Design And Implementation Of Path Planning For Mecanum Wheel Robot In Narrow Passage

Mobile robot path planning is the important research direction in the field of robot. When the robot used in space-constrained passage, due to the irregular edges of the passage, and there are often obstacles in the passage, therefore it often appears the situation of the low pass rate and needs the path planning. However, the existing methods of path planning are often failed which are used in the narrow passage, and the reason is that the robot needs larger space when it encounters obstacles. Therefore, the study of robot path planning at this scene improves the pass rate and efficiency, and has a high academic significance and application value.Aiming at the problem of low pass rate and efficiency, the paper starts from the analysis of distance and reflection intensity data which is output from laser radar, provides the method of dynamic error correction and puts forward the algorithm of data partition clustering based on reflection intensity. By introducing the local target and virtual gravity, the paper embarks an improved artificial potential field method model, and combines the kinematics model of Mecanum wheel robot to optimize path. Finally, the paper completes the design of Mecanum wheel robot path planning system. The main research contents are as follows:(1) Correction of radar data. Based on the analysis of the characteristics of the error data, this paper sets up the model of prediction compensation and then realizes the correction of error data. The experimental results show that this method improves the accuracy of the laser radar of environment detection, which laid a foundation for environmental data partition clustering.(2) Clustering method of Environmental data. Aiming at the lower obstacle recognition accuracy of common clustering method which is based on the common data partitioning clustering algorithm, the paper analyses the intensity information between different material which is reflected by laser radar. Then builds the environment data model and proposes an improved non-planar ABD clustering segmentation method. The experimental results show that the algorithm is superior to common algorithm both in instantaneity and accuracy.(3) Robot path planning of improved APF. Aiming at the local minimum defects of the traditional artificial potential field method, the paper chooses the local goal of local scene first, and imports the virtual driving force of predicted time to gravity model of APF, then provides an improved artificial potential field method which is based on virtual driving force of predicted time between current position to goal. Finally, the paper combines the kinematics features of Mecanum wheel robot to the path optimization and designs the path planning system of Mecanum wheel robot which is used in narrow passage. The experimental results show that the indicators of pass rate and efficiency are improved when the robot works in narrow passage.Finally, the paper summarizes the research results above, designs and realizes the obstacles avoidance system which is used in the narrow passage. Application results show that this system can identify the obstacles accurately and drive the robot safely from start position to end position.