Movable robot obstacle avoidance method based on bayesian network model

The invention provides a movable robot obstacle avoidance method based on a bayesian network model. The problem that the obstacle distance and angle are not considered sufficiently during obstacle avoidance of a movable robot in the prior art is solved. The method comprises the steps that a movable robot obstacle avoidance decision rule is built; the bayesian network model for obstacle avoidance behavior processing is built; a fuzzy subjection function of the distances and angles of obstacle targets around the movable robot is built, and target data are subjected to fuzzy processing; the distance and angle data of the fuzzy obstacle targets serve as evidences to be input into the bayesian network model for obstacle avoidance; and the deviation angle with the largest posterior probability value is the rotation angle of the movable robot so as to avoid the obstacles. The bayesian network model is used in obstacle avoidance behavior processing of the movable robot, the needed rotation angle is obtained, and thus