Research on Target Detection Method for Intelligent Mobile Robots based on Machine Visionr

Abstract

Intelligent mobile robots use machine vision and deep learning to achieve environmental perception, path planning, and obstacle avoidance, effectively improving autonomous decision-making capabilities. In this paper, you only look once (YOLO) and Faster Region-based Convolutional Neural Network (Faster R-CNN), Region of Interest Detection (ROB), and Direct Sparse Odometry (DSO) are compared and analyzed. The advantages and disadvantages of two Simultaneous Localization and Mapping (SLAM) algorithms, single-sensor and multi-sensor fusion are compared so that different object detection methods can be used in different application scenarios. In this paper, it is concluded that algorithms such as YOLO and Faster R-CNN meet the requirements of real-time and accuracy, respectively, visual SLAM supports autonomous navigation, and multi-sensor fusion enhances adaptability and stability. Current challenges include balancing real-time performance and accuracy, improving multi-target recognition capabilities in complex scenarios, and maintaining stability in harsh environments. Future research will combine deep learning, edge computing, and 5G technologies to further promote the intelligent and efficient application of machine vision in the field of robotics.