Application of machine learning methods to route planning and navigation for disabled people

  • Authors:
  • D. Karimanzira;P. Otto;J. Wernstedt

  • Affiliations:
  • Technical University of Ilmenau, Faculty for Computer - and System engineering, Ilmenau, Germany;Technical University of Ilmenau, Faculty for Computer - and System engineering, Ilmenau, Germany;Technical University of Ilmenau, Faculty for Computer - and System engineering, Ilmenau, Germany

  • Venue:
  • MIC'06 Proceedings of the 25th IASTED international conference on Modeling, indentification, and control
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Daily observation and common sense reveal that way finding and travel can be difficult and time consuming for persons with visual/limp impairments. Researchers still have little understanding of what the various degrees of difficulty are for various tasks, and how they affect travel and activity behaviour [1], [2]. This paper focuses on the research for the development of a new travel aid to increase the independent mobility of disabled and elderly people in potentially unfamiliar environments by providing useful information both before and during travelling. The route planning and navigation system consists mainly of four interrelated components: the module to assist travellers in pre-planning their journeys, the module to execute these plans to provide users with general information, orientation and navigation assistance during journeys. The route planning system itself consist of a geo-coding routine [11], a fuzzy decision system to eliminate unfeasible pathways in dependence of the type of disability and another fuzzy decision system to select a mobile equipment according to the users handicaps. The k-best paths are found using a modification of the well-known A* - algorithm. The study employs social science techniques to investigate and quantify urban barriers, and computer technologies to translate these into an operational route assessment and modelling package, using dynamic segmentation and network analysis techniques. This tool is to be applied and tested in Georgenthal, Germany.