Y., Hauptman, A., Shichel, Y, 2006. Designing an evolutionary strategizing machine for game playing and beyond. IEEE
  9. Togelius, J., Lucas, S. M., 2006. Evolving robust and specialized car racing skills in Proceedings of the IEEE Congress on Evolutionary Computation.
  10. Vose, M. D., 1999. The Simple Genetic Algorithm: Foundations and Theory. In IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews.
Download


Paper Citation


in Harvard Style

M. Eraqi H., Emad Eldin Y. and N. Moustafa M. (2016). Reactive Collision Avoidance using Evolutionary Neural Networks . In Proceedings of the 8th International Joint Conference on Computational Intelligence - Volume 1: ECTA, (IJCCI 2016) ISBN 978-989-758-201-1, pages 251-257. DOI: 10.5220/0006084902510257


in Bibtex Style

@conference{ecta16,
author={Hesham M. Eraqi and Youssef Emad Eldin and Mohamed N. Moustafa},
title={Reactive Collision Avoidance using Evolutionary Neural Networks},
booktitle={Proceedings of the 8th International Joint Conference on Computational Intelligence - Volume 1: ECTA, (IJCCI 2016)},
year={2016},
pages={251-257},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006084902510257},
isbn={978-989-758-201-1},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 8th International Joint Conference on Computational Intelligence - Volume 1: ECTA, (IJCCI 2016)
TI - Reactive Collision Avoidance using Evolutionary Neural Networks
SN - 978-989-758-201-1
AU - M. Eraqi H.
AU - Emad Eldin Y.
AU - N. Moustafa M.
PY - 2016
SP - 251
EP - 257
DO - 10.5220/0006084902510257