Page 146 - Biomimetics : Biologically Inspired Technologies

Bar-Cohen : Biomimetics: Biologically Inspired Technologies DK3163_c004 Final Proof page 132 21.9.2005 9:37am

132 Biomimetics: Biologically Inspired Technologies

4.3 MACHINE BODIES AND BRAINS

Many systems, including robotic systems in particular, are often viewed as comprising two major
parts: the morphology and the controller. The morphology is the physical structure of the system,
and the controller is a separate unit that governs the behavior of the morphology by setting the states
of actuators and reading sensory data. In nature, we often refer to these as the body and brain,
respectively. In control theory, we refer to these as the plant and the control (the term plant, as
in ‘‘manufacturing plant,’’ is used because of the original industrial applications). In computer
engineering terms, this often translates into hardware and software. This distinction is semantic; we
simply tend to refer to the part which is more easily adaptable as control and the part that is fixed as
the morphology. In practice, both the morphology and control contribute to the overall behavior of
the system and the distinction between them is blurred. Very often a particular morphology
accounts for some of the control and the control is embedded in the morphology. Nevertheless,
in describing the application of evolutionary design to systems, we find this distinction pedagogic-
ally useful.
In the following sections, we will see a series of examples of the application of evolutionary
processes to open-ended synthesis. These examples were chosen to illustrate the design of robotic
systems for their intuitiveness, starting at control and moving on to both control and morphology.
Following these examples, we will take a look at the common principles, and future challenges.

4.3.1 Evolving Controllers

It is perhaps easier, both conceptually and technically, to explore application of evolutionary
techniques to the design of robot controllers before using it to evolve their morphologies too.
Robot controllers can be represented in any one of a number of ways: as logic functions (‘‘if–then–
else’’ rules), as finite state machines, as programs, as sets of differential equations, or as neural
networks to name a few. Many of the experiments that follow represent the controller as a neural
network that maps sensory input to actuator outputs. These networks can have many architectures,
such as feed-forward or recurrent. Sometimes the choice of architecture is left to the synthesis
algorithm.
Some of the early experiments in this area performed by Beer and Gallagher (1992). Nolfi and
Floreano (2004), Harvey et al. (1997), and Meyer (1998) review many interesting experiments
evolving controllers for wheeled and gantry robots, but let us look at some examples with legged
robots. Consider a case where we have a legged robot morphology fitted with actuators and sensors,
and we would like to use evolutionary methods to evolve a controller that would make this machine
move (locomote) towards an area of high chemical concentration. Bongard (2002) explored this
concept on a legged robot in a physically realistic simulator. The robot has four legs and eight rotary
actuators as shown in Figure 4.1a. It has four touch sensors at the feet, which output a binary signal
depending on weather or not they are touching the ground. The machine also has four angle sensors
at the knees, outputting a graded signal depending on the actual angle of the knee. There are two
chemical sensors at the top, which output a value corresponding to the chemical level they sense
locally.
The behavior of the machine is determined by a neural controller that maps sensors to actuators,
as shown in Figure 4.1b. Inputs of candidate neural controllers were connected to the sensors, and
their output connected directly to the eight motors. Machines were rewarded for their ability to
reach the area with high concentration. The fitness was evaluated by trying out a candidate
controller in four different concentration fields, and summing up the distance between the final
position of the robot and the highest concentration point. The shorter the distance the better — and
in this sense the total distance is a performance error. In this experiment, 200 candidate controllers
were evolved for 50 generations. The variation operators could decide if and how to connect the
neurons. Figure 4.1c shows the progress of this error over generational time. The performance of