What does "cybernetics" actually mean?

This question often follows when we are talking about the Max Planck Institute for Biological Cybernetics or the Cyberneum. This article is intended to give an impression of what cybernetics is all about.

According to its founder Norbert Wiener, cybernetics is the science of control, regulation and information processing of machines and living organisms. Derived from the Greek words for helmsman and control, the term found its present meaning in the German language in the middle of the 20th century. In order to better illustrate the aspect of control and regulation, the classic example of a cybernetic system, the thermostat, can be used. The principle of the thermostat is based on entering a desired temperature value and comparing this setpoint with the current temperature, the actual value. The system constantly checks whether the setpoint has now been reached. If yes, the power of the heating is minimized, if not, it continues to run. The controller in the thermostat therefore has the task of regulating the heating in such a way that the actual value adjusts to the setpoint. This means quite clearly: If it is too cold in the room, the heating works. Once the desired temperature has been reached, the heating stops.

Similar to the thermostat in the home, our body also has a cybernetic system that permanently adjusts our body temperature to around 37 degrees Celsius - regardless of the ambient temperature. Not an easy task, if you look at the temperature fluctuations between inside and outside and throughout the year. However, there are control loops in humans not only with regard to body temperature, but also for every kind of sensory processing. Whether we stand, walk, talk or listen - it is always necessary to regulate incoming signals and the resulting actions. And in this way we are already building on the research priorities of the institute: Perception and action constantly influence each other, and it is precisely this dependency that is investigated here at the Institute using a cybernetic control loop. On the one hand, we are interested in the questions of how humans are able to absorb, process and store sensory information; on the other hand, we are interested in what behaviour results from it and what new sensory information results from it.

Another classic example of cybernetic systems are the Braitenberg vehicles, named after one of our former directors, Valentino Braitenberg. He designed simple artificial intelligences (the "vehicles") consisting of sensors and motors as well as simple control loops for the coupling of sensor (perception) and motor (action). The simplest model of such a vehicle has a light sensor and a motor. The control loop says that the more light the vehicle receives, the faster it moves. Less light therefore leads to a slowdown and darkness even to a standstill of the vehicle. More complex models, for example, have two independent light detectors on the left and right sides, each connected to a wheel on the respective side. More light on the right side results in the right wheel moving faster, so the vehicle moves to the left and away from the light source (it "flees"). If the wiring is changed so that the opposite wheel rotates faster, the vehicle turns to the light source (it "likes light"). With the help of just a few modules, Braitenberg designed many such vehicles, which showed purposeful, adaptable and complex behaviour, i.e. can be regarded as intelligent organisms in their own way. In fact, however, their complex behaviour can be explained by very simple (cybernetic) control mechanisms. After their publication in the eighties, the Braitenberg vehicles offered

Inspiration for generations of robotics and brain researchers. Similar to the Braitenberg vehicles, complex control cycles also take place in the human brain. The systematic investigation of the interdependent perception and action cycles with the help of cybernetic methods allows us to draw conclusions about how the outside world is represented in the brain and how complex behaviour is made possible. In machines, for example, such cybernetic models help to make automated driving possible. Cybernetics does indeed sound complicated, but it simply describes a cycle to be regulated - in our case the research and modelling of perception and action processes in humans. With the knowledge gained, new user-friendly human-machine interfaces can be developed that make life easier for us in an ever more technological world.

Go to Editor View