Multi-objective Optimization Of Intelligent Wheelchair Paths Taking Into Account User Preferences

With the increase of the aging population,and various unintentional injuries,more and more disabled people appear.At the same time,China is facing the challenge of aging.It places a huge burden on families and society.In order to improve the quality of the disabled people,the intelligent wheelchair has received more and more attention from people from all walks of life.For the optimization of intelligent wheelchair’s driving path,most studies focus on the length of the path.Although a few studies consider user preference and comfort,the length of the path is not taken into account in the case of user preference.Therefore,a multiobjective optimization method for intelligent wheelchair path considering user preference is proposed.Firstly,four path factors(channel width(CW),obstacles quantity(OQ),obstacle shape(OS)and obstacle distance(OD))were set up in VR virtual simulation environment,and subjects were asked to conduct experiments in the set virtual experiment scene.In different experimental scenarios,users’ preference data for different scenarios were obtained by questionnaire.Then based on the basic concept of evidence network,four path factor variables are taken as nodes of evidence network.The preference questionnaire data were obtained through four path factor experiments as the evidence network input to realize network parameterization.According to the structure of network nodes and network parameters,the network topology is determined,and the intelligent wheelchair path preference recognition framework based on evidence network is constructed.Secondly,A*and artificial potential field fusion method was used to generate the path of a certain scale,calculate the user’s path preference and path length,and build a multi-objective optimization mathematical model.Finally,the pareto-based multi-objective particle swarm optimization is used to optimize the optimization model and a set of optimal solutions is obtained.This set of solutions can realize path length optimization with consideration of user preferences.The results show that the proposed method can realize multi-objective optimization of intelligent wheelchair path while considering user preference.Afterwards,a solution is selected,which realizes the multi-objective optimization of the intelligent wheelchair path under the best consideration of user preference and path length.