Interacting with the environment

Computer systems have  to communicate with  the world around them,  getting information about  the external world, and  taking  actions  to change the world. The world is dynamic, so that as the machine is calculating, the world is modifying, needing future computation to take to the new part of the world.

There  are  a variety of different paths  to manage computations that  communicate with  a  outer world. Usually speaking, such a calculation requires  to:

1.  get details from sensors,

2.  perform calculation, remembering some of the results, and

3.  take appropriate actions  to modify  the outer world.

1.3.1.1 Sequential

The most  immediately style  for developing a program that  communicates with  the world is the basic imperative style, in which  the program provides a sequence of 'instructions' to the computer it is controlling. A library of special procedures is described, some of which read details from the sensors  and  others  of which  cause reactions  to be operated.

In this model,  we should  usually write  a program that  goes  an idealized robot  in a square, if there is area in front of it.

if noObstacleInFront:

 moveDistance(1)

turnAngle(90)

moveDistance(1)

turnAngle(90)

moveDistance(1)

 turnAngle(90)

 moveDistance(1)

turnAngle(90)

The related  problem with  this model of programming is that  the programme has to check the sensors  sufficiently frequently. If the robot  examine for free space in front, and  then  starts  rolling, it can  turn  out that  a subsequent sensor  reading will give  that  there is something in front  of the robot,  either  because  someone gone in front  of it or because  the last reading was erroneous. It is rigid  to have the discipline, as a programmer, to identify to check  the  sensor  situations frequently enough, and  the  answering programs can be quite difficult to understand and read.

For the robot rolling toward the light, we can  write a program like this:

while lightValue < desiredValue:

moveDistance(0.1)

This would have the robot creep up, one by one, toward the light. We may  want  to change it so that the robot goes to a distance that was correlated to the difference between the desired light values and current.  However, if it takes bigger  steps, then mean while the time that it is rolling it will not be sensitive to possible  modification in the light number  and may not  react immediately to them.

1.3.1.2 Event-Driven

User-interface function are usually best managed differently, as event-driven programs. In that case, the program is defined as a collection  of functions that  are related to particular function that  can give  place.  So, for example, there  may  kinds  into  a text area,  or when  the  temperature of a reactor  goes  too high. An "event loop"  executes  continuously, examine  to see whether any  of the triggering function  have occurred, and, if they have, calling the associated function.

1.3.1.3 Transducer

An alternative view  is that  programming a machine that  communicates with  an external world is like creating a transducer with  internal state.   Think  of a transducer as a operating box that  executes continuously. At normal intervals,  the transducer takes  all of the  sensors,  does  a short  amount of calculation, saves  some  numbers  it will require  for the  next calculation, and then creates output values  for the actions.

This calculations occur over  and over   again.   Complex behaviour may rise  out  of the temporal functions of outputs and inputs.

def step():

distToFront = min(frontSonarReadings)

motorOutput(gain * (distToFront - desiredDistance), 0.0)

Executed normally, this  program will automatically calculate the robot's  velocity to be proportional  to the free memory in front of it.

The basic  problem with  the  transducer method is that  it may be very difficult to do  programs  that  are fundamentally sequential.