Conditional Self-Organizing Map for Hierarchical Classification in Road-Vehicle Situation

  • Jaiprakash Narain Dwivedi


This paper proposes an approach of Conditional Self-Organizing Map (CSOM) algorithm for hierarchical classification. The application of this algorithm is in situation analysis. In this paper, situation is vehicles on the road such as to reduce the risk of collision on the road. The representation of the analysis data is by using Dot Distribution Representation (DDR) and two kinds of the data is being considered for hierarchical classification. First kind of the data is Cross shape road data and second one is object data (that is coordinates of the position of the vehicles) on the road. The first step of the hierarchical classification is to classify the different cross shapes of the road and the next step is to classify the group of number of vehicles corresponding to the cross shapes road. The main objective of hierarchical classification in road-vehicle environment is to anticipate the density of available objects on the road. This prediction can be fed to the decision mechanism of autonomous vehicle for taking decision of either changing the direction of vehicle or reducing the speed of vehicle. Thus the risk of collision of an autonomous vehicle can be minimized.

How to Cite
Dwivedi, J. N. (2020). Conditional Self-Organizing Map for Hierarchical Classification in Road-Vehicle Situation. International Journal of Advanced Science and Technology, 29(3), 3131- 3141. Retrieved from