Abstract

This article proposes a novel approach aimed at estimating the pose of a camera, affixed to a robotic manipulator, against a target object. Our approach provides a way to exploit the redundancy of the robotic arm kinematics by directly considering manipulator poses in the model formulation for camera pose estimation. We adopt a single camera multi-shot technique that minimizes the reprojection error over all the rigid poses. The results of the proposed method are compared to four other studies employing either monocular or stereo setup. The experimental results on synthetic and real data show that the proposed monocular approach achieves better and in some cases comparable results to the stereo approach. Moreover, the proposed approach is significantly more robust and precise compared to other methods.

Highlights

  • Camera pose estimation with respect to a target object/scene has been widely researched in the fields of computer and machine vision, photogrammetry and robotics

  • We focus on the prerequisites of visual servoing of a robotic arm for accurate manipulation

  • The authors claim that the presented approach achieves state-of-the-art accuracy in terms of pose estimation and its sub-sequent odometry

Read more

Summary

INTRODUCTION

Camera pose estimation with respect to a target object/scene has been widely researched in the fields of computer and machine vision, photogrammetry and robotics. Ali et al.: Multi-View Camera Pose Estimation for Robotic Arm Manipulation for estimating the pose of the camera relative to the object This approach is commonly known as Perspective from n points (PnP) [13]. Collet and Srinivasa [19] introduced a modified version of full multi-view, which they termed as introspective multiview approach This multistep approach first estimates object and camera pose using a single-view method. Some studies utilize multi-camera approaches to solve the pose estimation problem. This study is driven by the motivation to develop an accurate and robust pose estimation method for the International Thermonuclear Experimental Reactor (ITER) project using an eye-in-hand monocular approach. We attempt to achieve comparable results to stereo approaches by proposing a multi-view monocular approach considering the case of robotic arm manipulation.

PROBLEM FORMULATION
METHODS
EXPERIMENTAL SETUP AND RESULTS
ERROR METRICS
TESTS ON SYNTHETIC DATA
Findings
CONCLUSION
Full Text
Published version (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