Abstract
When locating wall-climbing robots with vision-based methods, locating and controlling the wall-climbing robot in the pixel coordinate of the wall map is an effective alternative that eliminates the need to calibrate the internal and external parameters of the camera. The estimation accuracy of the homography matrix between the camera image and the wall map directly impacts the pixel positioning accuracy of the wall-climbing robot in the wall map. In this study, we focused on the homography estimation between the camera image and wall map. We proposed HomographyFpnNet and obtained a smaller homography estimation error for a center-aligned image pair compared with the state of the art. The proposed hierarchical HomographyFpnNet for a non-center-aligned image pair significantly outperforms the method based on artificially designed features + Random Sample Consensus. The experiments conducted with a trained three-stage hierarchical HomographyFpnNet model on wall images of climbing robots also achieved small mean corner pixel error and proved its potential for estimating the homography between the wall map and camera images. The three-stage hierarchical HomographyFpnNet model has an average processing time of 10.8 ms on a GPU. The real-time processing speed satisfies the requirements of wall-climbing robots.
Highlights
In the past several decades, considerable research has been dedicated to the development of mobile systems that can traverse vertical surfaces
This paper focuses on the homography estimation between the camera image and the wall map
Our contributions in this work are as follows: (1) we proposed HomographyFpnNet to improve the homography estimation accuracy of a center-aligned image pair; (2) based on the proposed
Summary
In the past several decades, considerable research has been dedicated to the development of mobile systems that can traverse vertical surfaces. To locate and control wall-climbing robots using a wall map, the pixel coordinate of the wall-climbing robot, as detected in image_a (Figure 1, left), can be translated as a pixel coordinate in wall_map (Figure 1, right) using a homography matrix (or projective transformation matrix) H, which eliminates the need to calibrate the internal and external parameters of the camera. The estimation accuracy of H between the camera image and the wall map directly impacts the pixel positioning accuracy of the wall-climbing robot in the wall map. HomographyFpnNet, we used a hierarchical method composed of one HomographyFpnNet_A model and two HomographyFpnNet_B models to estimate the homography of a non-center-aligned image pair, and the mean corner error was significantly smaller than that of the classical. ORB/SIFT+RANSAC methods; and (3) we conducted experiments with our trained three-stage hierarchical HomographyFpnNet model on climbing robot wall images and achieved promising results.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
