| The complex and diverse experimental environment of outdoor mobile robot are studied and analyzed. And lane detection algorithm and track offset detection algorithm based on vision information are proposed in solving the occurrence of illumination changes and shadows and water and obstacles and a variety of intersection and the tract offset of robot on the lane.A detection method for unstructured road based on Intersecting Cortical Model(ICM) is proposed to solve the detection of complicated environment and a variety of intersection. The ICM is much closer to the information processing mechanism of biological vision and it can distinguish objective and background dynamically according to the relevance between pixel and its neighbor pixels. On the foundation of detecting road region, a T-type template method is proposed to detect and identify the intersection on the lane, the method can detect if there is an intersection and the type of intersection.Because of accumulated error of motor and wheels and so on, there will be track offset in the process of the moving robot, resulting in the failure of lane detection. Therefore the track detection algorithm is designed based on lane’s edge detecting and following algorithms and monocular measurement of distance focusing on side camera. The algorithm can provide the edge information of the lane and the robot’s horizontal estimate position in the road by monocular camera on both sides of the robot in real time. The edge information of the lane is exacted basing on the combination of Canny operator and Hough transformation. This method used the virtue of the edge which the Canny operator examined having single-pixel width and the virtue of its strong filtering ability. Then Hough transformation is used to fit the straight line. The distance between robot and both sides of lane edge in real time is provided, and then estimate the robot’s horizontal position on the road, so when the robot is yawing and the yawing information is given.The experimental result indicated that the algorithm has better real-time ability and robustness. It is suitable and feasible. |
