Research on autonomous inspection control system of robot combined with improved YOLOv8 n and SLAM
A scheme of indoor security robot autonomous inspection control system based on improved YOLOv8n and SLAM is proposed.YOLOv8n algorithm of lightweight processing is adopted for target detection,and GhostNet is used to carry out lightweight improvement of YOLOv8n.Thus,the computation and memory consumption are reduced,and the speed of target detection is improved at the same time.The frame rate of the improved YOLOv8n is 50％ higher than that before the improvement,the model is reduced to 63％ of the original,and the recognition accuracy is only reduced by about 5％.In order to enhance the perception and autonomy of robots to unknown environment,Cartographer algorithm is used.Based on this algorithm,the robot can achieve autonomous navigation and mapping,and the lateral deviation between the estimated value and the actual value in the positioning process is less than 0.06 m;the longitudinal deviation is less than 0.08 m;course deflection angle is less than 16°.The experimental results show that the system can realize the accurate mapping and the rapid detection of target flame,and realize the real-time early warning.
万方数据
