Research On Environment Perception Of Mobile Robot Based On SLAM

With the continuous progress of society and the vigorous development of science and technology,the pursuit of mankind has become more intelligent in various industries.Nowadays,intelligent robots have been applied in all walks of life,making robots more intelligent has become a hot research direction.With the further entry of intelligent robots into people’s life and work,people also put forward higher requirements for the working ability of intelligent mobile robots.In order for mobile robots to perform different tasks intelligently and autonomously in different work scenarios,they must be able to sense their environment and have the ability to localize themselves.In the field of mobile robotics,we usually use SLAM(Simultaneous Localization and Mapping)to implement such basic functions for mobile robots.Most of the current visual SLAM research,which implements simultaneous localization and mapping of robots,mainly uses detection and matching of feature points in images to accomplish localization and mapping.However,it often only uses the geometric information in the environment,but cannot obtain the semantic information in the environment,resulting in the robot not being able to perceive the surrounding environment well.Location and mapping can only be done by detecting and matching the feature points in the image.Only the geometric information in the environment is used,and the linguistic information in the environment cannot be obtained,which makes the robot unable to perceive the environment well.In this paper,a three-dimensional semantic map is implemented based on the system design.The data uses RGBD image sequence,and the target detection algorithm YOLOv4 is applied to ORB-SLAM2 system to extract the semantic information in the environment.A three-dimensional dense point cloud map constructed by ORB-SALM2 system is segmented using an improved LCCP algorithm.Semantic information in the environment is projected to the three-dimensional point cloud according to the globally consistent camera posture,and a three-dimensional dense point cloud map with semantic information is constructed.The main work and research methods of the environmental awareness system designed in this paper are as follows:1.Based on RGB-D images,a visual SLAM based on ORB-SLAM2 system is designed to obtain globally consistent motion and posture.The ORB features of two adjacent frames are extracted and calculated according to the geometric relationship of the ORB features,and the BA(Bundle Adjustment)algorithm is used to adjust the position and position to optimize it.The word bag model is used to detect whether the mobile robot has loops or not,to correct the motion track drift of the mobile robot,and to eliminate the accumulated error in the process of the robot motion.2.Target detection algorithm is incorporated into ORB-SLAM2 system to enhance the mobile robot’s awareness of the environment by detecting and identifying RGB images to obtain semantic information in the environment.Target detection algorithm labels objects with moving attributes in the environment and only relies on the key frames of static objects for map construction,effectively removing the impact of dynamic objects on semantic map construction.3.The point cloud is cut based on an improved LCCP algorithm,and the geometric and semantic information is projected into three-dimensional space according to the globally consistent camera position and posture.Semantic maps are constructed and stored in the form of color octree,which improves the perception of environment by mobile robots.