Model predictive tracking control for wheeled mobile robots considering limitation and continuity of steering angles

Robots with independent four-wheel steering have many advantages such as high mobility, energy efficiency and strong stability. They can move toward arbitrary direction by conforming the wheel velocity to the velocity field of the locomotion. If the steering angles are limited, or if the instantaneo...

Full description

Bibliographic Details
Main Authors: Yuki HAGIMORI, Naoki TAKAHASHI, Kenichiro NONAKA
Format: Article
Language:Japanese
Published: The Japan Society of Mechanical Engineers 2015-03-01
Series:Nihon Kikai Gakkai ronbunshu
Subjects:
Online Access:https://www.jstage.jst.go.jp/article/transjsme/81/825/81_14-00484/_pdf/-char/en
Description
Summary:Robots with independent four-wheel steering have many advantages such as high mobility, energy efficiency and strong stability. They can move toward arbitrary direction by conforming the wheel velocity to the velocity field of the locomotion. If the steering angles are limited, or if the instantaneous center of motion passes through the steering axis, however, abrupt turn of wheels may appear. It requires unacceptable load torque and the steering mechanism may be damaged. To avoid rapid steering, in this study, we propose a tracking control method for robots having independent four-wheel steering with limited range of steering angles. We use a model predictive control based on the model with the steering and wheel velocity dynamics which explicitly deals with the limitation of the steering angle. We conduct experiments on two trajectories; S-shaped curve and V-shaped turn. In the S-shaped curve, the steering angle has to turn over due to their limitation, and in the V-shaped turn, the instantaneous center of motion passes through the steering axis; both of them may cause discontinuous steering in the conventional methods. Comparing with the conventional method, the explicit advantage of the proposed method is that the continuity of the steering angle and limited wheel velocity is guaranteed. We implement the proposed method into an embedded CPU on an experimental robot to verify the efficacy and feasibility of the proposed method.
ISSN:2187-9761