An adaptive algorithm for deploying self-tuning traffic control systems

Currently, a considerable amount of human effort and time is spent for initialization or calibration of operational tra ffic control systems. Typically, this optimization (fine-tunin g) procedure is conducted manually, via trial-and-error, r elying on expertise and human judgment and does not always lead to a d esirable outcome. This paper presents a new learning/adaptive algorithm that enables automatic fine-tuning of eneral traffic control systems. The efficiency and online feasibility of the algorithm is investigated through exten sive simulation experiments. The fine-tuning problem of thr ee mutually-interacting control modules – each one with its di tinct design parameters – of an urban traffic signal control strategy is thoroughly investigated. Simulation results i ndicate that the learning algorithm can provide efficient au tomatic fine-tuning, guaranteeing safe and convergent behavior.

[1]  Yu. M. Ermol’ev.  On the method of generalized stochastic gradients and quasi-Féjer sequences , 1969 .

[2]  J. Kiefer,et al.  Stochastic Estimation of the Maximum of a Regression Function , 1952 .

[3]  Markos Papageorgiou,et al.  Extensions and New Applications of the Traffic-Responsive Urban Control Strategy: Coordinated Signal Control for Urban Networks , 2003 .

[4]  J. Spall.  Multivariate stochastic approximation using a simultaneous perturbation gradient approximation , 1992 .

[5]  Elias B. Kosmatopoulos,et al.  International comparative field evaluation of a traffic-responsive signal control strategy in three cities. , 2006 .

[6]  Ron Meir,et al.  Approximation bounds for smooth functions in C(Rd) by neural and mixture networks , 1998, IEEE Trans. Neural Networks.

[7]  Elias B. Kosmatopoulos,et al.  Large Scale Nonlinear Control System Fine-Tuning Through Learning , 2009, IEEE Transactions on Neural Networks.

[8]  Manolis A. Christodoulou,et al.  Correction on 'Dynamical neural networks that ensure exponential identification error convergence' , 1998, Neural Networks.

[9]  Markos Papageorgiou,et al.  A Multivariable Regulator Approach to Traffic-Responsive Network-Wide Signal Control , 2000 .

[10]  Markos Papageorgiou,et al.  A rolling-horizon quadratic-programming approach to the signal control problem in large-scale congested urban road networks , 2008 .

[11]  H. Robbins.  A Stochastic Approximation Method , 1951 .

[12]  Marios M. Polycarpou,et al.  High-order neural network structures for identification of dynamical systems , 1995, IEEE Trans. Neural Networks.

[13]  Chee Kheong Siew,et al.  Universal Approximation using Incremental Constructive Feedforward Networks with Random Hidden Nodes , 2006, IEEE Transactions on Neural Networks.

[14]  Markos Papageorgiou,et al.  Adaptive Fine-Tuning of Nonlinear Control Systems With Application to the Urban Traffic Control Strategy TUC , 2007, IEEE Transactions on Control Systems Technology.

[15]  Markos Papageorgiou,et al.  A rolling-horizon quadratic-programming approach to the signal control problem in large-scale conges , 2009 .

[16]  Manolis A. Christodoulou,et al.  Dynamical Neural Networks that Ensure Exponential Identification Error Convergence , 1997, Neural Networks.