0.25
0.5
0.75
1.25
1.5
1.75
2
Translation-based approaches to conformant and contingent planning
Published on Jul 21, 20113899 Views
Conformant planning is the problem of finding a sequence of actions for achieving a goal in the presence of uncertainty in the initial state or in action effects. On the other hand, Contingent plannin
Related categories
Chapter list
Translation-based approaches to Conformant and Contingent Planning00:00
Get it real! (1)00:13
Get it real! (2)00:42
Get it real! (3)01:03
Problem addressed in this tutorial01:30
Classical Planning (1)02:18
Conformant Planning (1)02:58
Classical problem for one state of a Conformant (1)03:33
Classical problem for one state of a Conformant (2)04:02
Conformant Planning (2)04:35
Contingent Planning04:41
Translation to Classical planning05:23
Classical Planning (2)06:16
Classical Planning (3)06:18
Classical Planning Syntax06:53
Classical Planning Model07:13
Classical Planning (4)07:45
State-of-the-art Classical Planning (1)07:55
State-of-the-art Classical Planning (2)08:16
Conformant Planning (3)08:55
Examples10:13
Omit precondition if true12:12
Look-n-grab 8x812:31
Conformant Planning: The Trouble with Incomplete Info13:30
Why it’s important? (1)14:40
Why it’s important? (2)14:49
Why it’s important? (3)15:12
Why it’s important? (4)15:26
Conformant Planning Syntax16:02
Conformant Planning: Semantic (1)16:24
Conformant Planning: Semantic (2)16:40
Conformant Planning: Semantic (3)16:47
Conformant Planning (4)16:51
Belief space search17:02
Roadmap of First Part18:35
Translation from P into K0(P) (1)19:40
Translation from P into K0(P) (2)20:20
Translation from P into K0(P) (3)20:26
Translation from P into K0(P) (4)21:13
Translation from P into K0(P) (5)21:23
Translation from P into K0(P) (6)21:40
Translation from P into K0(P) (7)21:45
Translation from P into K0(P) (8)21:56
Translation from P into K0(P) (9)22:27
K0 example (1)23:17
K0 example (2)24:43
K0 example. Cancellation rules (1)26:55
K0 example. Cancellation rules (2)28:20
K0 example. Cancellation rules (3)29:42
Basic Properties and Extensions (1)32:20
Basic Properties and Extensions (2)32:33
Key elements in Translation KT;M(P)32:37
Intuition of merge actions (1)33:20
Intuition of merge actions (2)33:33
Intuition of merge actions (3)33:58
Translation from P into KT;M(P) (1)34:14
Translation from P into KT;M(P) (2)34:17
Translation from P into KT;M(P) (3)34:37
Translation from P into KT;M(P) (4)34:41
Translation from P into KT;M(P) (5)35:51
Translation from P into KT;M(P) (6)35:53
Translation from P into KT;M(P) (7)35:54
Translation from P into KT;M(P) (8)36:01
Idea of KT;M(P) (1)36:47
Idea of KT;M(P) (2)36:50
Idea of KT;M(P) (3)36:51
Idea of KT;M(P) (4)37:41
Idea of KT;M(P) (5)37:45
Example of T;M (1)37:59
Interesting properties of the translation KT;M?38:42
Properties of Translation KT;M39:27
Soundness (1)39:44
Soundness (2)39:47
Soundness (3)40:23
A complete but exponential instance of KT;M(P): Ks0 (1)41:26
A complete but exponential instance of KT;M(P): Ks0 (2)41:51
A complete but exponential instance of KT;M(P): Ks0 (3)41:53
A complete but exponential instance of KT;M(P): Ks0 (4)42:32
Example: complete but compact instance of KT;M (1)42:46
Example: complete but compact instance of KT;M (1)42:54
Example: complete but compact instance of KT;M (2)43:18
Example: complete but compact instance of KT;M (3)43:45
Covering Translation (1)43:48
Covering44:29
Relevance44:41
Relevant Clauses (1)46:06
Relevant Clauses (2)46:20
Relevant Clauses (3)46:34
Relevant Clauses (4)47:20
Relevant Clauses (5)47:47
Satisfy48:01
Example Satisfy (1)49:21
Grid problem51:32
Example Satisfy (2)51:34
Covering Translation (2)51:35
Example of Covering Translation (1)52:07
Example of Covering Translation (2)52:21
Cover it!52:25
Width (1)53:20
Width (2)53:22
Width (3)53:23
Width (4)53:24
Width (5)53:25
Width (6)54:33
Width (examples) (1)54:44
Width (examples) (2)54:47
Width (examples) (3)54:48
Translation Ki(P) (1)54:48
Translation Ki(P) (2)55:02
Translation Ki(P) (3)55:46
Width of Conformant Benchmarks56:20
Width of some problems57:05
Conformant Width: intuitions (1)57:10
Conformant Width: intuitions (2)57:11
Basis (1)58:13
Basis (2)58:59
Basis examples59:55
Monotonicity59:56
There exist a Basis! (1)59:57
There exist a Basis! (2)59:57
Other instances of KT;M? (1)59:58
Other instances of KT;M? (2)01:00:17
Other instances of KT;M? (3)01:00:23
Other instances of KT;M? (4)01:00:24
Translation Kmodels(P) (1)01:00:52
Translation Kmodels(P) (2)01:00:57
The planner T001:01:04
Digression: on conditional effects01:01:25
Sampling (1)01:02:04
Sampling (2)01:02:05
Sampling (3)01:02:05
Sampling (4)01:02:05
Sampling (5)01:02:06
Sampling (6)01:02:07
Related Work (1)01:02:07
Related Work (2)01:02:08
T0 vs CpA (1)01:02:09
T0 vs CpA (2)01:02:09
T0 vs CpA (3)01:02:09
Summary of first part01:02:10
Translation-based Approaches to Conformant and Contingent Planning (Part 2)01:02:36
Contingent Planning01:02:43
Action selection in Wumpus01:03:10
Contingent Planning: Sensing and Incomplete Information01:03:57
A Translation-based approach to Contingent Planning01:04:31
Compiling into classical planning: the CLG approach01:06:07
Translation XT,M(P)01:07:25
Complete Translation XS0 (P)01:08:56
Example: Problem P (1)01:09:42
Example: Problem P (2)01:10:10
Example Problem P - Actions01:10:38
Example XS0 (P) translation01:11:08
Example with XS0 (P) translation - A possible Plan (1)01:11:48
Example with XS0 (P) translation - A possible Plan (2)01:12:03
Example with XS0 (P) translation goto(corridor, panel) (1)01:12:26
Example with XS0 (P) translation goto(corridor, panel) (2)01:12:35
Example with XS0 (P) translation goto(corridor, panel) (3)01:12:40
Example with XS0 (P) translation inspect-panel (1)01:12:43
Example with XS0 (P) translation inspect-panel (2)01:12:45
Example with XS0 (P) translation inspect-panel (3)01:13:14
Example with XS0 (P) translation tag-refutation: KL/t ∧K ¬L →K¬t (1)01:13:19
Example with XS0 (P) translation tag-refutation: KL/t ∧K ¬L →K¬t (2)01:13:20
Example with XS0 (P) translation tag-refutation: KL/t ∧K ¬L →K¬t (3)01:13:37
Example with XS0 (P) translation tag-refutation: KL/t ∧K ¬L →K¬t (4)01:13:41
General Translations that are Complete01:13:52
Width and Complexity01:14:33
where are we? (1)01:15:35
where are we? (2)01:15:36
where are we? (3)01:15:44
Relaxation X+(P)01:15:49
Relaxing on action preconditions01:17:22
Example with X+(P) (1)01:17:49
Example with X+(P) (2)01:17:51
Example with X+(P) (3)01:17:52
Example with X+(P) (4)01:18:12
Example with X+(P) applying derivation rules (1)01:18:15
Example with X+(P) applying derivation rules (2)01:18:18
Example with X+(P) applying derivation rules (3)01:18:35
Example with X+(P) a possible plan01:18:36
Closed Loop Greedy Planner01:19:20
Using assumptions on sensing outcome01:20:03
Another approach on how to Solve contingent problems with classical planners01:21:30
Planning under optimism (1)01:22:38
Planning under optimism (2)01:24:06
Dead-ends [Albore & Geffner 2009]01:24:15
Example with no full solution plan01:24:54
Planners for Problems with No Strong Solutions01:25:29
Encoding Assumptions Into CLG+ (pay-for-tags)01:26:06
Use of Assumptions01:26:41
Problems with Dead-end States01:27:14
Problems with Pure Dead-end States01:27:23
Problems with High Contingent Width01:27:24
Summary of Second Part01:27:24
Summary of the tutorial01:28:19
Conclusions01:29:00