A neural network approach to navigation of a mobile robot and obstacle avoidance in dynamic and unknown environments

Mobile robot navigation and obstacle avoidance in dynamic and unknown environments is one of the most challenging problems in the field of robotics. Considering that a robot must be able to interact with the surrounding environment and respond to it in real time, and given the limited sensing range, inaccurate data, and noisy sensor readings, this problem becomes even more acute. In this paper, we attempt to develop a neural network approach equipped with statistical dimension reduction techniques to perform exact and fast robot navigation, as well as obstacle avoidance in such a manner. In order to increase the speed and precision of the network learning and reduce the noise, kernel principal component analysis is applied to the training patterns of the network. The proposed method uses two feedforward neural networks based on function approximation with a backpropagation learning algorithm. Two different data sets are used for training the networks. In order to visualize the robot environment, 180 ◦ laser range sensor (SICK) readings are employed. The method is tested on real-world data and experimental results are included to verify the effectiveness of the proposed method.

PDF

___

[1] Mahadevan S. Machine learning for robots: a comparison of different paradigms. In: IEEE/RSJ 1996 International Conference on Intelligent Robots and Systems; 8 November 1996; Osaka, Japan. New York, NY, USA: IEEE. pp. 1-14.
[2] Tang L, Dian S, Gu G, Zhou K, Wang S, Feng X. A novel potential field method for obstacle avoidance and path planning of mobile robot. In: IEEE 2010 International Conference on Computer Science and Information Technology; 911 July 2010; Chengdu, China. New York, NY, USA: IEEE. pp. 633-637.
[3] Wang Y, Chirikjian GS. A new potential field method for robot path planning. In: IEEE 2000 International Conference on Robotics and Automation; 2428 April 2000; San Francisco, CA, USA. New York, NY, USA: IEEE. pp. 977-982.
[4] Alander JT. On robot navigation using a genetic algorithm. In: Albrecht RF, Reeves C, Steele NC, editors. Artificial Neural Nets and Genetic Algorithms. Innsbruck, Austria: Springer, 1993. pp. 471-478.
[5] Manikas W, Ashenayi K, Wainwright R. Genetic algorithms for autonomous robot navigation. Instrum Meas Mag 2007; 10: 26-31.
[6] Geisler T, Manikas TW. Autonomous robot navigation system using a novel value encoded genetic algorithm. In: IEEE 2002 Midwest Symposium on Circuits and Systems; 47 August 2002; Tulsa, OK, USA. New York, NY, USA: IEEE. pp. III-45.
[7] Zhang Y, Gong DW, Zhang JH. Robot path planning in uncertain environment using multi-objective particle swarm optimization. Neurocomputing 2013; 103: 172-185.
[8] Qin YQ, Sun DB, Li N, Ma Q. Path planning for mobile robot based on particle swarm optimization. Robot 2004; 26: 222-225.
[9] Breed MD, Moore J. Animal Behavior. 2nd ed. San Diego, CA, USA: Academic Press, 2016.
[10] Haykin SS. Neural Networks and Learning Machines. 3rd ed. Upper Saddle River, NJ, USA: Pearson Education, 2009.
[11] Hagan MT, Demuth HB, Beale MH, De Jes´us O. Neural Network Design. Boston, MA, USA: PWS, 1996.
[12] Wang Y, Wang S, Tan R, Jiang Y. Motion control of a wheeled mobile robot using digital acceleration control method. Int J Innov Comput I 2013; 9: 387-396.
[13] Parhi DR, Singh MK. Real-time navigational control of mobile robots using an artificial neural network. J Mech Eng Sci 2009; 7: 1713-1725.
[14] Dahm P, Bruckhoff C, Joublin F. A neural field approach for robot motion control. In: IEEE 1998 International Conference on Systems, Man, and Cybernetics; 1114 October 1998; San Diego, CA, USA. New York, NY, USA: IEEE. pp. 3460-3465.
[15] Zou AM, Hou ZG, Fu SY, Tan M. Neural networks for mobile robot navigation: a survey. In: Wang J, Yi Z, Zurada JM, Lu BL, Hujun Y, editors. Advances in Neural Networks. Chengdu, China: Springer, 2006. pp. 1218-1226.
[16] Chi KH, Lee MF. Obstacle avoidance in mobile robot using Neural Network. In: IEEE 2011 International Conference on Consumer Electronics, Communications and Networks; 1618 April 2011; Xianning, China. New York, NY, USA: IEEE. pp. 5082-5085.
[17] Ortega JG, Camacho EF. Mobile robot navigation in a partially structured static environment, using neural predictive control. Control Eng Pract 1996; 12: 1669-1679.
[18] Jung IK, Hong KB, Hong SK, Hong SC. Path planning of mobile robot using neural network. In: IEEE 1999 International Symposium on Industrial Electronics; 1216 July 1999; Bled, Slovenia. New York, NY, USA: IEEE. pp. 979-983.
[19] Ni J, Li X, Fan X, Shen J. A dynamic risk level based bioinspired neural network approach for robot path planning. In: IEEE 2014 World Automation Congress; 37 August 2014; Waikoloa, HI, USA. New York, NY, USA: IEEE. pp. 829-833.
[20] Quinonez Y, Ramirez M, Lizarraga C, Tostado I, Bekios J. Autonomous robot navigation based on pattern recognition techniques and artificial neural networks. In: Ferr´andez Vicente JM, Alvarez-S´anchez JR, de la Paz ´ L´opez F, Toledo-Moreo FJ, Adeli H, editors. Bioinspired Computation in Artificial Systems. Elche, Spain: Springer, 2015. pp. 320-329.
[21] Fierro R, Lewis FL. Control of a nonholonomic mobile robot using neural networks. IEEE T Neural Networ 1998; 7: 589-600.
[22] Chen Xi, El Kamel A. A reinforcement learning method of obstacle avoidance for industrial mobile vehicles in unknown environments using neural network. In: Qi E, Jiang SH, Runliang D, editors. Proceedings of the 21st International Conference on Industrial Engineering and Engineering Management. Amsterdam, the Netherlands: Atlantis Press, 2015. pp. 671-675.
[23] Pomerleau DA. Neural network based autonomous navigation. In: Charles E, editor. Vision and Navigation: The Carnegie Mellon Navlab. Pittsburg, PA, USA: Springer, 1990. pp. 83-93.
[24] Crowley JL, Wallner F, Schiele B. Position estimation using principal components of range data. In: IEEE 1998 International Conference on Robotics and Automation; 1620 May 1998; Leuven, Belgium. New York, NY, USA: IEEE. pp. 3121-3128.
[25] Vlassis N, Motomura Y, Kr¨ose B. Supervised dimension reduction of intrinsically low-dimensional data. Neural Comput 2002; 1: 191-215.
[26] Dezfoulian SH, Wu D, Ahmad IS. A generalized neural network approach to mobile robot navigation and obstacle avoidance. In: Lee S, Cho H, Yoon KJ, Lee J, editors. Intelligent Autonomous Systems. Jeju Island, Korea: Springer, 2013. pp. 25-42.
[27] Siegwart R, Nourbakhsh IR, Scaramuzza D. Introduction to Autonomous Mobile Robots. Cambridge, MA, USA: MIT Press, 2011.
[28] Sch¨olkopf B, Smola A, M¨uller KR. Kernel principal component analysis. In: Gerstner W, Germond A, Hasler M, Nicoud JD, editors. Artificial Neural Networks. Lausanne, Switzerland: Springer, 1997. pp. 583-588.
[29] Sch¨olkopf B, Smola A, M¨uller KR. Nonlinear component analysis as a kernel eigenvalue problem. Neural Comput 1998; 7: 1299-1319.
[30] Schiilkopf B. The kernel trick for distances. In: Todd KL, Thomas GD, Volker T, editors. Advances in Neural Information Processing Systems 13. Cambridge, MA, USA: MIT Press, 2000. pp. 301-307.
[31] Urˇsiˇc P, Kristan M, Skoˇcaj D, Leonardis A. Room classification using a hierarchical representation of space. In: IEEE/RSJ 2012 International Conference on Intelligent Robots and Systems; 712 October 2012; Vilamoura, Portugal. New York, NY, USA: IEEE. pp. 1371-1378.