Study On Navigation Methods For Intelligent Robots Based On Cognitive Map

Intelligent robots are getting more and more important in daily life of human being recently.They have entered various fields instead of only industrial areas.In the research of Intelligent robots,navigation technologies are one of the core research area,since high-precision navigation technologies can play a central role to safe navigation and efficient completion of tasks.Based on the biological cognitive maps,from the biological navigation point of view,this paper studies theoretical methods for the intelligent robots navigation problem.The main contents can be stated as follows:First of all,we introduce the cells of making up brain network.They are grid cells,place cells,head direction cells,Hebb rules and remapping.Corresponding to a particular spatial location,grid cells conduct repetitive regularity firing.The firing places constitute a node.No matter at which node animals arrive in the space,the grid cell of the node has the largest firing.Place cells are pyramidal neurons.When the animal head in a relatively narrow range,the pyramidal neurons can exhibit site-specific firing.Head direction cells are neurons that rely on head direction.The firing activity is influenced by only head direction and is not affected by the animal's posture,location,and behavior.Then,we apply the simplified model of grid cells to the spatial localization model.Spatial localization function mainly decided by place cells.The simplified model of grid cells and LEC are taken as the outside input of the place cells network model,which constitute the grid cells-place cells Spatial localization model,to achieve location function.After that,we use the place cells to establish a model of partial route exploration.However,the time and place of the model are random.For these two issues,we change it to solve the problems.In such a way,the desired goal is achieved.Finally,the main results of the thesis are summarized and further research problems are pointed out.