Research On VFH For Mobile Robot Obstacle-avoidance Based On Kalman Filtering Prediction

The intelligent obstacle-avoidance of autonomous mobile robots is the key technology for mobile robots to finish given task autonomously.This thesis researches on mainly two aspects,obstacle-avoiding of mobile robot in real-time and the kalman filtering prediction of the state of obstacles in dynamic uncertain environments.The simulation of obstacle-avoiding algorithm,prediction algorithm and obstacle-avoiding algorithm based on the forecasting model are provided in this paper.Firstly,this thesis summarized the obstacle avoidance approach of mobile robot and indicates the research emphases and its development trend.And then,obstacle avoidance based on Vector Field Histogram method is introduced in detail,the basic idea and the framework of the algorithm are introduced.In VFH method,the work environment of mobile robot is decomposed into a series of grid units with binary information,there is a accumulated value in every rectangular grid which represents the credibility of the existing obstacles,high accumulated value represents high credibility of the existing obstacles,and then the robot searches for the optimal path without collision through the control of the velocity and direction of the robot.Secondly,the prediction of the dynamic obstacles based on kalman filtering approach is studied;the establishment of the kalman filtering prediction model and the steps of kalman filtering prediction is introduced in detail.The simulation results show that the kalman filtering prediction model can forecast the state information of the obstacles effectively and the simulation experiment achieve expected results.This thesis analyzes the VFH method which based on kalman filtering prediction on the basis of the researches on VFH approach and kalman filtering prediction,and the design and algorithm implementation of the VFH method based on kalman filtering prediction are also introduced.The obstacle state information which was forecasted by kalman filtering predictor is one of the inputs of VFH method,and then the robot will get prior knowledge of the next moment state information of the obstacles to improve the accuracy of the obstacle avoidance strategy implementation. Finally,the simulation of the VFH method based on kalman filtering prediction is realized in Matlab Environment.Such essential technology as the design of grid map and the operation interface of the simulation,updating of the obstacle CV value distribution,moving direction updating of the mobile robot,dynamic obstacle avoidance effect demonstration and so on can be realized.And the obstacle avoidance simulation of the VFH method based on kalman filtering prediction in uncertain environments is implemented.Simulation results show that the proposed method is effective and reasonable.Simulation results show the effectiveness of this method,the accuracy and rationality.