We present a method to describe the behaviour of a mobile robot quantitatively, using methods from dynamical systems theory, time series analysis and deterministic chaos theory. Experimental results obtained with a Pioneer II mobile robot demonstrate the use of the method, and show that robot behaviour exhibits deterministic chaos, and is substantially influenced by the control program executed by the robot, while changes to the environment have far less influence. 1 Background 1.1 Motivation Research in mobile robotics to date has, with very few exceptions, been based on trial-and-error experimentation and the presentation of existence proofs. Task-achieving robot control programs are obtained through a process of iterative refinement, typically involving the use of computer models of the robot, the robot itself, and program refinements based on observations made using the model and the robot. This process is iterated until the robot’s behaviour resembles the desired behaviour to a sufficient degree of accuracy. Typically, the results of these iterative refinement processes are valid within a very narrow band of application scenarios, they constitute “existence proofs”. As such, they demonstrate that a particular behaviour can be achieved, but not, how that particular behaviour can in general be achieved for any experimental scenario. The purpose of this paper is to introduce quantitative measures of robot behaviour, as components of a theory of robot-environment interaction. Using dynamical systems theory and methods of analysis derived from chaos theory, we investigate quantitatively in what way the behaviour of a mobile robot changes if a) the robot’s environment is modified, and b) the robot’s control code is modified. Underlying this research, however, is the fundamental question: How can the interaction of a mobile robot with its environment be characterised quantitatively?
[1]
D. A. Bell,et al.
Information Theory and Reliable Communication
,
1969
.
[2]
M. Pascual.
Understanding nonlinear dynamics
,
1996
.
[3]
F. Takens.
Detecting strange attractors in turbulence
,
1981
.
[4]
M. Yamaguti,et al.
Chaos and Fractals
,
1987
.
[5]
L. Tsimring,et al.
The analysis of observed chaotic data in physical systems
,
1993
.
[6]
A. Wolf,et al.
Determining Lyapunov exponents from a time series
,
1985
.
[7]
J. A. Stewart,et al.
Nonlinear Time Series Analysis
,
2015
.
[8]
Fraser,et al.
Independent coordinates for strange attractors from mutual information.
,
1986,
Physical review. A, General physics.
[9]
Mehmet Emre Çek,et al.
Analysis of observed chaotic data
,
2004
.
[10]
Gregor Schöner,et al.
Dynamics of behavior: Theory and applications for autonomous robot architectures
,
1995,
Robotics Auton. Syst..
[11]
Tim Smithers,et al.
On quantitative performance measures of robot behaviour
,
1995,
Robotics Auton. Syst..