Journal of Robotics and Mechatronics
Online ISSN : 1883-8049
Print ISSN : 0915-3942
ISSN-L : 0915-3942
Regular Papers
Multi-Legged Inspection Robot with Twist-Based Crouching and Fine Adjustment Mechanism
Maroay PhlernjaiPhotchara Ratsamee
Author information
JOURNAL OPEN ACCESS

2022 Volume 34 Issue 3 Pages 588-598

Details
Abstract

The monitoring and inspection of buildings, towers, and large chemical tanks can be dangerous and costly. A wall climbing multi-legged robot can climb and crouch to move the inspection probe closer to the wall surface. The typical wall-sticking crouching motion of multi-legged robot is unable to achieve this without an additional actuator at the foot base. This limits the design of a magnetic leg that is bulky enough to keep the robot stable during inspections. This paper provides the concept and implementation details of a coarse-to-fine inspection using a novel twisting motion to crouch the robot towards the steel surface without the need for an additional actuator at the magnetized foot joint. The center position error of this method is reported as less than 1 cm in radius. Additionally, this paper suggests using a linear motor to provide fine-tuning capability for the gentle touch of the ultrasonic inspection probe to the steel surface, protecting the probe from a direct crash during the coarse twisting crouch movement. Experiments have been conducted to confirm the validity of this concept on horizontally flat steel plates. The robot successfully crouch towards the surface and gently touch it with the ultrasonic probe. The results were reliable and comparable to manual inspection. For the 2 mm and 3 mm sample plates, the average measurement error compared to the ground truth from a vernier caliper was less than 2%, which is better than the measurement error (3%) of humans. The total time used was 2.475 seconds per inspection spot, which is 5.6 times faster compared to the expected time used to cover the same distance with a typical probe extension using a lead screw drive.

Content from these authors

This article cannot obtain the latest cited-by information.

© 2022 Fuji Technology Press Ltd.

This article is licensed under a Creative Commons [Attribution-NoDerivatives 4.0 International] license (https://creativecommons.org/licenses/by-nd/4.0/).
The journal is fully Open Access under Creative Commons licenses and all articles are free to access at JRM official website.
https://www.fujipress.jp/jrobomech/rb-about/#https://creativecommons.org/licenses/by-nd
Previous article Next article
feedback
Top