Velocity Control of an Omnidirectional Wheeled Mobile Robot Using Computed Voltage Control with Visual Feedback: Experim
- PDF / 2,658,619 Bytes
- 14 Pages / 594.77 x 793.026 pts Page_size
- 34 Downloads / 365 Views
ISSN:1598-6446 eISSN:2005-4092 http://www.springer.com/12555
Velocity Control of an Omnidirectional Wheeled Mobile Robot Using Computed Voltage Control with Visual Feedback: Experimental Results Armando Saenz*, Victor Santibañez, Eusebio Bugarin, Alejandro Dzul, Héctor Ríos, and Jorge Villalobos-Chin Abstract: In this paper, we propose a novel computed voltage based control law for an omnidirectional wheeled mobile robot equipped with four mecanum wheels. First, the dynamical model for the mechanical part is originally introduced by means of the Euler-Lagrange formulation. The motion constraints are added to the dynamic model using the Lagrange multipliers. Then, the dynamic model is completed by incorporating the dynamics of the actuators. Therefore, the dynamical model input signals are the armature voltage of the motors. The control law requires only the feedback of position and velocity of the whole robot, unlike most controllers in the literature that also need wheel speed feedback. The position and velocity are obtained by means of a multicam vision system, so measurements from motor encoders are not needed. A high-order sliding modes differentiator is included to estimate linear and angular velocities in a finite time. A stability proof is presented by means of the direct Lyapunov method, and furthermore, an analysis about parametric uncertainties in the mechanical parameters is introduced. The experimental results validate the theoretical proposal and show the good performance of the approach. Keywords: Actuator dynamics, computed voltage control, omnidirectional wheeled mobile robots, sliding-mode observer.
1.
INTRODUCTION
In the 50’s, Arthur Barret invented the first wheeled mobile robot (WMR). The WMR was called Guide-o-Matic [1]. This WMR was a truck used to transport tools. The truck was guided by a wire and had continuous stops at work stations. In general, the applications of WMRs are very widespread nowadays, e.g., entertainment, security, war, rescue missions, spacial missions, assistant healthcare, etc. A state of the art could be revised in the references [1–3]. A WMR could be thought as a robotic system that can be separated into three smaller subsystems [4]. The subsystems are depicted in Fig. 1 and described
Fig. 1. Subsystems of a WMR.
as follow: 1) Power converter, which supplies the electric power to activate the actuators. 2) Actuator system, this subsystem carries out the motion of the kinematic configuration. Such a subsystem is equipped, in most of the cases, with Direct Current (DC) motors. 3) Kinematic configuration, which represents all relatives configurations of the wheels that the robot can possess. The most important configurations are: unicycle, tricycle, Ackerman, differential, omnidirectional and skid steer. Before using a WMR, the following two questions must be answered: • Where is the robot? • Where is the robot going? The first question is answered by defining how to acquire the WMR posture (i.e., position and orientation). Posture acquisition can be done using diffe
Data Loading...
