Exceptional Status Detection Of Robot And Negative Elevation Area Detection Under Unknown Environments

To ensure the implement of the control methods and actional strategy in Navigation Control for Mobile Robots under Unknown Environments. It is necessary to detect the status of robot for the exception. This paper is foucus on dealing with several familiar exceptions,include 1) the exception of negative elevation terrain and countermeasure.2)the status of the tyre exception and it's corresponding danger avoiding strategy.First,we introduced the hole system of our experiment, an autonomous robot that could self-navigation in outdoor environment. Implement the 3D detection with a 2D lidar combined with our control mechanism.second, we discuss the exceptional status detection of robot and it's obstacle avoiding strategy.Using four PNN to detect the exceptional status of action. Extracted eight characteristic which reflects the normal and abnormity status of robot(skid,suffocated,blocked), input this characteristic to PNN, and distinguish the status type of tyres. We gived an avoiding strategy based on this,this strategy is suitabe for the obstacles which robot can overcome.The accuracy and efficiency of the method are testified on mobile robot MORCS-1.Last,by switch relevant coordinate, we analyse the status of negative elevation,obtained the highness of detected points under the world coordinate,identify the negative elevation area efficiently,compute the length and width of the area and give a speed contral mathod for robot based on this,then introduced the way how to filting the dynamic obstacle in map biulding,Represented a Linear Interpolation approach of dynamics obstacle filtering in map building.The accuracy and efficiency of the method are testified by the experiments.