jason-users Mailing List for jason

Hi All,

Well, made some progress. If I override the executeAction method  
inherited from Environment, e.g.

public class TankCodersEnvironment extends Environment {

@Override
     public boolean executeAction(String agName, Structure act) {
        // System.out.println("The action is not implemented in the  
default environment.");
         return true;
     }

Then I don't get these errors:
[Environment] The action turnToAngle(40.907071492138506) done by  
central is not implemented in the default environment.

Don't know why I suddenly started getting this error in 1.3.4, maybe  
the previous "[Atom] You should not add annot 'source(self)' in atom  
cruise" message was blocking it?

But, is this approach OK? Currently all of the actions are in an  
extension of AgArch, and that seems to work ok. But it seems that  
there is also an attempt to execute actions from the Environment class.

I guess maybe its nothing major, everything seems to work alot better  
now, but I don't want to be breaking intended implementation approach  
/ using some costly implementation.

Cheers!

Vin

Quoting Vin Baines <v.f...@ba...>:

> Hi Rafael,
>
> I thought I was.. but maybe not. mas2j is basically:
>
> 	infrastructure: Centralised
> 	environment:   
> br.furb.inf.tcc.tankcoders.jason.TankCodersEnvironment("-debug")
> 	agents:
> 	  central2 agentArchClass br.furb.inf.tcc.tankcoders.jason.TankAgArch;
> 	  central agentArchClass br.furb.inf.tcc.tankcoders.jason.TankAgArch;
>
> Have attached those java files, not sure if helps much but might give
> a clue to the implementation.
>
> In the simulated environment, the vehicles perform exactly as they did
> with jason 1.3.3 (I would add a link to the video but I'm having codec
> problems, scenario with 1.3.3 is here
> http://people.bath.ac.uk/vb216/follow.avi), the only differnce with
> 1.3.4 is this new error message. Is that some additional error
> checking? I can see the methods are being called at each cycle as the
> vehicles behave as expected.
>
> I'm trying to think through the code, I'm not sure what kind of return
> would be expected?
>
> +Jomi - sorry, something went wrong with the zip file extraction on my
> PC, it hadn't extracted the javacc lib. I re-ran the extraction, and
> it worked fine the second time and no need for the additional javacc
> download.
>
> Vin
>
> Quoting Rafael H Bordini <r.b...@ac...>:
>
>> Hi Vin,
>>
>> Did you specify the correct class with your simulated environment in the
>> mas2j file? If so, you probably forgot to return (with true, probably)
>> when your implementation of  "turnToAngle" in the environment finished
>> changing the simulated model. The environment gives out that message for
>> (external) actions that were not implemented yet.
>>
>> BTW, there are some strange parts in your code. For example, in a plan
>> context "true & started(Owner)" could be written just "started(Owner)"
>> (recall the identity laws of classical logic)
>>
>> Cheers,
>>
>> Rafael
>>
>> On 22/06/2011 08:27, Vin Baines wrote:
>>> Hi Jomi,
>>>
>>> Hmm.. OK, download 1.3.4 and compiled it (had to add in a javacc.jar
>>> which seems a new dependency?)
>>>
>>> Looks like there are a few changes too, I've changed over
>>> Literal info = new Literal("info");
>>> to
>>> LiteralImpl info = new LiteralImpl("info");
>>>
>>> and when getting agent names,
>>> Class c = Class.forName(classPar.className);
>>> is now
>>> Class c = Class.forName(classPar.getClassName());
>>>
>>> Everything compiles OK, but now I get alot of strange output:
>>>
>>> [driver_a] driver agent has started by central2
>>> [convoyFollower] I've started
>>> [convoyFollower] got told something about which vehicle to follow: central
>>> [driver] driver agent has started by central
>>> [Environment] The action turnToAngle(40.907071492138506) done by
>>> central is not implemented in the default environment.
>>> [Environment] The action turnToAngle(40.907071492138506) done by
>>> central is not implemented in the default environment.
>>> [Environment] The action turnToAngle(40.907071492138506) done by
>>> central is not implemented in the default environment.
>>> [Environment] The action turnToAngle(40.907071492138506) done by
>>> central is not implemented in the default environment.
>>> [convoyLeader] I've started
>>>
>>> The vehicle drives off, so some of the requests seem to be handled by
>>> the driver/central agent into simulatiom. I think this output has
>>> maybe replaced the Atom error from before?
>>>
>>> FYI, the turnToAngle comes from:
>>> +!requestTurnToAngle(A) : true<- //.print("Central Agent told to
>>> achieve angle ", A);
>>> turnToAngle(A).
>>>
>>> Which is a call made back to the simulation environment where
>>> turnToAngle is implemented (to physically turn the wheels of the
>>> simulated vehicle)
>>>
>>> Has something else changed between versions? Do I need to change some
>>> declarations from the simulated environment?
>>>
>>> Thanks!
>>>
>>> Vin
>>>
>>> Quoting Jomi Hubner<jom...@gm...>:
>>>
>>>> On 18/06/2011, at 02:43, Vin Baines wrote:
>>>>
>>>>> Hi Jomi,
>>>>>
>>>>> Many thanks for looking at this!
>>>>>
>>>>> I'm using 1.3.3, do you think I should upgrade?
>>>> Yes, if you don't mind. I remember to have fixed a related bug quite
>>>>  recently...
>>>>
>>>> Jomi
>>>>
>>>>> It's strange, the error doesn't crash anything, just seems to occur
>>>>>  for the first addition of the belief, these lines are run by the
>>>>> central control agent:
>>>>> .send(driver, tell, started(N));
>>>>> .send(driver, achieve, moveToXZ(100,2100)).
>>>>>
>>>>> and you can see the affect of them from the output:
>>>>>
>>>>> [driver] driver agent has started by central
>>>>> [driver] planning to move to 100 2100 and at 8.40046501159668
>>>>> 2019.8753662109375
>>>>> [Atom] You should not add annot 'source(self)' in atom cruise
>>>>>
>>>>> java.lang.Exception
>>>>> 	at jason.asSyntax.Atom.addAnnot(Atom.java:141)
>>>>> 	at jason.asSyntax.VarTerm.addAnnot(VarTerm.java:426)
>>>>> 	at
>>>>> jason.asSemantics.TransitionSystem.applyExecInt(TransitionSystem.java:466)
>>>>> 	at
>>>>> jason.asSemantics.TransitionSystem.applySemanticRule(TransitionSystem.java:126)
>>>>> 	at
>>>>> jason.asSemantics.TransitionSystem.reasoningCycle(TransitionSystem.java:868)
>>>>> 	at
>>>>> jason.infra.centralised.CentralisedAgArch.run(CentralisedAgArch.java:197)
>>>>> 	at java.lang.Thread.run(Thread.java:619)
>>>>>
>>>>> Vin
>>>>>
>>>>>
>>>>> Quoting Jomi Hubner<jom...@gm...>:
>>>>>
>>>>>> Dear Vin,
>>>>>>
>>>>>> I couldn't reproduce your error. Which version of Jason are you using?
>>>>>>
>>>>>> Jomi
>>>>>>
>>>>>> On 15/06/2011, at 08:11, Vin Baines wrote:
>>>>>>
>>>>>>> Hi All,
>>>>>>>
>>>>>>> Having a few problems with a scenario I'm working on, wondering if I'm
>>>>>>> doing something fundamentally wrong with the architecture..
>>>>>>>
>>>>>>> I've got a number of 'vehicles' which are collections of agents e.g.
>>>>>>> driver agent, control agent, fuel management agent. Control agent
>>>>>>> receives position updates from simulation, and can perform acts on
>>>>>>> simulated vehicle to control it.
>>>>>>>
>>>>>>> Problem I've got is with the driver agent, it should be able to
>>>>>>> achieve a goal of moving to an X Z position. I've got the underlying
>>>>>>> move controls working, but I'm not sure about the layout of my plan.
>>>>>>>
>>>>>>> I've cut out the basics of the driver agent below.
>>>>>>>
>>>>>>> But, what I'm not sure about: is this a reasonable approach? would I
>>>>>>> be better with 'if' statements? Something just doesn't feel quite
>>>>>>> right about the looping..
>>>>>>>
>>>>>>> It works, but I get a few errors,
>>>>>>> [Atom] You should not add annot 'source(self)' in atom cruise
>>>>>>> and usually a fail plan event when it arrives.
>>>>>>> [convoyFollower] Found a goal for which there is no
>>>>>>> relevant/applicable plan:+!followXZ[source(self)]
>>>>>>> [convoyFollower] No fail event was generated for     
>>>>>>> +!followXZ[source(self)]
>>>>>>>
>>>>>>> What I'm trying to do is have one vehicle follow another, by having
>>>>>>> the follower ask the leader its position. Next step is to get it
>>>>>>> following by asking the leader about its plans. The first step is
>>>>>>> working, but I'd be interested to hear whether I'm being sensible, or
>>>>>>> its just luck that its working!
>>>>>>>
>>>>>>>
>>>>>>> distanceRemaining(0, 50, arrivedAtDestination).
>>>>>>> distanceRemaining(51, 10000, cruise).
>>>>>>>
>>>>>>> +!arrivedAtDestination : true&  started(Owner)<- .print("arrived");
>>>>>>> 					.send(Owner, achieve, chosenSpeed(0));
>>>>>>> 					print("stopped");
>>>>>>> 					//ADD BELIEF WE'VE ARRIVED
>>>>>>> 					+arrived(true);
>>>>>>> 				  .drop_desire(moveToXZ(X,Z)).
>>>>>>> +!cruise : true&  started(Owner)<- //.print("cruising default");
>>>>>>> 					.send(Owner, achieve, chosenSpeed(10)).
>>>>>>>
>>>>>>> +!moveToXZ(X1,Z1) : arrived(true)<- .print("called the alternative
>>>>>>> moveToXZ to drop desire");						.drop_desire(moveToXZ(X,Z)).
>>>>>>>
>>>>>>> +!moveToXZ(X1,Z1) : entityUpdate(X,_,Z,_,_,_,_)
>>>>>>> &  started(Owner)
>>>>>>> &  DX=(X-X1)&  DZ=(Z-Z1)
>>>>>>> &   test.test1(X,X1,Z,Z1,R2)
>>>>>>> &  D1 = math.abs(DX)+math.abs(DZ)
>>>>>>> //AND CHECK NOT EXIST BELIEF WE'VE ARRIVED
>>>>>>> &  distanceRemaining(MIN,MAX,GOAL)&  MIN<= D1&  MAX>= D1
>>>>>>> 	<- .send(Owner,achieve,requestTurnToAngle(R2));
>>>>>>> 	.wait(1500);
>>>>>>> 	!GOAL;
>>>>>>> 	!!moveToXZ(X1,Z1).
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> ------------------------------------------------------------------------------
>>>>>>> EditLive Enterprise is the world's most technically advanced content
>>>>>>> authoring tool. Experience the power of Track Changes, Inline Image
>>>>>>> Editing and ensure content is compliant with Accessibility Checking.
>>>>>>> http://p.sf.net/sfu/ephox-dev2dev
>>>>>>> _______________________________________________
>>>>>>> Jason-users mailing list
>>>>>>> Jas...@li...
>>>>>>> https://lists.sourceforge.net/lists/listinfo/jason-users
>>>>>> --
>>>>>> Jomi Fred Hubner
>>>>>> Federal University of Santa Catarina
>>>>>> Department of Automation and Systems Engineering
>>>>>> PO Box 476, Florianópolis, SC
>>>>>> 88040-900 Brazil
>>>>>> http://www.das.ufsc.br/~jomi
>>>>>>
>>>>>>
>>>>>> ------------------------------------------------------------------------------
>>>>>> EditLive Enterprise is the world's most technically advanced content
>>>>>> authoring tool. Experience the power of Track Changes, Inline Image
>>>>>> Editing and ensure content is compliant with Accessibility Checking.
>>>>>> http://p.sf.net/sfu/ephox-dev2dev
>>>>>> _______________________________________________
>>>>>> Jason-users mailing list
>>>>>> Jas...@li...
>>>>>> https://lists.sourceforge.net/lists/listinfo/jason-users
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>> --
>>>> Jomi Fred Hubner
>>>> Federal University of Santa Catarina
>>>> Department of Automation and Systems Engineering
>>>> PO Box 476, Florianópolis, SC
>>>> 88040-900 Brazil
>>>> http://www.das.ufsc.br/~jomi
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Simplify data backup and recovery for your virtual environment   
>>> with vRanger.
>>> Installation's a snap, and flexible recovery options mean your   
>>> data is safe,
>>> secure and there when you need it. Data protection magic?
>>> Nope - It's vRanger. Get your free trial download today.
>>> http://p.sf.net/sfu/quest-sfdev2dev
>>> _______________________________________________
>>> Jason-users mailing list
>>> Jas...@li...
>>> https://lists.sourceforge.net/lists/listinfo/jason-users
>>
>>
>> ------------------------------------------------------------------------------
>> Simplify data backup and recovery for your virtual environment with vRanger.
>> Installation's a snap, and flexible recovery options mean your data is safe,
>> secure and there when you need it. Data protection magic?
>> Nope - It's vRanger. Get your free trial download today.
>> http://p.sf.net/sfu/quest-sfdev2dev
>> _______________________________________________
>> Jason-users mailing list
>> Jas...@li...
>> https://lists.sourceforge.net/lists/listinfo/jason-users
>>
>>

Hi Rafael,

I thought I was.. but maybe not. mas2j is basically:

	infrastructure: Centralised
	environment: br.furb.inf.tcc.tankcoders.jason.TankCodersEnvironment("-debug")
	agents:
	  central2 agentArchClass br.furb.inf.tcc.tankcoders.jason.TankAgArch;
	  central agentArchClass br.furb.inf.tcc.tankcoders.jason.TankAgArch;

Have attached those java files, not sure if helps much but might give  
a clue to the implementation.

In the simulated environment, the vehicles perform exactly as they did  
with jason 1.3.3 (I would add a link to the video but I'm having codec  
problems, scenario with 1.3.3 is here  
http://people.bath.ac.uk/vb216/follow.avi), the only differnce with  
1.3.4 is this new error message. Is that some additional error  
checking? I can see the methods are being called at each cycle as the  
vehicles behave as expected.

I'm trying to think through the code, I'm not sure what kind of return  
would be expected?

+Jomi - sorry, something went wrong with the zip file extraction on my  
PC, it hadn't extracted the javacc lib. I re-ran the extraction, and  
it worked fine the second time and no need for the additional javacc  
download.

Vin

Quoting Rafael H Bordini <r.b...@ac...>:

> Hi Vin,
>
> Did you specify the correct class with your simulated environment in the
> mas2j file? If so, you probably forgot to return (with true, probably)
> when your implementation of  "turnToAngle" in the environment finished
> changing the simulated model. The environment gives out that message for
> (external) actions that were not implemented yet.
>
> BTW, there are some strange parts in your code. For example, in a plan
> context "true & started(Owner)" could be written just "started(Owner)"
> (recall the identity laws of classical logic)
>
> Cheers,
>
> Rafael
>
> On 22/06/2011 08:27, Vin Baines wrote:
>> Hi Jomi,
>>
>> Hmm.. OK, download 1.3.4 and compiled it (had to add in a javacc.jar
>> which seems a new dependency?)
>>
>> Looks like there are a few changes too, I've changed over
>> Literal info = new Literal("info");
>> to
>> LiteralImpl info = new LiteralImpl("info");
>>
>> and when getting agent names,
>> Class c = Class.forName(classPar.className);
>> is now
>> Class c = Class.forName(classPar.getClassName());
>>
>> Everything compiles OK, but now I get alot of strange output:
>>
>> [driver_a] driver agent has started by central2
>> [convoyFollower] I've started
>> [convoyFollower] got told something about which vehicle to follow: central
>> [driver] driver agent has started by central
>> [Environment] The action turnToAngle(40.907071492138506) done by
>> central is not implemented in the default environment.
>> [Environment] The action turnToAngle(40.907071492138506) done by
>> central is not implemented in the default environment.
>> [Environment] The action turnToAngle(40.907071492138506) done by
>> central is not implemented in the default environment.
>> [Environment] The action turnToAngle(40.907071492138506) done by
>> central is not implemented in the default environment.
>> [convoyLeader] I've started
>>
>> The vehicle drives off, so some of the requests seem to be handled by
>> the driver/central agent into simulatiom. I think this output has
>> maybe replaced the Atom error from before?
>>
>> FYI, the turnToAngle comes from:
>> +!requestTurnToAngle(A) : true<- //.print("Central Agent told to
>> achieve angle ", A);
>> turnToAngle(A).
>>
>> Which is a call made back to the simulation environment where
>> turnToAngle is implemented (to physically turn the wheels of the
>> simulated vehicle)
>>
>> Has something else changed between versions? Do I need to change some
>> declarations from the simulated environment?
>>
>> Thanks!
>>
>> Vin
>>
>> Quoting Jomi Hubner<jom...@gm...>:
>>
>>> On 18/06/2011, at 02:43, Vin Baines wrote:
>>>
>>>> Hi Jomi,
>>>>
>>>> Many thanks for looking at this!
>>>>
>>>> I'm using 1.3.3, do you think I should upgrade?
>>> Yes, if you don't mind. I remember to have fixed a related bug quite
>>>   recently...
>>>
>>> Jomi
>>>
>>>> It's strange, the error doesn't crash anything, just seems to occur
>>>>   for the first addition of the belief, these lines are run by the
>>>> central control agent:
>>>> .send(driver, tell, started(N));
>>>> .send(driver, achieve, moveToXZ(100,2100)).
>>>>
>>>> and you can see the affect of them from the output:
>>>>
>>>> [driver] driver agent has started by central
>>>> [driver] planning to move to 100 2100 and at 8.40046501159668
>>>> 2019.8753662109375
>>>> [Atom] You should not add annot 'source(self)' in atom cruise
>>>>
>>>> java.lang.Exception
>>>> 	at jason.asSyntax.Atom.addAnnot(Atom.java:141)
>>>> 	at jason.asSyntax.VarTerm.addAnnot(VarTerm.java:426)
>>>> 	at
>>>> jason.asSemantics.TransitionSystem.applyExecInt(TransitionSystem.java:466)
>>>> 	at
>>>> jason.asSemantics.TransitionSystem.applySemanticRule(TransitionSystem.java:126)
>>>> 	at
>>>> jason.asSemantics.TransitionSystem.reasoningCycle(TransitionSystem.java:868)
>>>> 	at
>>>> jason.infra.centralised.CentralisedAgArch.run(CentralisedAgArch.java:197)
>>>> 	at java.lang.Thread.run(Thread.java:619)
>>>>
>>>> Vin
>>>>
>>>>
>>>> Quoting Jomi Hubner<jom...@gm...>:
>>>>
>>>>> Dear Vin,
>>>>>
>>>>> I couldn't reproduce your error. Which version of Jason are you using?
>>>>>
>>>>> Jomi
>>>>>
>>>>> On 15/06/2011, at 08:11, Vin Baines wrote:
>>>>>
>>>>>> Hi All,
>>>>>>
>>>>>> Having a few problems with a scenario I'm working on, wondering if I'm
>>>>>> doing something fundamentally wrong with the architecture..
>>>>>>
>>>>>> I've got a number of 'vehicles' which are collections of agents e.g.
>>>>>> driver agent, control agent, fuel management agent. Control agent
>>>>>> receives position updates from simulation, and can perform acts on
>>>>>> simulated vehicle to control it.
>>>>>>
>>>>>> Problem I've got is with the driver agent, it should be able to
>>>>>> achieve a goal of moving to an X Z position. I've got the underlying
>>>>>> move controls working, but I'm not sure about the layout of my plan.
>>>>>>
>>>>>> I've cut out the basics of the driver agent below.
>>>>>>
>>>>>> But, what I'm not sure about: is this a reasonable approach? would I
>>>>>> be better with 'if' statements? Something just doesn't feel quite
>>>>>> right about the looping..
>>>>>>
>>>>>> It works, but I get a few errors,
>>>>>> [Atom] You should not add annot 'source(self)' in atom cruise
>>>>>> and usually a fail plan event when it arrives.
>>>>>> [convoyFollower] Found a goal for which there is no
>>>>>> relevant/applicable plan:+!followXZ[source(self)]
>>>>>> [convoyFollower] No fail event was generated for   
>>>>>> +!followXZ[source(self)]
>>>>>>
>>>>>> What I'm trying to do is have one vehicle follow another, by having
>>>>>> the follower ask the leader its position. Next step is to get it
>>>>>> following by asking the leader about its plans. The first step is
>>>>>> working, but I'd be interested to hear whether I'm being sensible, or
>>>>>> its just luck that its working!
>>>>>>
>>>>>>
>>>>>> distanceRemaining(0, 50, arrivedAtDestination).
>>>>>> distanceRemaining(51, 10000, cruise).
>>>>>>
>>>>>> +!arrivedAtDestination : true&  started(Owner)<- .print("arrived");
>>>>>> 					.send(Owner, achieve, chosenSpeed(0));
>>>>>> 					print("stopped");
>>>>>> 					//ADD BELIEF WE'VE ARRIVED
>>>>>> 					+arrived(true);
>>>>>> 				  .drop_desire(moveToXZ(X,Z)).
>>>>>> +!cruise : true&  started(Owner)<- //.print("cruising default");
>>>>>> 					.send(Owner, achieve, chosenSpeed(10)).
>>>>>>
>>>>>> +!moveToXZ(X1,Z1) : arrived(true)<- .print("called the alternative
>>>>>> moveToXZ to drop desire");						.drop_desire(moveToXZ(X,Z)).
>>>>>>
>>>>>> +!moveToXZ(X1,Z1) : entityUpdate(X,_,Z,_,_,_,_)
>>>>>> &  started(Owner)
>>>>>> &  DX=(X-X1)&  DZ=(Z-Z1)
>>>>>> &   test.test1(X,X1,Z,Z1,R2)
>>>>>> &  D1 = math.abs(DX)+math.abs(DZ)
>>>>>> //AND CHECK NOT EXIST BELIEF WE'VE ARRIVED
>>>>>> &  distanceRemaining(MIN,MAX,GOAL)&  MIN<= D1&  MAX>= D1
>>>>>> 	<- .send(Owner,achieve,requestTurnToAngle(R2));
>>>>>> 	.wait(1500);
>>>>>> 	!GOAL;
>>>>>> 	!!moveToXZ(X1,Z1).
>>>>>>
>>>>>>
>>>>>>
>>>>>> ------------------------------------------------------------------------------
>>>>>> EditLive Enterprise is the world's most technically advanced content
>>>>>> authoring tool. Experience the power of Track Changes, Inline Image
>>>>>> Editing and ensure content is compliant with Accessibility Checking.
>>>>>> http://p.sf.net/sfu/ephox-dev2dev
>>>>>> _______________________________________________
>>>>>> Jason-users mailing list
>>>>>> Jas...@li...
>>>>>> https://lists.sourceforge.net/lists/listinfo/jason-users
>>>>> --
>>>>> Jomi Fred Hubner
>>>>> Federal University of Santa Catarina
>>>>> Department of Automation and Systems Engineering
>>>>> PO Box 476, Florianópolis, SC
>>>>> 88040-900 Brazil
>>>>> http://www.das.ufsc.br/~jomi
>>>>>
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> EditLive Enterprise is the world's most technically advanced content
>>>>> authoring tool. Experience the power of Track Changes, Inline Image
>>>>> Editing and ensure content is compliant with Accessibility Checking.
>>>>> http://p.sf.net/sfu/ephox-dev2dev
>>>>> _______________________________________________
>>>>> Jason-users mailing list
>>>>> Jas...@li...
>>>>> https://lists.sourceforge.net/lists/listinfo/jason-users
>>>>>
>>>>>
>>>>
>>>>
>>> --
>>> Jomi Fred Hubner
>>> Federal University of Santa Catarina
>>> Department of Automation and Systems Engineering
>>> PO Box 476, Florianópolis, SC
>>> 88040-900 Brazil
>>> http://www.das.ufsc.br/~jomi
>>>
>>>
>>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Simplify data backup and recovery for your virtual environment with vRanger.
>> Installation's a snap, and flexible recovery options mean your data is safe,
>> secure and there when you need it. Data protection magic?
>> Nope - It's vRanger. Get your free trial download today.
>> http://p.sf.net/sfu/quest-sfdev2dev
>> _______________________________________________
>> Jason-users mailing list
>> Jas...@li...
>> https://lists.sourceforge.net/lists/listinfo/jason-users
>
>
> ------------------------------------------------------------------------------
> Simplify data backup and recovery for your virtual environment with vRanger.
> Installation's a snap, and flexible recovery options mean your data is safe,
> secure and there when you need it. Data protection magic?
> Nope - It's vRanger. Get your free trial download today.
> http://p.sf.net/sfu/quest-sfdev2dev
> _______________________________________________
> Jason-users mailing list
> Jas...@li...
> https://lists.sourceforge.net/lists/listinfo/jason-users
>
>

Hi Vin,

Did you specify the correct class with your simulated environment in the 
mas2j file? If so, you probably forgot to return (with true, probably) 
when your implementation of  "turnToAngle" in the environment finished 
changing the simulated model. The environment gives out that message for 
(external) actions that were not implemented yet.

BTW, there are some strange parts in your code. For example, in a plan 
context "true & started(Owner)" could be written just "started(Owner)" 
(recall the identity laws of classical logic)

Cheers,

Rafael

On 22/06/2011 08:27, Vin Baines wrote:
> Hi Jomi,
>
> Hmm.. OK, download 1.3.4 and compiled it (had to add in a javacc.jar
> which seems a new dependency?)
>
> Looks like there are a few changes too, I've changed over
> Literal info = new Literal("info");
> to
> LiteralImpl info = new LiteralImpl("info");
>
> and when getting agent names,
> Class c = Class.forName(classPar.className);
> is now
> Class c = Class.forName(classPar.getClassName());
>
> Everything compiles OK, but now I get alot of strange output:
>
> [driver_a] driver agent has started by central2
> [convoyFollower] I've started
> [convoyFollower] got told something about which vehicle to follow: central
> [driver] driver agent has started by central
> [Environment] The action turnToAngle(40.907071492138506) done by
> central is not implemented in the default environment.
> [Environment] The action turnToAngle(40.907071492138506) done by
> central is not implemented in the default environment.
> [Environment] The action turnToAngle(40.907071492138506) done by
> central is not implemented in the default environment.
> [Environment] The action turnToAngle(40.907071492138506) done by
> central is not implemented in the default environment.
> [convoyLeader] I've started
>
> The vehicle drives off, so some of the requests seem to be handled by
> the driver/central agent into simulatiom. I think this output has
> maybe replaced the Atom error from before?
>
> FYI, the turnToAngle comes from:
> +!requestTurnToAngle(A) : true<- //.print("Central Agent told to
> achieve angle ", A);
> turnToAngle(A).
>
> Which is a call made back to the simulation environment where
> turnToAngle is implemented (to physically turn the wheels of the
> simulated vehicle)
>
> Has something else changed between versions? Do I need to change some
> declarations from the simulated environment?
>
> Thanks!
>
> Vin
>
> Quoting Jomi Hubner<jom...@gm...>:
>
>> On 18/06/2011, at 02:43, Vin Baines wrote:
>>
>>> Hi Jomi,
>>>
>>> Many thanks for looking at this!
>>>
>>> I'm using 1.3.3, do you think I should upgrade?
>> Yes, if you don't mind. I remember to have fixed a related bug quite
>>   recently...
>>
>> Jomi
>>
>>> It's strange, the error doesn't crash anything, just seems to occur
>>>   for the first addition of the belief, these lines are run by the
>>> central control agent:
>>> .send(driver, tell, started(N));
>>> .send(driver, achieve, moveToXZ(100,2100)).
>>>
>>> and you can see the affect of them from the output:
>>>
>>> [driver] driver agent has started by central
>>> [driver] planning to move to 100 2100 and at 8.40046501159668
>>> 2019.8753662109375
>>> [Atom] You should not add annot 'source(self)' in atom cruise
>>>
>>> java.lang.Exception
>>> 	at jason.asSyntax.Atom.addAnnot(Atom.java:141)
>>> 	at jason.asSyntax.VarTerm.addAnnot(VarTerm.java:426)
>>> 	at
>>> jason.asSemantics.TransitionSystem.applyExecInt(TransitionSystem.java:466)
>>> 	at
>>> jason.asSemantics.TransitionSystem.applySemanticRule(TransitionSystem.java:126)
>>> 	at
>>> jason.asSemantics.TransitionSystem.reasoningCycle(TransitionSystem.java:868)
>>> 	at
>>> jason.infra.centralised.CentralisedAgArch.run(CentralisedAgArch.java:197)
>>> 	at java.lang.Thread.run(Thread.java:619)
>>>
>>> Vin
>>>
>>>
>>> Quoting Jomi Hubner<jom...@gm...>:
>>>
>>>> Dear Vin,
>>>>
>>>> I couldn't reproduce your error. Which version of Jason are you using?
>>>>
>>>> Jomi
>>>>
>>>> On 15/06/2011, at 08:11, Vin Baines wrote:
>>>>
>>>>> Hi All,
>>>>>
>>>>> Having a few problems with a scenario I'm working on, wondering if I'm
>>>>> doing something fundamentally wrong with the architecture..
>>>>>
>>>>> I've got a number of 'vehicles' which are collections of agents e.g.
>>>>> driver agent, control agent, fuel management agent. Control agent
>>>>> receives position updates from simulation, and can perform acts on
>>>>> simulated vehicle to control it.
>>>>>
>>>>> Problem I've got is with the driver agent, it should be able to
>>>>> achieve a goal of moving to an X Z position. I've got the underlying
>>>>> move controls working, but I'm not sure about the layout of my plan.
>>>>>
>>>>> I've cut out the basics of the driver agent below.
>>>>>
>>>>> But, what I'm not sure about: is this a reasonable approach? would I
>>>>> be better with 'if' statements? Something just doesn't feel quite
>>>>> right about the looping..
>>>>>
>>>>> It works, but I get a few errors,
>>>>> [Atom] You should not add annot 'source(self)' in atom cruise
>>>>> and usually a fail plan event when it arrives.
>>>>> [convoyFollower] Found a goal for which there is no
>>>>> relevant/applicable plan:+!followXZ[source(self)]
>>>>> [convoyFollower] No fail event was generated for +!followXZ[source(self)]
>>>>>
>>>>> What I'm trying to do is have one vehicle follow another, by having
>>>>> the follower ask the leader its position. Next step is to get it
>>>>> following by asking the leader about its plans. The first step is
>>>>> working, but I'd be interested to hear whether I'm being sensible, or
>>>>> its just luck that its working!
>>>>>
>>>>>
>>>>> distanceRemaining(0, 50, arrivedAtDestination).
>>>>> distanceRemaining(51, 10000, cruise).
>>>>>
>>>>> +!arrivedAtDestination : true&  started(Owner)<- .print("arrived");
>>>>> 					.send(Owner, achieve, chosenSpeed(0));
>>>>> 					print("stopped");
>>>>> 					//ADD BELIEF WE'VE ARRIVED
>>>>> 					+arrived(true);
>>>>> 				  .drop_desire(moveToXZ(X,Z)).
>>>>> +!cruise : true&  started(Owner)<- //.print("cruising default");
>>>>> 					.send(Owner, achieve, chosenSpeed(10)).
>>>>>
>>>>> +!moveToXZ(X1,Z1) : arrived(true)<- .print("called the alternative
>>>>> moveToXZ to drop desire");						.drop_desire(moveToXZ(X,Z)).
>>>>>
>>>>> +!moveToXZ(X1,Z1) : entityUpdate(X,_,Z,_,_,_,_)
>>>>> &  started(Owner)
>>>>> &  DX=(X-X1)&  DZ=(Z-Z1)
>>>>> &   test.test1(X,X1,Z,Z1,R2)
>>>>> &  D1 = math.abs(DX)+math.abs(DZ)
>>>>> //AND CHECK NOT EXIST BELIEF WE'VE ARRIVED
>>>>> &  distanceRemaining(MIN,MAX,GOAL)&  MIN<= D1&  MAX>= D1
>>>>> 	<- .send(Owner,achieve,requestTurnToAngle(R2));
>>>>> 	.wait(1500);
>>>>> 	!GOAL;
>>>>> 	!!moveToXZ(X1,Z1).
>>>>>
>>>>>
>>>>>
>>>>> ------------------------------------------------------------------------------
>>>>> EditLive Enterprise is the world's most technically advanced content
>>>>> authoring tool. Experience the power of Track Changes, Inline Image
>>>>> Editing and ensure content is compliant with Accessibility Checking.
>>>>> http://p.sf.net/sfu/ephox-dev2dev
>>>>> _______________________________________________
>>>>> Jason-users mailing list
>>>>> Jas...@li...
>>>>> https://lists.sourceforge.net/lists/listinfo/jason-users
>>>> --
>>>> Jomi Fred Hubner
>>>> Federal University of Santa Catarina
>>>> Department of Automation and Systems Engineering
>>>> PO Box 476, Florianópolis, SC
>>>> 88040-900 Brazil
>>>> http://www.das.ufsc.br/~jomi
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> EditLive Enterprise is the world's most technically advanced content
>>>> authoring tool. Experience the power of Track Changes, Inline Image
>>>> Editing and ensure content is compliant with Accessibility Checking.
>>>> http://p.sf.net/sfu/ephox-dev2dev
>>>> _______________________________________________
>>>> Jason-users mailing list
>>>> Jas...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/jason-users
>>>>
>>>>
>>>
>>>
>> --
>> Jomi Fred Hubner
>> Federal University of Santa Catarina
>> Department of Automation and Systems Engineering
>> PO Box 476, Florianópolis, SC
>> 88040-900 Brazil
>> http://www.das.ufsc.br/~jomi
>>
>>
>>
>
>
>
> ------------------------------------------------------------------------------
> Simplify data backup and recovery for your virtual environment with vRanger.
> Installation's a snap, and flexible recovery options mean your data is safe,
> secure and there when you need it. Data protection magic?
> Nope - It's vRanger. Get your free trial download today.
> http://p.sf.net/sfu/quest-sfdev2dev
> _______________________________________________
> Jason-users mailing list
> Jas...@li...
> https://lists.sourceforge.net/lists/listinfo/jason-users

Hi Jomi,

Hmm.. OK, download 1.3.4 and compiled it (had to add in a javacc.jar  
which seems a new dependency?)

Looks like there are a few changes too, I've changed over
Literal info = new Literal("info");
to
LiteralImpl info = new LiteralImpl("info");

and when getting agent names,
Class c = Class.forName(classPar.className);
is now
Class c = Class.forName(classPar.getClassName());

Everything compiles OK, but now I get alot of strange output:

[driver_a] driver agent has started by central2
[convoyFollower] I've started
[convoyFollower] got told something about which vehicle to follow: central
[driver] driver agent has started by central
[Environment] The action turnToAngle(40.907071492138506) done by  
central is not implemented in the default environment.
[Environment] The action turnToAngle(40.907071492138506) done by  
central is not implemented in the default environment.
[Environment] The action turnToAngle(40.907071492138506) done by  
central is not implemented in the default environment.
[Environment] The action turnToAngle(40.907071492138506) done by  
central is not implemented in the default environment.
[convoyLeader] I've started

The vehicle drives off, so some of the requests seem to be handled by  
the driver/central agent into simulatiom. I think this output has  
maybe replaced the Atom error from before?

FYI, the turnToAngle comes from:
+!requestTurnToAngle(A) : true <- //.print("Central Agent told to  
achieve angle ", A);
turnToAngle(A).

Which is a call made back to the simulation environment where  
turnToAngle is implemented (to physically turn the wheels of the  
simulated vehicle)

Has something else changed between versions? Do I need to change some  
declarations from the simulated environment?

Thanks!

Vin

Quoting Jomi Hubner <jom...@gm...>:

>
> On 18/06/2011, at 02:43, Vin Baines wrote:
>
>> Hi Jomi,
>>
>> Many thanks for looking at this!
>>
>> I'm using 1.3.3, do you think I should upgrade?
>
> Yes, if you don't mind. I remember to have fixed a related bug quite  
>  recently...
>
> Jomi
>
>>
>> It's strange, the error doesn't crash anything, just seems to occur  
>>  for the first addition of the belief, these lines are run by the   
>> central control agent:
>> .send(driver, tell, started(N));
>> .send(driver, achieve, moveToXZ(100,2100)).
>>
>> and you can see the affect of them from the output:
>>
>> [driver] driver agent has started by central
>> [driver] planning to move to 100 2100 and at 8.40046501159668   
>> 2019.8753662109375
>> [Atom] You should not add annot 'source(self)' in atom cruise
>>
>> java.lang.Exception
>> 	at jason.asSyntax.Atom.addAnnot(Atom.java:141)
>> 	at jason.asSyntax.VarTerm.addAnnot(VarTerm.java:426)
>> 	at   
>> jason.asSemantics.TransitionSystem.applyExecInt(TransitionSystem.java:466)
>> 	at   
>> jason.asSemantics.TransitionSystem.applySemanticRule(TransitionSystem.java:126)
>> 	at   
>> jason.asSemantics.TransitionSystem.reasoningCycle(TransitionSystem.java:868)
>> 	at   
>> jason.infra.centralised.CentralisedAgArch.run(CentralisedAgArch.java:197)
>> 	at java.lang.Thread.run(Thread.java:619)
>>
>> Vin
>>
>>
>> Quoting Jomi Hubner <jom...@gm...>:
>>
>>> Dear Vin,
>>>
>>> I couldn't reproduce your error. Which version of Jason are you using?
>>>
>>> Jomi
>>>
>>> On 15/06/2011, at 08:11, Vin Baines wrote:
>>>
>>>> Hi All,
>>>>
>>>> Having a few problems with a scenario I'm working on, wondering if I'm
>>>> doing something fundamentally wrong with the architecture..
>>>>
>>>> I've got a number of 'vehicles' which are collections of agents e.g.
>>>> driver agent, control agent, fuel management agent. Control agent
>>>> receives position updates from simulation, and can perform acts on
>>>> simulated vehicle to control it.
>>>>
>>>> Problem I've got is with the driver agent, it should be able to
>>>> achieve a goal of moving to an X Z position. I've got the underlying
>>>> move controls working, but I'm not sure about the layout of my plan.
>>>>
>>>> I've cut out the basics of the driver agent below.
>>>>
>>>> But, what I'm not sure about: is this a reasonable approach? would I
>>>> be better with 'if' statements? Something just doesn't feel quite
>>>> right about the looping..
>>>>
>>>> It works, but I get a few errors,
>>>> [Atom] You should not add annot 'source(self)' in atom cruise
>>>> and usually a fail plan event when it arrives.
>>>> [convoyFollower] Found a goal for which there is no
>>>> relevant/applicable plan:+!followXZ[source(self)]
>>>> [convoyFollower] No fail event was generated for +!followXZ[source(self)]
>>>>
>>>> What I'm trying to do is have one vehicle follow another, by having
>>>> the follower ask the leader its position. Next step is to get it
>>>> following by asking the leader about its plans. The first step is
>>>> working, but I'd be interested to hear whether I'm being sensible, or
>>>> its just luck that its working!
>>>>
>>>>
>>>> distanceRemaining(0, 50, arrivedAtDestination).
>>>> distanceRemaining(51, 10000, cruise).
>>>>
>>>> +!arrivedAtDestination : true & started(Owner) <- .print("arrived");
>>>> 					.send(Owner, achieve, chosenSpeed(0));
>>>> 					print("stopped");
>>>> 					//ADD BELIEF WE'VE ARRIVED
>>>> 					+arrived(true);
>>>> 				  .drop_desire(moveToXZ(X,Z)).
>>>> +!cruise : true & started(Owner) <- //.print("cruising default");
>>>> 					.send(Owner, achieve, chosenSpeed(10)).
>>>>
>>>> +!moveToXZ(X1,Z1) : arrived(true) <- .print("called the alternative
>>>> moveToXZ to drop desire");						.drop_desire(moveToXZ(X,Z)).
>>>>
>>>> +!moveToXZ(X1,Z1) : entityUpdate(X,_,Z,_,_,_,_)
>>>> & started(Owner)
>>>> & DX=(X-X1) & DZ=(Z-Z1)
>>>> &  test.test1(X,X1,Z,Z1,R2)
>>>> & D1 = math.abs(DX)+math.abs(DZ)
>>>> //AND CHECK NOT EXIST BELIEF WE'VE ARRIVED
>>>> & distanceRemaining(MIN,MAX,GOAL) & MIN <= D1 & MAX >= D1
>>>> 	<- .send(Owner,achieve,requestTurnToAngle(R2));
>>>> 	.wait(1500);
>>>> 	!GOAL;
>>>> 	!!moveToXZ(X1,Z1).
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> EditLive Enterprise is the world's most technically advanced content
>>>> authoring tool. Experience the power of Track Changes, Inline Image
>>>> Editing and ensure content is compliant with Accessibility Checking.
>>>> http://p.sf.net/sfu/ephox-dev2dev
>>>> _______________________________________________
>>>> Jason-users mailing list
>>>> Jas...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/jason-users
>>>
>>> --
>>> Jomi Fred Hubner
>>> Federal University of Santa Catarina
>>> Department of Automation and Systems Engineering
>>> PO Box 476, Florianópolis, SC
>>> 88040-900 Brazil
>>> http://www.das.ufsc.br/~jomi
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> EditLive Enterprise is the world's most technically advanced content
>>> authoring tool. Experience the power of Track Changes, Inline Image
>>> Editing and ensure content is compliant with Accessibility Checking.
>>> http://p.sf.net/sfu/ephox-dev2dev
>>> _______________________________________________
>>> Jason-users mailing list
>>> Jas...@li...
>>> https://lists.sourceforge.net/lists/listinfo/jason-users
>>>
>>>
>>
>>
>>
>
> --
> Jomi Fred Hubner
> Federal University of Santa Catarina
> Department of Automation and Systems Engineering
> PO Box 476, Florianópolis, SC
> 88040-900 Brazil
> http://www.das.ufsc.br/~jomi
>
>
>

Hi Tina,

There is one data structure that records the percepts that all agents 
receive (everything that is visible to ALL agents) and then there are 
separate structures for each of the agents. This way you can model 
things that are publicly perceptible as well as things that only one or 
a group of agents can perceive (e.g. because of their location within an 
environment). To more more practical, if you wanted to completely clear 
up all percepts of all agents you need to execute clearPercepts() (to 
clearly only the global percepts) and then the individual percepts for 
each agent, e.g. clearPercepts(ag1), clearPercepts(ag2), etc. See also 
method

getEnvironmentInfraTier().getRuntimeServices().getAgentsNames()

in the API, if you need it.

Cheers,

Rafael


On 19/06/2011 11:50, jas...@li... wrote:
> The attached message has been automatically discarded.
>
>
> clearPercepts().eml
>
> Subject:
> clearPercepts()
> From:
> Tina Balke <t....@ba...>
> Date:
> Sun, 19 Jun 2011 15:48:59 +0100
>
> To:
> jas...@li...
>
>
> Hi Jomi and Rafael,
> I have a short question concerning the clearPercepts-method. I try to
> use clearPercepts(String ag) to clear the percepts of one particular
> agent, however looking at the simulation, the percepts are not always
> cleared when the method is called. That's why I'm currently trying to
> find a reason for this. Looking into your book, I saw that as
> explanation for the method you write "clearPercepts(A): delete all
> percepts in the list of percepts that are exclusive to agent A". Just
> for clarification: What does "exclusive" mean in this context, i.e. if
> two agents hold the same percept and and one used the clearPercepts(A)
> method on one of the agents, is this percept deleted?
> What could be other reasons for percepts nor being properly deleted?
> Thanks a lot for your help in advance,
> Tina
>
>

Dear Vin, 

I couldn't reproduce your error. Which version of Jason are you using?

Jomi

On 15/06/2011, at 08:11, Vin Baines wrote:

> Hi All,
> 
> Having a few problems with a scenario I'm working on, wondering if I'm  
> doing something fundamentally wrong with the architecture..
> 
> I've got a number of 'vehicles' which are collections of agents e.g.  
> driver agent, control agent, fuel management agent. Control agent  
> receives position updates from simulation, and can perform acts on  
> simulated vehicle to control it.
> 
> Problem I've got is with the driver agent, it should be able to  
> achieve a goal of moving to an X Z position. I've got the underlying  
> move controls working, but I'm not sure about the layout of my plan.
> 
> I've cut out the basics of the driver agent below.
> 
> But, what I'm not sure about: is this a reasonable approach? would I  
> be better with 'if' statements? Something just doesn't feel quite  
> right about the looping..
> 
> It works, but I get a few errors,
> [Atom] You should not add annot 'source(self)' in atom cruise
> and usually a fail plan event when it arrives.
> [convoyFollower] Found a goal for which there is no  
> relevant/applicable plan:+!followXZ[source(self)]
> [convoyFollower] No fail event was generated for +!followXZ[source(self)]
> 
> What I'm trying to do is have one vehicle follow another, by having  
> the follower ask the leader its position. Next step is to get it  
> following by asking the leader about its plans. The first step is  
> working, but I'd be interested to hear whether I'm being sensible, or  
> its just luck that its working!
> 
> 
> distanceRemaining(0, 50, arrivedAtDestination).
> distanceRemaining(51, 10000, cruise).
> 
> +!arrivedAtDestination : true & started(Owner) <- .print("arrived");
> 					.send(Owner, achieve, chosenSpeed(0));
> 					print("stopped");
> 					//ADD BELIEF WE'VE ARRIVED
> 					+arrived(true);
> 				  .drop_desire(moveToXZ(X,Z)).
> +!cruise : true & started(Owner) <- //.print("cruising default");
> 					.send(Owner, achieve, chosenSpeed(10)).
> 
> +!moveToXZ(X1,Z1) : arrived(true) <- .print("called the alternative  
> moveToXZ to drop desire");						.drop_desire(moveToXZ(X,Z)).
> 
> +!moveToXZ(X1,Z1) : entityUpdate(X,_,Z,_,_,_,_)
> & started(Owner)
> & DX=(X-X1) & DZ=(Z-Z1)
> &  test.test1(X,X1,Z,Z1,R2)
> & D1 = math.abs(DX)+math.abs(DZ)
> //AND CHECK NOT EXIST BELIEF WE'VE ARRIVED
> & distanceRemaining(MIN,MAX,GOAL) & MIN <= D1 & MAX >= D1
> 	<- .send(Owner,achieve,requestTurnToAngle(R2));
> 	.wait(1500);
> 	!GOAL;
> 	!!moveToXZ(X1,Z1).
> 
> 
> 
> ------------------------------------------------------------------------------
> EditLive Enterprise is the world's most technically advanced content
> authoring tool. Experience the power of Track Changes, Inline Image
> Editing and ensure content is compliant with Accessibility Checking.
> http://p.sf.net/sfu/ephox-dev2dev
> _______________________________________________
> Jason-users mailing list
> Jas...@li...
> https://lists.sourceforge.net/lists/listinfo/jason-users

-- 
Jomi Fred Hubner
Federal University of Santa Catarina
Department of Automation and Systems Engineering
PO Box 476, Florianópolis, SC
88040-900 Brazil
http://www.das.ufsc.br/~jomi

Hi All,

Having a few problems with a scenario I'm working on, wondering if I'm  
doing something fundamentally wrong with the architecture..

I've got a number of 'vehicles' which are collections of agents e.g.  
driver agent, control agent, fuel management agent. Control agent  
receives position updates from simulation, and can perform acts on  
simulated vehicle to control it.

Problem I've got is with the driver agent, it should be able to  
achieve a goal of moving to an X Z position. I've got the underlying  
move controls working, but I'm not sure about the layout of my plan.

I've cut out the basics of the driver agent below.

But, what I'm not sure about: is this a reasonable approach? would I  
be better with 'if' statements? Something just doesn't feel quite  
right about the looping..

It works, but I get a few errors,
[Atom] You should not add annot 'source(self)' in atom cruise
and usually a fail plan event when it arrives.
[convoyFollower] Found a goal for which there is no  
relevant/applicable plan:+!followXZ[source(self)]
[convoyFollower] No fail event was generated for +!followXZ[source(self)]

What I'm trying to do is have one vehicle follow another, by having  
the follower ask the leader its position. Next step is to get it  
following by asking the leader about its plans. The first step is  
working, but I'd be interested to hear whether I'm being sensible, or  
its just luck that its working!


distanceRemaining(0, 50, arrivedAtDestination).
distanceRemaining(51, 10000, cruise).

+!arrivedAtDestination : true & started(Owner) <- .print("arrived");
					.send(Owner, achieve, chosenSpeed(0));
					print("stopped");
					//ADD BELIEF WE'VE ARRIVED
					+arrived(true);
				  .drop_desire(moveToXZ(X,Z)).
+!cruise : true & started(Owner) <- //.print("cruising default");
					.send(Owner, achieve, chosenSpeed(10)).

+!moveToXZ(X1,Z1) : arrived(true) <- .print("called the alternative  
moveToXZ to drop desire");						.drop_desire(moveToXZ(X,Z)).

+!moveToXZ(X1,Z1) : entityUpdate(X,_,Z,_,_,_,_)
& started(Owner)
& DX=(X-X1) & DZ=(Z-Z1)
&  test.test1(X,X1,Z,Z1,R2)
& D1 = math.abs(DX)+math.abs(DZ)
//AND CHECK NOT EXIST BELIEF WE'VE ARRIVED
& distanceRemaining(MIN,MAX,GOAL) & MIN <= D1 & MAX >= D1
	<- .send(Owner,achieve,requestTurnToAngle(R2));
	.wait(1500);
	!GOAL;
	!!moveToXZ(X1,Z1).

Hi Francisco, 

For user defined functions, we have one instance for each agent. For some predefined functions (as abs, floor, ...) the instance is shared by all agents in the same JVM.

The current version of Jason does not provide an (easy) way to change that. If need such a feature, we can discuss it implementation.

Cheers,

Jomi

On 08/06/2011, at 13:37, Francisco Grimaldo Moreno wrote:

> 
> Hi all,
> 
> Regarding this issue, do ArithFunctions work the same way?
> 
> Thanks,
> Francisco Grimaldo
> 
> 
> A Dimecres, 8 de juny de 2011 14:37:38, Jomi Hubner va escriure:
>> Hi Alberto,
>> 
>> the default implementation creates one instance of the internal action for
>> each agent (15 instances in your case). If you want the agents in the same
>> JVM to share the instance, you have to include
>> 
>>    private static InternalAction singleton = null;
>>    public static InternalAction create() {
>>        if (singleton == null)
>>            singleton = new println(); // replace println for your class
>> name return singleton;
>>    }
>> 
>> in your class.
>> 
>> HTH,
>> 
>> Jomi
>> 
>> On 07/06/2011, at 19:29, Alberto Signoretti wrote:
>>> Hi,
>>> 
>>> i'm using some customized internal actions in a Jason Project. I have a
>>> doubt about the relationship of the Jason interpreter with the class that
>>> implements my customized internal action.
>>> 
>>> I'm using 15 diferent agents that, in diferent points of the simulation,
>>> need to call the customized IA. When the Jason starts, it will creat 15
>>> diferent instances (one for each individual agent) of the class that
>>> implements the IA? Can I use the instance of the IA to keep some state
>>> variables belonging to a particular agent?
>>> 
>>> Best regards.
>>> -------------------------------------------------------------------------
>>> ----- EditLive Enterprise is the world's most technically advanced content
>>> authoring tool. Experience the power of Track Changes, Inline Image
>>> Editing and ensure content is compliant with Accessibility Checking.
>>> http://p.sf.net/sfu/ephox-dev2dev________________________________________
>>> _______ Jason-users mailing list
>>> Jas...@li...
>>> https://lists.sourceforge.net/lists/listinfo/jason-users
>> 
> 
> -- 
> -------------------------------------------
> Dr. Francisco Grimaldo Moreno (PhD)
> (Id. GnuPG: 0xBAEC75A7)
> 
> Departament d'Informàtica
> Escola Tècnica Superior d'Enginyeria (ETSE)
> Universitat de València
> Avda. Vicent Andrés Estellés, s/n
> 46100 Burjassot-València (SPAIN)
> 
> Tel: +34 96 354 44 87 (+34 96 354 45 65)
> Fax: +34 96 354 47 68
> Email: fra...@uv...
> Web: http://www.uv.es/grimo
> -------------------------------------------
> 
> ------------------------------------------------------------------------------
> EditLive Enterprise is the world's most technically advanced content
> authoring tool. Experience the power of Track Changes, Inline Image
> Editing and ensure content is compliant with Accessibility Checking.
> http://p.sf.net/sfu/ephox-dev2dev
> _______________________________________________
> Jason-users mailing list
> Jas...@li...
> https://lists.sourceforge.net/lists/listinfo/jason-users

-- 
Jomi Fred Hubner
Federal University of Santa Catarina
Department of Automation and Systems Engineering
PO Box 476, Florianópolis, SC
88040-900 Brazil
http://www.das.ufsc.br/~jomi

Hi all,

Regarding this issue, do ArithFunctions work the same way?

Thanks,
Francisco Grimaldo


A Dimecres, 8 de juny de 2011 14:37:38, Jomi Hubner va escriure:
> Hi Alberto,
> 
> the default implementation creates one instance of the internal action for
>  each agent (15 instances in your case). If you want the agents in the same
>  JVM to share the instance, you have to include
> 
>     private static InternalAction singleton = null;
>     public static InternalAction create() {
>         if (singleton == null)
>             singleton = new println(); // replace println for your class
>  name return singleton;
>     }
> 
> in your class.
> 
> HTH,
> 
> Jomi
> 
> On 07/06/2011, at 19:29, Alberto Signoretti wrote:
> > Hi,
> >
> > i'm using some customized internal actions in a Jason Project. I have a
> > doubt about the relationship of the Jason interpreter with the class that
> > implements my customized internal action.
> >
> > I'm using 15 diferent agents that, in diferent points of the simulation,
> > need to call the customized IA. When the Jason starts, it will creat 15
> > diferent instances (one for each individual agent) of the class that
> > implements the IA? Can I use the instance of the IA to keep some state
> > variables belonging to a particular agent?
> >
> > Best regards.
> > -------------------------------------------------------------------------
> >----- EditLive Enterprise is the world's most technically advanced content
> > authoring tool. Experience the power of Track Changes, Inline Image
> > Editing and ensure content is compliant with Accessibility Checking.
> > http://p.sf.net/sfu/ephox-dev2dev________________________________________
> >_______ Jason-users mailing list
> > Jas...@li...
> > https://lists.sourceforge.net/lists/listinfo/jason-users
> 

-- 
-------------------------------------------
Dr. Francisco Grimaldo Moreno (PhD)
(Id. GnuPG: 0xBAEC75A7)

Departament d'Informàtica
Escola Tècnica Superior d'Enginyeria (ETSE)
Universitat de València
Avda. Vicent Andrés Estellés, s/n
46100 Burjassot-València (SPAIN)

Tel: +34 96 354 44 87 (+34 96 354 45 65)
Fax: +34 96 354 47 68
Email: fra...@uv...
Web: http://www.uv.es/grimo
-------------------------------------------

Hi Alberto,

the default implementation creates one instance of the internal action for each agent (15 instances in your case). If you want the agents in the same JVM to share the instance, you have to include 

    private static InternalAction singleton = null;
    public static InternalAction create() {
        if (singleton == null) 
            singleton = new println(); // replace println for your class name
        return singleton;
    }

in your class.

HTH,

Jomi
 
On 07/06/2011, at 19:29, Alberto Signoretti wrote:

> Hi,
> 
> i'm using some customized internal actions in a Jason Project. I have a doubt about the relationship of the Jason interpreter with the class that implements my customized internal action.
> 
> I'm using 15 diferent agents that, in diferent points of the simulation, need to call the customized IA. When the Jason starts, it will creat 15 diferent instances (one for each individual agent) of the class that implements the IA? Can I use the instance of the IA to keep some state variables belonging to a particular agent? 
> 
> Best regards.
> ------------------------------------------------------------------------------
> EditLive Enterprise is the world's most technically advanced content
> authoring tool. Experience the power of Track Changes, Inline Image
> Editing and ensure content is compliant with Accessibility Checking.
> http://p.sf.net/sfu/ephox-dev2dev_______________________________________________
> Jason-users mailing list
> Jas...@li...
> https://lists.sourceforge.net/lists/listinfo/jason-users

-- 
Jomi Fred Hubner
Federal University of Santa Catarina
Department of Automation and Systems Engineering
PO Box 476, Florianópolis, SC
88040-900 Brazil
http://www.das.ufsc.br/~jomi

Hi,

i'm using some customized internal actions in a Jason Project. I have a doubt 
about the relationship of the Jason interpreter with the class that implements 
my customized internal action.

I'm using 15 diferent agents that, in diferent points of the simulation, need to 
call the customized IA. When the Jason starts, it will creat 15 diferent 
instances (one for each individual agent) of the class that implements the IA? 
Can I use the instance of the IA to keep some state variables belonging to a 
particular agent? 

Best regards.