Research On 2D Semantic Map Construction And Path Planning Of Indoor Environment

With the development of robot technology and sensor technology,mobile robots are becoming more and more intelligent.Acquire the environment information of mobile robots while modeling the surrounding environment and planning the path in the established map model.is one of the key technologies for the intelligentization mobile robots.In this paper,the research on the construction and navigation technology of mobile robots has been carried out,and some achievements have been obtained.The main research contents of this paper are as follows:(1)This paper first introduces the establishment methods of various observation models and the selection of experimental platform sensors and hardware and software of mobile robots.Then,based on the robot operating system(ROS),a new mobile robot platform is designed.meanwhile,the mobile robot experimental platform is also found.(2)This paper first proposes a new method to jointly calibrate the monocular camera and the laser radar,complete the joint calibration through the newly designed calibration plate,and achieve certain effects through experiments.Then this paper proposes a semantic construction method of the map,the target object is identified by the monocular camera during the construction of the image,and the laser beam hitting the target object is segmented,and the semantic road sign of the corresponding object is established by counting on the grid corresponding to the grid map,thereby the semantic map of the surrounding environment is established.Finally,the feasibility of the method is verified by the experimental study in the actual environment.(3)Aiming at the problem that the mobile robot passes the traditional A* algorithm path planning through the door and the positioning is lost due to the planned path close to the obstacle,this paper proposes a segmented path planning method based on double-layer map.The method firstly draws a door point layer by pre-processing the already established grid map,and then judges whether the path passes through the gate by calculating the generation value of the path on the door point layer.When it is judged that the path passes through the door,the segment path planning is performed on the original path.Finally,the feasibility of the method is verified by experiments and simulations.