Changes have been pushed for the repository "fawkes.git".
(Fawkes Robotics Software Framework)

Clone:  g...@git.fawkesrobotics.org:fawkes.git
Gitweb: http://git.fawkesrobotics.org/fawkes.git
Trac:   http://trac.fawkesrobotics.org

The branch, mgomaa/clips-executive-possible-bugs-and-where-to-find-them has 
been created
        at  ab4109ce5f0483a032bcbb8b672170fe71d05ed9 (commit)

http://git.fawkesrobotics.org/fawkes.git/mgomaa/clips-executive-possible-bugs-and-where-to-find-them

- *Log* ---------------------------------------------------------------
commit ab4109ce5f0483a032bcbb8b672170fe71d05ed9
Author:     Mostafa Gomaa <go...@kbsg.rwth-aachen.de>
AuthorDate: Tue Apr 17 00:59:57 2018 +0200
Commit:     Mostafa Gomaa <go...@kbsg.rwth-aachen.de>
CommitDate: Tue Apr 17 01:06:29 2018 +0200

    clips-exec: comments describing some possible/- bugs
    
    Will be used as ground for trying/discussing different solutions

http://git.fawkesrobotics.org/fawkes.git/commit/ab4109c
http://trac.fawkesrobotics.org/changeset/ab4109c

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -


- *Summary* -----------------------------------------------------------


- *Diffs* -------------------------------------------------------------

- *commit* ab4109ce5f0483a032bcbb8b672170fe71d05ed9 - - - - - - - - - -
Author:  Mostafa Gomaa <go...@kbsg.rwth-aachen.de>
Date:    Tue Apr 17 00:59:57 2018 +0200
Subject: clips-exec: comments describing some possible/- bugs

 src/plugins/clips-executive/clips/domain.clp |   11 +++++++++++
 1 files changed, 11 insertions(+), 0 deletions(-)

_Diff for modified files_:
diff --git a/src/plugins/clips-executive/clips/domain.clp 
b/src/plugins/clips-executive/clips/domain.clp
index 6a5cc7c..44c08a4 100644
--- a/src/plugins/clips-executive/clips/domain.clp
+++ b/src/plugins/clips-executive/clips/domain.clp
@@ -326,6 +326,8 @@
 )
 
 (defrule domain-check-if-atomic-precondition-is-satisfied
+  ;possible-bug:: No other rule exists to make is-satisfiable FALSE 
+  ;               if the domain-fact does no longer exist
   ?precond <- (domain-atomic-precondition
                 (goal-id ?g) (plan-id ?p)
                 (is-satisfied FALSE)
@@ -341,6 +343,8 @@
   "A negative precondition is satisfied iff its (only) child is not satisfied.
    Note that we need a second rule that retracts the fact if the child is
    asserted."
+   ;possible-bug:: This could fire on the initial false value of is-satisifed.
+   ;               Even if, it was satisfied but yet to be set
   ?precond <- (domain-precondition
                 (type negation)
                 (grounded TRUE)
@@ -364,6 +368,7 @@
   (modify ?precond (is-satisfied TRUE))
 )
 
+
 (defrule domain-retract-negative-precondition-if-child-is-satisfied
   "If a negative precondition's child is satisfied, the precondition is not
    satisfied anymore."
@@ -392,6 +397,9 @@
 
 (defrule domain-check-if-conjunctive-precondition-is-satisfied
   "All the precondition's children must be satisfied."
+  ;BUG:: if the atomic-preconditions/nested-precondition where never grounded,
+  ; this will still fire making the conjunction satisfiable and hence the 
action executable
+  ; and not EVEN check if the atomic preconditions where satisfied at all
   ?precond <- (domain-precondition
                 (name ?pn)
                 (type conjunction)
@@ -587,6 +595,9 @@
   "If the precondition of an action is satisfied, the action is executable."
   ?action <- (plan-action (id ?action-id) (goal-id ?g) (plan-id ?p)
                           (executable FALSE))
+  ;possible-bug::: (part-of ?action-id) is missing which could cause any random
+  ;                 satisfied precondition that was grounded by that actions-id
+  ;                 to render the action executable
   (domain-precondition (plan-id ?p) (goal-id ?g) (grounded-with ?action-id)
                        (is-satisfied TRUE))
 =>




-- 
Fawkes Robotics Framework                 http://www.fawkesrobotics.org
_______________________________________________
fawkes-commits mailing list
fawkes-commits@lists.kbsg.rwth-aachen.de
https://lists.kbsg.rwth-aachen.de/listinfo/fawkes-commits

Reply via email to