Abstract:A new optimization technique for dynamic system is proposed to achieve autonomous control under complicated environment. Firstly, Dynamic Bayesian Network (DBN) is incorporated into evolutionary algorithm as a transfer network from t to t+1 generation. Through DBN, the original static optimization process of evolutionary algorithm based on Bayesian optimization algorithm (BOA) is effectively changed into the dynamic process. Using this scheme, the DBN transfer network can reestablish optimization direction for system to adapt to various changes of environment. The scheme can help agent to complete a series of complex tasks without intervention from users. The experimental results clearly demonstrate the accuracy and effectiveness of method. Secondly, new concepts are introduced to increase optimization speed and meet realtime requirement. One is Restriction Function, which is used to cut off unnecessary nodes during evolutionary computation, and the other is Replacement, which is used to inherit part of good results of former generation evolutionary. The new concepts are used to make the evolutionary optimization process more efficient .
[1] Pachter M, Chandler P R. Challenges of Autonomous Control. IEEE Control Systems Society Magazine, 1998,18(4): 92-97 [2] Yang H. Investigation of Advanced Unmanned Aerial Vehicle Flight Control Techniques. Flight Dynamics, 2002, 20(1): 1-4 (in Chinese) (杨 晖.先进无人机飞行控制技术研究.飞行力学, 2002, 20(1): 1-4) [3] Breitner M H, et al. Robust Optimal On-Board Reentry Guidance of a Space Shuttle: Dynamic Game Approach and Guidance Synthesis via Neural Networks. Joural of Optimization Theory and Applications, 2000,107(3): 481-503 [4] Fu X W, Gao X G. Study on a Kind of Path Planning Algorithm for UAV. Journal of System Simulation, 2004, 16(1): 20-21,34 (in Chinese) (符小卫,高晓光.一种无人机路径规划算法研究.系统仿真学报, 2004, 16(1): 20-21,34) [5] Pelikan M, Goldberg D E, Cantu-Paz E. Linkage Problem, Distribution Estimation and Bayesian Networks. Evolutionary Computation, 2000, 8(3): 311- 340 [6] Pelikan M, Goldberg D E, Tsutsui S. Combining the Strengths of the Bayesian Optimization Algorithm and Adaptive Evolution Strategies. In: Proc of the Genetic and Evolutionary Computation Conference. San Francisco, USA: Morgan Kaufmann, 2002, 512-519 [7] Harik G R, Lobo F G, Goldberg D E. The Compact Genetic Algorithm. IEEE Trans on Evolutionary Computation, 1999, 3(4): 287-297 [8] Pavlovi’c V, Frey B J, Huang T S. Time-Series Classification Using Mix-State Dynamic Bayesian Networks. IEEE Trans on Robotics and Automation, 1999,16(3): 609-615 [9] Mühlenbein H, Mahnig T. Evolutionary Synthesis of Bayesian Networks for Optimization. In: Patel M, Honavar V, Balakrishnan K, eds. Advance in the Evolutionary Synthesis of Intelligent Agents. Cambridge, USA: MIT Press, 2001, 429-455 [10] de Bonet J S, Isbell C L, Viola P. MIMIC: Finding Optima by Estimating Probability Densities. In: Mozer M C, Jordan M I, Petsche T, eds. Advances in Neural Information Processing Systems. Cambridge, USA: MIT Press, 2000, 424-430 [11] Rudolph G. Convergence Analysis of Canonical Genetic Algorithms. IEEE Trans on Neural Networks, 1994, 5(1): 96-101 [12] Gao X G, Yang Y L. Initial Path Planning Based on Different Threats for Unmanned Combat Air Vehicles. Acta Aeronautica ET Astronautica Sinica, 2003, 24(5): 435-438 (in Chinese) (高晓光,杨有龙.基于不同威胁体的无人作战飞机初始路径规划. 航空学报, 2003, 24(5): 435-438) [13] Pavlovi’c V, Rehg J M, et al. A Dynamic Bayesian Network Approach to Figure Tracking Using Learned Dynamic Models. In: Proc of the International Conference on Computer Vision. Corfu, Greece, 1999, 94-101 [14] Neal R M, Hinton G E. A New View of the EM Algorithm that Justifies Incremental and Other Variants. In: Jordan M I, ed. Learning in Graphical Models. Dordrecht, Netherlands: Kluwer Academic Publishers, 1998, 355-368 [15] Hanson M L, Harperk K A. An Intelligent Agent for Supervisory Control of Teams of Unmanned Combat Air Vehicles. In: Proc of the AUVSI 2000 Unmanned Systems Symposium and Exhibition. Orlando, USA, 2000, 456-458 [16] Rauch H E. Solutions to the Linear Smoothing Problem. IEEE Trans on Automatic Control, 1963, 8(3): 371-372 [17] Jordan M I, Ghahramani Z,et al. An Introduction to Variational Methods for Graphical Methods. Machine Learning, 1999, 37(2): 183-233