The vacuum environments in the preceding
exercises have all been deterministic. Discuss possible agent programs
for each of the following stochastic versions:
1. Murphy’s law: twenty-five percent of the time, the Suck action fails to clean the floor if it is dirty and deposits dirt onto the floor if the floor is clean. How is your agent program affected if the dirt sensor gives the wrong answer 10% of the time?
2. Small children: At each time step, each clean square has a 10% chance of becoming dirty. Can you come up with a rational agent design for this case?
1. Murphy’s law: twenty-five percent of the time, the Suck action fails to clean the floor if it is dirty and deposits dirt onto the floor if the floor is clean. How is your agent program affected if the dirt sensor gives the wrong answer 10% of the time?
2. Small children: At each time step, each clean square has a 10% chance of becoming dirty. Can you come up with a rational agent design for this case?
The vacuum environments in the preceding
exercises have all been deterministic. Discuss possible agent programs
for each of the following stochastic versions:
1. Murphy’s law: twenty-five percent of the time, the Suck action
fails to clean the floor if it is dirty and deposits dirt onto the
floor if the floor is clean. How is your agent program affected if
the dirt sensor gives the wrong answer 10% of the time?
2. Small children: At each time step, each clean square has a 10%
chance of becoming dirty. Can you come up with a rational agent
design for this case?