Journal of Zhejiang University-SCIENCE A

, Volume 7, Supplement 2, pp 287–292

Fractal modelling of off-road terrain oriented to vehicle virtual test

  • Wang Qian-ting 
  • Guo Jian 
  • Chen Yi-zhi 

DOI: 10.1631/jzus.2006.AS0287

Cite this article as:
Wang, Q., Guo, J. & Chen, Y. J. Zhejiang Univ. - Sci. A (2006) 7(Suppl 2): 287. doi:10.1631/jzus.2006.AS0287


In order to reconstruct typical off-road terrain surface for vehicle performance virtual test, a terrain generation method with controllable roughness was proposed based on fractal dimension. Transverse profile sampling and unevenness characteristics of typical off-road terrain were discussed according to the choices of appropriate wavelength and sampling interval. Since the off-road terrain in virtual environment is self-similar, the method of calculating the discrete fractal Gauss noise and its auto-correlation function were analyzed. The terrain surface fractal dimension was estimated by determining the Hurst coefficient. As typical off-road terrain is rugged terrain, the method of reconstructing it using fractal modelling is presented. The steps include calculating statistical variations in the absolute value of the difference in elevation between two points, plotting the points in log-log space, identifying linear segments and estimating fractal dimension from the linear segments slope. The constructed surface includes information on potholes, bumps, trend and unevenness of terrain, and can be used as the excitation of vehicle performance virtual test.

Key words

Off-road terrain Transverse profile of terrain Terrain surface reconstruction Fractal dimension 

CLC number


Copyright information

© Zhejiang University 2006

Authors and Affiliations

  • Wang Qian-ting 
    • 1
  • Guo Jian 
    • 2
  • Chen Yi-zhi 
    • 3
  1. 1.School of Mechanical and Energy EngineeringZhejiang UniversityHangzhouChina
  2. 2.Department of Civil EngineeringZhejiang UniversityHangzhouChina
  3. 3.College of Statistics & Computing ScienceZhejiang Gongshang UniversityHangzhouChina

Personalised recommendations