Two-dimensional road shape optimisation using genetic algorithms

P. Liatsis, H. M. Tawfik

Research output: Contribution to journalArticlepeer-review

7 Scopus citations


This research describes the development of a generic road shape modelling algorithm for simulation of traffic scenes. The proposed algorithm simulates horizontal road alignment, according to the road geometry design standards, in the form of alternating straight section and curve segment arrangement. We utilise genetic algorithms to optimise the road shape with regard to real road design considerations such as road curvature and visibility distance. The input to the system is a set of critical points representing points with extreme road curvature. The output is the two-dimensional coordinates of the road centreline (alignment), optimised for driver's safety and comfort. The algorithm produces successful results for various road types and shapes, and exhibits good tolerance to noise.

Original languageBritish English
Pages (from-to)19-31
Number of pages13
JournalMathematics and Computers in Simulation
Issue number1-2
StatePublished - 22 Dec 1999


  • Genetic algorithms
  • Horizontal road alignment
  • Road modelling
  • Road simulation environment


Dive into the research topics of 'Two-dimensional road shape optimisation using genetic algorithms'. Together they form a unique fingerprint.

Cite this