Multi-Scale Fully Convolutional Network-Based Semantic Segmentation for Mobile Robot Navigation

In computer vision and mobile robotics, autonomous navigation is crucial. It enables the robot to navigate its environment, which consists primarily of obstacles moving objects. Robot employing impediment detections, such as walls pillars, not only essential but also challenging due real-world complications. This study provides a real-time solution problem obtaining hallway scenes from an exclusive image. The authors predict dense scene using multi-scale fully convolutional network (FCN). output image with pixel-by-pixel predictions that can be used for various strategies. addition, method comparing computational cost precision FCN architectures VGG-16 introduced. binary semantic segmentation optimal obstacle avoidance robots are two areas in our outperforms methods competing works. successfully apply perspective correction segmented order construct frontal view general area, identifies available area. strategy comprised collision-free path planning, reasonable processing time, smooth steering low angle changes.