Accident data provide some evidence on the relation between geometric design and traffic safety. Horizontal curvature is a main factor, but many potentially intervening variables can be listed. Accident data have not been explicitly used in the derivation of design rules for minimum radii, transition lengths or sight distances. The design rules are mainly based on the characteristics of vehicle-road interaction, assuming error-free steering behaviour. Since the design criteria are rather sensitive to the assumptions made, clearly more insight into actual behaviour is needed. This also applies to the way design speeds are used. This raises the question about the perception of curve characteristics and the control variables of curve driving behaviour. The literature does not contain such experimental evidence on these questions, and only more or less substantial suggestions are given. Driver behaviour while approaching curves is only studied in relation to the presence of warning signs, advisory speed signs or advance information and not in relation to curve characteristics. Usually changes in speed behaviour are taken as the criterion variables. Research on driver behaviour in curves also is not very complete. Vehicle speeds are not constant along a curve but vary in complex ways. Much controversy exists about the causal determination of speed choice. Measurements of actual vehicle paths show also that vehicle paths are not congruent with the road lay-out as is assumed in the design rules, but deviate in a systematic way. Studies of eye movements point to a complex dual control strategy for negotiating curves, and some modelling along these lines is discussed, especially in relation to adverse visibility conditions. Some special problems of curve driving at night are briefly discussed. The review suggests several possible lines of needed research. Of these, the study of forward speed, especially in the approach zone of curves, and the perception of optically given curve characteristics as they dynamically change in the approach of a curve, are considered the primary ones to pursue. (TRRL)

  • Corporate Authors:


    KAMPWEG 5, PO BOX 23
    SOESTERBERG,   Netherlands  3769 ZG
  • Authors:
  • Publication Date: 1984

Media Info

  • Features: Figures; References; Tables;
  • Pagination: 89 p.
  • Serial:
    • Issue Number: C-12

Subject/Index Terms

Filing Info

  • Accession Number: 00395016
  • Record Type: Publication
  • Source Agency: Institute for Road Safety Research, SWOV
  • Report/Paper Numbers: Monograph
  • Files: ITRD, TRIS
  • Created Date: Oct 31 1985 12:00AM