Navigation Dialog of Blind People: Recovery from Getting Lost

Proceedings of the of the EACL 2014 Workshop on Dialogue in Motion (DM), pages 58–62, Gothenburg, Sweden, April 26-30 2014. c⃝2014 Association for Computational Linguistics Navigation Dialog of Blind People: Recovery from Getting Lost Jan Vystrcil, Ivo Maly, Jan Balata and Zdenek Mikovec Czech Technical University in Prague, Faculty Of Electrical Engineering Karlovo nam. 13, 121 35 Praha 2 Czech Republic {vystrjan, malyi1, balatjan, xmikovec}@fel.cvut.cz Abstract Navigation of blind people is different from the navigation of sighted people and there is also difference when the blind per- son is recovering from getting lost. In this paper we focus on qualitative analysis of dialogs between lost blind person and nav- igator, which is done through the mobile phone. The research was done in two out- door and one indoor location. The analy- sis revealed several areas where the dialog model must focus on detailed information, like evaluation of instructions provided by blind person and his/her ability to reliably locate navigation points. 1 Introduction When blind people travel independently, it may happen that they get lost. This happens when they can not find any useful navigation points. Use of existing GPS based navigation systems is of no use as the maps do not provide appropriate navigation points and the GPS localization is imprecise (tens of meters, where the lost blind person needs preci- sion at highest in meters). In such situation blind people typically use one of two following meth- ods to recover. First they can ask people in their surrounding for help. Second they can call friend or dedicated assistance center. The first method is currently more favorable for blind people, but they have experience with both methods. In each method the dialog has different structure due to the different context information available to the help- ing person (called navigator) and lost blind person. In our research we focus on the second method, navigation through mobile phone call. Balata et al. (2013b) showed that such method is usable and navigator can successfully guide blind person in outdoor environment. This is because the blind person is able to efficiently describe his/her po- sition. Balata et al. (2013a) found that there is quite good coverage of locations that are very well known to blind persons and that they should be able to navigate other lost blind person there. These findings show that building some kind of assistance center where blind people can help each other is a promising idea. Our intention is to ex- tend such a center in a way that the helping per- son will be replaced by natural language based di- alog system. According to Pittermann (2005) this dialog system belongs to the category “Dialog as purposeful activity” with overlapping to the cate- gory “Dialogue as collaborative activity”. The key questions we focus on are: • How the selection of an appropriate form of language depends on aspects of the environ- ment? • What is the structure of the dialog with re- spect to the environment? In the initial step …

Content truncated to 500 words