Loading ...
Sorry, an error occurred while loading the content.

RE: [aima-talk] Simple question regarding task environment

Expand Messages
  • E etech058
    Hello bob, Do you mean the 3 concept s relationship? 1. environment 2. agent head 3. agent body My idea are as follows: 1. Every agent contains 2 parts, 1 is
    Message 1 of 4 , Apr 20, 2004
    • 0 Attachment
      Hello bob,
      Do you mean the 3 concept's relationship?
      1. environment
      2. agent head
      3. agent body

      My idea are as follows:
      1. Every agent contains 2 parts, 1 is "agent head", the other is "agent body".
      2. The agent through its body's sensors to get percept from environment. The percept has been sent to "agent head" for process, and "head" produce out the "action". The "action" has been sent to "agent body" for execution. "execuation" will change some aspect of environment.

      3. In the sourcecode of "aima-lisp-code\agents\agents\agents.lisp" file, it contains Mr. Author's design Decision mark.
      ====quotation start==============
      ;; We have decided that the agent and its body are two separate objects.
      ;; We could have combined the agent and its body into one object. But then
      ;; each new type of agent would need to inherit from both AGENT and some
      ;; body type, such as OBJECT. This would require multiple inheritance,
      ;; which is part of CLOS, but is not in our simple implementation for those
      ;; who don't have CLOS. In any case, it would get messy. We think that
      ;; separating agents from agent-bodies is a good thing for this
      ;; implementation. (Just don't take it too far and assume that this says
      ;; anything about the mind-body problem.)
      ====quotation end================


      I hope the above information is useful to you.
      Kind regards
      chenyu
    • Mauricio Amaral de Almeida
      Someon please correct me if I wrong, but in my interpretation the actuators as in conceptualy part of the agent. But when you make a software implementation on
      Message 2 of 4 , Apr 20, 2004
      • 0 Attachment
        Someon please correct me if I wrong, but in my interpretation the actuators as
        in conceptualy part of the agent.
        But when you make a software implementation on the pair agent-environment you
        have to implement the actuatror, and the sensor for that matter, as methods
        of the class environment.
        That is so for you to be able to decouple the agent and the environment. In
        that way when an agent want to act over an environment it should call the
        proper method in the class environment.
        This is in fact the best way to do this because doing things this way the
        agent doesn't need to know the environment internal structure.

        Mauricio


        On Monday 19 April 2004 23:25, Bob Riley wrote:
        > I am starting on some of the computing exercises in chapter 2 and
        > have a question regarding task environment. According to the book,
        > task environment is specified by performance measure, environment,
        > actuators, and sensors. The 1st two of these I can understand - but
        > why are acuators and sensors included in the task environment?
        > Shouldn't the task environment be the elements of interest outisde
        > of the agent? Aren't sensors and actuators part of the agent
        > itself? Upon first seeing this, I thought that my interpretation of
        > agent may be wrong - maybe the agent is simply the program/function
        > and the actuators and sensors are external entities that the agent
        > interacts with. But there are several examples in this chapter
        > where the actuators and sensors are considered to be part of the
        > agent. Can someone clear this up? Is it just me or does the book
        > seem to be altering the definition of agent in different parts of
        > the chapter?
        >
        > Thanks,
        >
        > Bob
        >

        --
        Prof. Dr. Maurício Amaral de Almeida
        malmeida@...
      • ³ÂÓí
        I think you are right. Because the environment also our program, rather than a real-world s thing. kind regards chenyu ... actuators as ... environment you ...
        Message 3 of 4 , Apr 23, 2004
        • 0 Attachment
          I think you are right. Because the environment also our program,
          rather than a real-world's thing.


          kind regards
          chenyu



          --- In aima-talk@yahoogroups.com, Mauricio Amaral de Almeida
          <malmeida@i...> wrote:
          > Someon please correct me if I wrong, but in my interpretation the
          actuators as
          > in conceptualy part of the agent.
          > But when you make a software implementation on the pair agent-
          environment you
          > have to implement the actuatror, and the sensor for that matter, as
          methods
          > of the class environment.
          > That is so for you to be able to decouple the agent and the
          environment. In
          > that way when an agent want to act over an environment it should
          call the
          > proper method in the class environment.
          > This is in fact the best way to do this because doing things this
          way the
          > agent doesn't need to know the environment internal structure.
          >
          > Mauricio
          >
          >
          > On Monday 19 April 2004 23:25, Bob Riley wrote:
          > > I am starting on some of the computing exercises in chapter 2 and
          > > have a question regarding task environment. According to the
          book,
          > > task environment is specified by performance measure, environment,
          > > actuators, and sensors. The 1st two of these I can understand -
          but
          > > why are acuators and sensors included in the task environment?
          > > Shouldn't the task environment be the elements of interest outisde
          > > of the agent? Aren't sensors and actuators part of the agent
          > > itself? Upon first seeing this, I thought that my interpretation
          of
          > > agent may be wrong - maybe the agent is simply the
          program/function
          > > and the actuators and sensors are external entities that the agent
          > > interacts with. But there are several examples in this chapter
          > > where the actuators and sensors are considered to be part of the
          > > agent. Can someone clear this up? Is it just me or does the book
          > > seem to be altering the definition of agent in different parts of
          > > the chapter?
          > >
          > > Thanks,
          > >
          > > Bob
          > >
          >
          > --
          > Prof. Dr. Maurício Amaral de Almeida
          > malmeida@i...
        Your message has been successfully submitted and would be delivered to recipients shortly.