Abstract:For the autonomous collision-avoidance of mobile robots in dynamic environments, a neural collision-avoidance decision system consisting of a sensory module, a threat-avoidance module and a target-directed module is constructed. The relative velocity between threats and robot, effectual detection range of lidar, and prediction azimuth of threat within the sampling period are investigated, and their effects on the maneuvering efficiency and security of collision avoidance are analyzed deeply. Based on this, the risk level of each maneuvering area is determined, and a quantitative evaluation model of threat degree is established. Then, it is introduced into the local decision maker to correct the output of each node. Furthermore, a neural dynamic collision-avoidance method based on evaluation of threat degree of the maneuvering strategy is proposed. Simulation experiment results indicate that, compared with traditional neural network based navigation approaches, the proposed approach can optimize avoidance paths with a lower cost.
[1] Savkin A V, Hoy M. Reactive and the shortest path navigation of a wheeled mobile robot in cluttered environments[J]. Robotica, 2013, 31(2):323-330.
[2] Lapierre L, Zapata R, Lepinay P. Combined path-following and obstacle avoidance control of a wheeled robot[J]. International Journal of Robotics Research, 2007, 26(4):361-375.
[3] Huang B Q, Cao G Y, Guo M. Reinforcement learning neural network to the problem of autonomous mobile robot obstacle avoidance[C]//4th International Conference on Machine Learning and Cybernetics. Piscataway, USA:IEEE, 2005:85-89.
[4] Watkins C J C H, Dayan P. Q-learning[J]. Machine Learning, 1992, 8(3-4):279-292.
[5] Yun S C, Parasuraman S, Ganapathy V. Mobile robot navigation:Neural Q-learning[C]//2nd International Conference on Advances in Computing and Information Technology. Berlin, Germany:Springer, 2013:259-268.
[6] Figueroa J, Posada J, Soriano J, et al. A type-2 fuzzy logic controller for tracking mobile objects in the context of robotic soccer games[C]//IEEE International Conference on Fuzzy Systems. Piscataway, USA:IEEE, 2005:359-364.
[7] Junratanasiri S, Auephanwiriyakul S, Theera-Umpon N. Navigation system of mobile robot in an uncertain environment using type-2 fuzzy modelling[C]//IEEE International Conference on Fuzzy Systems. Piscataway, USA:IEEE, 2011:1171-1178.
[8] Shi P, Cui Y J. Dynamic path planning for mobile robot based on genetic algorithm in unknown environment[C]//22nd Chinese Control and Decision Conference. Piscataway:IEEE, 2010:4325-4329.
[9] Kang W S, Yun S, Kwon H O, et al. Stable path planning algorithm for avoidance of dynamic obstacles[C]//9th Annual IEEE International Systems Conference. Piscataway, USA:IEEE, 2015:578-581.
[10] Yadav V, Wang X H, Balakrishnan S N. Neural network approach for obstacle avoidance in 3-D environments for UAVs[C]//American Control Conference. Piscataway, USA:IEEE, 2006:3667-3672.
[11] Janglova D. Neural networks in mobile robot motion[J]. International Journal of Advanced Robotic Systems, 2004, 1(1):15-22.
[12] 郑一力,孙汉旭,刘晋浩.球形移动机器人基于神经网络的反馈线性化运动控制研究与实验[J].机器人,2012,34(4):455-459.Zheng Y L, Sun H X, Liu J H. Study and experiment on neural-network-based feedback linearization motion control for a spherical mobile robot[J]. Robot, 2012, 34(4):455-459.
[13] 乔俊飞,樊瑞元,韩红桂,等.机器人动态神经网络导航算法的研究和实现[J].控制理论与应用,2010,27(1):111-115.Qiao J F, Fan R Y, Han H G, et al. Research and realization of dynamic neural network navigation algorithm for mobile robot[J]. Control Theory and Applications, 2010, 27(1):111-115.
[14] Sariff N B, Abd Wahab N H N B. Automatic mobile robot obstacles avoidance in a static environment by using a hybrid technique based on fuzzy logic and artificial neural network[C]//4th IEEE International Conference on Artificial Intelligence with Applications in Engineering and Technology. Piscataway,USA:IEEE, 2014:137-142.
[15] Hacene N, Mendil B. Toward safety navigation in cluttered dynamic environment:A robot neural-based hybrid autonomous navigation and obstacle avoidance with moving target tracking[C]//3rd International Conference on Control, Engineering and Information Technology. Piscataway, USA:IEEE, 2015:6pp.
[16] 任红格,阮晓钢.Skinner操作条件反射的一种仿生学习算法与机器人控制[J].机器人,2010,32(1):132-137.Ren H G, Ruan X G. A bionic learning algorithm based on Skinner's operant conditioning and control of robot[J]. Robot, 2010, 32(1):132-137.
[17] Glasius R, Komoda A, Gielen S C A M. Neural-network dynamics for path planning and obstacle avoidance[J]. Neural Networks, 1995, 8(1):125-133.
[18] Yang S X, Yuan G F, Meng M Q H. Real-time collision-free path planning and tracking control of a nonholonomic mobile robot using a biologically inspired approach[C]//IEEE International Symposium on Computational Intelligence in Robotics and Automation. Piscataway, USA:IEEE, 2001:113-118.
[19] Yang S X, Meng M Q H. Real-time collision-free motion planning of a mobile robot using a neural dynamics-based approach[J]. IEEE Transactions on Neural Networks, 2003, 14(6):1541-1552.
[20] 朱大奇,孙兵,李利.基于生物启发模型的AUV三维自主路径规划与安全避障算法[J].控制与决策,2015,30(5):798-806.Zhu D Q, Sun B, Li L. Algorithm for AUV's 3-D path planning and safe obstacle avoidance based on biological inspired model[J]. Control and Decision, 2015, 30(5):798-806.
[21] Grossberg S. Nonlinear neural networks:Principles, mechanisms, and architectures[J]. Neural Networks, 1988, 1(1):17-61.
[22] Peng Y, Qu D, Zhong Y X, et al. The obstacle detection and obstacle avoidance algorithm based on 2-D Lidar[C]//IEEE International Conference on Information and Automation. Piscataway, USA:IEEE, 2015:1648-1653.
[23] Li X R, Jilkov V P. Survey of maneuvering target tracking. Part I:Dynamic models[J]. IEEE Transactions on Aerospace and Electronic Systems, 2003, 39(4):1333-1364.
[24] 付梦印,邓志红,闫莉平.Kalman滤波理论及其在导航系统中的应用[M].北京:科学出版社,2010:1-81.Fu M Y, Deng Z H, Yan L P. Theory of Kalman filtering and its applications in navigation systems[M]. Beijing:Science Press, 2010:1-81.