Abstract
The N-stages control problem for isolated signalized intersections is defined as the control problem to disperse initial queue lengths to their optimal steady-state values in N cycles. Based on a discrete-event model of a simplified isolated signalized intersection, the N-stages control problem is formulated as a linear programming problem as well as a quadratic programming problem. A new algorithm is proposed for solving the discrete optimization problem by simple calculations, based on the optimal solution of the corresponding continuous-time problem. Numerical comparisons between the continuous-time optimal solution and the discrete-event optimal solutions, obtained from linear programming and sequential quadratic programming, are given for a few examples.
| Original language | English |
|---|---|
| Pages (from-to) | 1553-1563 |
| Number of pages | 11 |
| Journal | Control Engineering Practice |
| Volume | 21 |
| Issue number | 11 |
| DOIs | |
| State | Published - Nov 2013 |
Keywords
- Discrete-event systems
- Optimal control
- Optimization
- Traffic signal control
- Transient analysis
All Science Journal Classification (ASJC) codes
- Applied Mathematics
- Electrical and Electronic Engineering
- Control and Systems Engineering
- Computer Science Applications
