Paper Infomation
The Task Allocation and Path Planning Based on BISOM
Full Text(PDF, 345KB)
Author: Feifei Chen, Daqi Zhu
Abstract: For the task allocation and path planning of multi-AUV (Autonomous Underwater Vehicle) system in three dimensional underwater environments with obstacles, a novel biological inspired self-organizing task allocation and path planning algorithm is proposed. Firstly, the SOM neural network (the self-organizing map) is developed to assign targets in underwater environment to a team of AUVs. The working process involves the definition of the initial neural weights of the SOM network, the rule of selecting the winner, the computation of the neighborhood function. Then, in order to solve the problem of obstacle avoidance and sharp speed jump, the biological inspired neural dynamics model (BINM) is embedded into the SOM neural network, the biological inspired neural dynamic model is used to update weights of the SOM neural network, and plan a path without sharp speed jump and obstacles. As a result, adaptive task allocation and effective path planning are achieved. Finally, to demonstrate the effectiveness of the proposed approach, simulation results are given in this paper.
Keywords: Multi-AUV System; Self-organizing Map (SOM); Task Allocation; Biological Inspired Neural Dynamics Model (BINM); Obstacle Avoidance
References:
[1] O'Hara K J, Walker D B, Balch T R. Physical path planning using a pervasive embedded network[J]. IEEE Transactions on Robotics, 24(3): 741-746. 2008
[2] Parker L E. Heterogeneous multi-robot cooperation[R]. Massachusetts Inst of Tech Cambridge Artificial Intelligence Lab, 1994
[3] Akkiraju R, Keskinocak P, Murthy S, et al. An agent-based approach for scheduling multiple machines[J]. Applied Intelligence, 14(2): 135-144.2001
[4] Turner R M. Intelligent control of autonomous underwater vehicles: IEEE International Conference on Systems, Man and Cybernetics. IEEE, 2: 1717-1722.1995
[5] Zhu A, Yang S X. A neural network approach to dynamic task assignment of multi-robot[J]. IEEE Transactions on Neural Networks, 17(5): 1278-1287.2006
[6] Sato M, Kanda A, Ishii K. A switching controller system for a wheeled mobile robot[J]. Journal of Bionic Engineering, 4(4): 281-289. 2007
[7] Hendzel Z. Collision free path planning and control of wheeled mobile robot using Kohonen self-organizing map[J]. Technical Sciences, 53(1). 2005
[8] Zhu A, Yang S X. An improved som-based approach to dynamic task assignment of multi-robots[C]. World Congress on Intelligent control and Automation. IEEE, 2168-2173. 2010
[9] Chow B. Assigning closely spaced targets to multiple autonomous underwater vehicles[D]. Master Thesis, University of Waterloo, 2009
[10] Zhu DQ, Huang H, Yang S X. Dynamic task assignment and path planning of multi-AUV system based on an improved self-organizing map and velocity synthesis method in three-dimensional underwater workspace[J]. IEEE Transactions on Cybernetics, 43(2): 504-514.2013
[11] Huang H, Zhu DQand DingF. Dynamic task assignment and path planning for multi-AUV system in variable ocean current environment[J], Journal of Intelligent & Robotic Systems. 74(2): 999-1012. 2014
[12] Luo C and YangS X. A Bio-inspired Neural Network for Real-Time Concurrent Map Building and Complete Coverage Robot Navigation in Unknown Environments[J], IEEE Transactions on Neural Networks, 19(7): 1279-1298. 2008