The purpose of this exercise is to play with emergence, agent-environment interaction, control architectures, and system complexity.
Return a URL to your runnable code.
by midnight (23:59) Sep. 13 (the evening before Friday's class next week).
Optionally you may return a short explanatory description of your system and the result.
Your task is to modify the Braitenberg Vehicles Scratch simulation (Links to an external site.)Links to an external site.. You may choose to do ONE of these modifications:
- a. Group Behaviors. Vehicles search for lights ("food"). Put lights on Vehicles so that Vehicles start to form groups. Make it so that they self-organize into groups that form long snakes. Can you make them create circles instead of snakes - or squares?
- b. Dynamic Environment. Food grows and rots; disappears when eaten. Vehicles search for food, and when they find it they eat it, at which point their control architecture changes in some way and makes them do other things, until they get hungry again, at which point their old control structure returns and they start to search for food again.
- c. Memory & Learning. Vehicles search for food. There are three kinds of edibles, one being poison, one being normal food, and the third being super-food which gives them triple energy (with side effects for their behavior). Their architecture may be simple but it is augmented with memory that enables them to learn to avoid poison and prefer super-food. NOTE: If you choose this option you may request to return the assignment on Sep. 21st.
If you think there are some parameters that would be fun to experiment with you may choose to make them tunable at the beginning of the simulation.
Make your demo as interesting as you can make it!