Design and visual servo control of a non-holonomic mobile manipulator for visual inspection

Date

2023-5-03

Author

Şen, Mert

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

280
views

327
downloads

Cite This

In this study, we design a system to perform a visual servo control using a 6 DoF robotic manipulator with an RGB-D camera, mounted on a non-holonomic mobile robot capable of performing localization and mapping in an indoor environment. The aim of this study is to utilize user-provided RGB/grayscale image of the target object in order to move the end-effector to the same relative pose with respect to the target object at the time the target image was captured. The user is requested to provide the system with a rough estimate of a target object position in world frame and three angles which describe a rough estimate of the gaze direction of the camera when the target image was taken. Our algorithm not only computes the optimal pose at which the camera can gaze at the target object with a direction as close as possible to that provided by the user, subject to kinematic constraints, while the system approaches the target object, but also ensures that the camera does not collide with any wall-like obstacles present in the environment. When the target object is visually recognized by the visual perception algorithms, the system computes the position of the target object and gaze direction and corrects the rough estimate of the object position that user provided. If the computed final position of the camera is in the reach of the arm, system activates IBVS to move the camera to same relative pose when the target image was taken. Our proposed algorithms are implemented in integration with ROS, and tested in both real and simulation environments.

Subject Keywords

Visual servoing of non-holonomic mobile manipulator, Image based visual servoing (IBVS), Position based visual servoing (PBVS), Feature tracking, Feature extraction

URI

https://hdl.handle.net/11511/103720

Collections

Graduate School of Natural and Applied Sciences, Thesis