Back to Search Start Over

Cooperative step-climbing strategy using an autonomous wheelchair and a robot

Authors :
Daisuke Maki
Takafumi Toyama
Keisuke Sato
Eiji Nakano
Hidetoshi Ikeda
Source :
Robotics and Autonomous Systems. 135:103670
Publication Year :
2021
Publisher :
Elsevier BV, 2021.

Abstract

This report describes an automatic control system that allows an assistive robot pushing a wheelchair to climb steps. The robot is equipped with a wheeled mechanism and dual manipulators. The wheelchair is a commercially available model that has been equipped with sensors, circuits, and batteries. The robot and wheelchair are connected when the vehicles climb a step. In that operation, the front wheels of the wheelchair are lifted and placed on the step using the velocity differences between the wheelchair and the robot. Next, when the rear wheels of the wheelchair ascend the step, the robot imitates the upper arm motions of a human pushing against his/her chest that commonly occurs when maneuvering a wheelchair up a step. Similarly, the front wheels of the robot are lifted and placed on the step using the velocity differences between the vehicles and the robot’s front wheels. After that, with the assistance of the wheelchair, the other wheels of the robot climb onto the step. In an effort to ensure safety, we also performed a theoretical analysis to determine the most suitable distance for lifting the front wheels of the robot when approaching and climbing a step. Our newly developed cooperative step-climbing system makes it possible to eliminate the complicated operations that were required by previous methods and can also prevent collisions between the wheelchair’s front wheels and the step, thus drastically improving the convenience of the operation. The test subject riding the wheelchair was an able-bodied male, and the experiment conducted to evaluate our system was performed on a 120 mm step height that had a friction coefficient of 0.72. This setup was sufficient for demonstrating the overall effectiveness of our system.

Details

ISSN :
09218890
Volume :
135
Database :
OpenAIRE
Journal :
Robotics and Autonomous Systems
Accession number :
edsair.doi...........0a01b3ad15d4ab0b3745a1ffecee0dcc
Full Text :
https://doi.org/10.1016/j.robot.2020.103670