论文部分内容阅读
For a robot working in a complicated environment,it is virtually impossible to predict all possible situations and to pre-program the robot with all suitable reaction patterns for each of the possible situations,because the robot is required to act differently in different situations.Furthermore,the robot should be able to adapt to different environments by deciding upon the course of action to take depending on the situation,in addition to pre-registered commands,in a manner similar to humans.However,hardware and the limited computational resources pose a physical limitation,so the robot needs some time to decide its course of action.In this regard,predicting the future state of a robot would be effective.The purpose of this research is to state achieve advance prediction using Online SVR as a learner.Online SVR predicts the future internal robot state-i.e.,the robots next internal state and the appropriate action to be taken.Furthermore,this predictor will help predict the distant future internal robot state,using the internal state and action that the robot adopts repeatedly.This paper presents the results of these studies and discusses methods that allow the robot decide its desirable behavior quickly,using the state predicted.As an application example,we used inverted pendulum “NXTway-GS” model and compared the predicted results with the actual response.