Path Preference Recognition For Intelligent Robotic Wheelchair Based On Evidence Network

As we all know,the aging of the population is becoming more and more serious in China.In April 2011,the data from the sixth census showed that the number of elderly people over 60 years reached 178 million in China,in which account for 13.26% of the total population.The date shown that the fifth population census data of 2000 rose by 2.93%.At the end of 2010,The China Aging Research Center's survey data showed that there were 33 million disable people in China and account for 19% of the elderly population.There are 10.8 million elderly people that are totally disabled elderly and need to care,accounting for 6.23 percent of the country's elderly population.Due to the loss in self-care ability of the disabled elderly,the need for care and long-term care for the elderly will increase in the future.In the case of an aging population,the intelligent robot wheelchair can improve the quality of life of the disabled and the elderly who cannot walk independently.How to choose the favorite path for the user is very important in the process of helping these users move from one location to another.And existing method is mainly to provide users with a safe path,but it is not enough,if you want to allow the user to accept,choice of path not only safe,comfortable and fast,and should be the path of the user favorite.However,because of the differences between users,the smart wheelchair is difficult to identify preferred paths for different users.On this basis,this paper proposes a path selection bias identification method based on the evidence network.The intelligent robot wheelchair is used to simulate human thought and make decisions instead of users.This article extracted user preference data from test and questionnaire.These data are affected by four path characteristics: path width,number of obstacles,obstacle shape and distance between users and obstacles.Firstly,the user's preference data of different paths is collected,and the data input into the evidence network is integrated to obtain the user's preference.The results show that there is a correlation between the user and the proposed method.