Research On HEnvironment Recognition And Path Planning For Multi-legged Mobile Robots With Global Vision

With the development of mobile robot technology Hat home and abroad, robot navigationtechnology Hhas become a hot research field. The effective recognition of scene information isthe key of navigation technology based on the principle of human eyes, robot visionnavigation which can recognize the scene information within a Hspecified rangeH has made agreat progress in recent 20 years. However, it is not formed mature theory and methods tocompletely transform 2D information into 3D information. Researchers try their best to seekmore alternative methods to guide the robot motion, and achieved prominent results.Focusing on bottleneck problem of robot vision information transformation, combinedwith template multi-legged walking robot platform, this paper uses a simplified method ofHglobal visionH. Different objects are labeled by different colors, such as red, green, blue, black,etc. Thus vision system can recognize different objects based on the color information. Afterthe target point of robot motion is determined, a reasonable motion path can be planned toachieve the robot vision navigation.Based on these ideas, by using c# language in Microsoft visual studio.net Hintegrateddevelopment environmentH, the Hsimulation platformH of global vision navigation introduced inthe paper is created. Using the camera calibration method of Zhang Zhengyou, a model oftransformation from 2D image information into 3D scene information is established. DHigitalHstore framework of scene image information is set up by using GeneralMatrix classes. Usingmatured image processing technology, scene Hidentification moduleH of global vision iscompiled, and on this base A* algorithm is used to find available HpathH and make navigationplan.The results of experiment show that the Hsimulation platformH of global vision navigationsystem has Hhigh efficiencyH in completing the path recognition and motion planning, andreaches the Hthe initial aimH of our HlaboratoryH vision researcher. The imagine processing module,cameraH calibrationH module and the components of digital information store frameworkmodule in this system have HscalabilityH and actual HapplicabilityH, which can be used in futurework.