a cognitive map for an artificial agent

44
A Cognitive Map for an Artificial Agent Unmesh Kurup RPI [email protected] B. Chandrasekaran The Ohio State University [email protected]

Upload: tilden

Post on 09-Jan-2016

22 views

Category:

Documents


1 download

DESCRIPTION

A Cognitive Map for an Artificial Agent. Unmesh Kurup RPI [email protected] B. Chandrasekaran The Ohio State University [email protected]. Overview. Cognitive map Features Goals biSoar architecture Cognitive Map in biSoar Examples. Cognitive Map. Cognitive Map. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: A Cognitive Map for an Artificial Agent

A Cognitive Map for an Artificial Agent

Unmesh KurupRPI

[email protected]

B. ChandrasekaranThe Ohio State University

[email protected]

Page 2: A Cognitive Map for an Artificial Agent

Overview

• Cognitive map – Features– Goals

• biSoar architecture– Cognitive Map in biSoar– Examples

Page 3: A Cognitive Map for an Artificial Agent

Cognitive Map

Page 4: A Cognitive Map for an Artificial Agent

Cognitive Map

• Representation of large-scale space

Page 5: A Cognitive Map for an Artificial Agent

Cognitive Map

• Representation of large-scale space– Layout of a city

Page 6: A Cognitive Map for an Artificial Agent

Cognitive Map

• Representation of large-scale space– Layout of a city or a building

Page 7: A Cognitive Map for an Artificial Agent

Cognitive Map

• Representation of large-scale space– Layout of a city or a building

• Supports a number of problem solving tasks.

Page 8: A Cognitive Map for an Artificial Agent

Cognitive Map

• Representation of large-scale space– Layout of a city or a building

• Supports a number of problem solving tasks.– Route-finding: How can I get to the Radisson from

here? • Exit the hotel, Take a left, Right at 23rd St, Right on Clark.

Page 9: A Cognitive Map for an Artificial Agent

Cognitive Map

• Representation of large-scale space– Layout of a city or a building

• Supports a number of problem solving tasks.– Geo Recall: Is your hotel to the west of this hotel?

Page 10: A Cognitive Map for an Artificial Agent

Cognitive Map

• Representation of large-scale space– Layout of a city or a building

• Supports a number of problem solving tasks.– Finding shortcuts: Is there a shorter way to my

hotel?• Possible: Take a right on 20th st.

Page 11: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

Page 12: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic

Page 13: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic

If you take a left on Crystal Dr, you will get to the intersection of Crystal Dr and 23rd St

If you take a right on 23rd St, you will get to the intersection of 23rd St and Clark.

Page 14: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic– manageable

Page 15: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic– Manageable

Page 16: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic– Manageable

If you take a left on Crystal Dr, you will get to the intersection of Crystal Dr and 23rd St

If you take a right on 23rd St, you will get to the intersection of 23rd St and Clark.

If you take a right on Clark you will get to your hotel

vs

Page 17: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic– Manageable, updateable,

Page 18: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic– Manageable, updateable,

If you take a left on Crystal Dr, you will get to the intersection of Crystal Dr and 23rd St

If you take a right on 23rd St, you will get to the intersection of 23rd St and Clark.

If you take a right on Clark you will get to your hotel

If you take a right on 20th St, you will get to the intersection of 20th St and Clark.

New info? Just add it!

Page 19: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic– Manageable, updateable, composable

Page 20: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic– Manageable, updateable, composable

If you take a left on Crystal Dr, you will get to the intersection of Crystal Dr and 23rd St

If you take a right on 20th St, you will get to the intersection of 20th St and Clark.

If you take a right on Clark you will get to your hotel

Page 21: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic– Manageable, updateable, composable

• Has both symbolic and metric aspects

Page 22: A Cognitive Map for an Artificial Agent

Features of the Cognitive Map

• Non-holistic– Manageable, updateable, composable

• Has both symbolic and metric aspects

If you take a left on Crystal Dr, you will get to the intersection of Crystal Dr and 23rd St

Page 23: A Cognitive Map for an Artificial Agent

Goals

• Capture features– A non–holistic representation with both

symbolic and metric aspects

Page 24: A Cognitive Map for an Artificial Agent

Goals

• Capture features– A non–holistic representation with both

symbolic and metric aspects

• Cognitive architecture approach

Page 25: A Cognitive Map for an Artificial Agent

biSoar

A Bimodal Cognitive Architecture

Page 26: A Cognitive Map for an Artificial Agent

biSoar

• Soar + Diagrammatic Representation System (DRS)

Page 27: A Cognitive Map for an Artificial Agent

DRS - Diagrammatic Representation System (Chandra et. al. 2004)

• Diagrams consist of three types of objects – Points, Curves & Regions.

Page 28: A Cognitive Map for an Artificial Agent

DRS - Diagrammatic Representation System (Chandra et. al. 2004)

• Perceptual routines allow extraction of relationships between objects in the diagram.– Ex: LeftOf, RightOf etc

• Action routines allow the diagram to be modified– AddPoint, AddCurve etc

Page 29: A Cognitive Map for an Artificial Agent

biSoar

C

B

A

Symbolic component:Block (A), Block (B), Block (C), On (A,B), On (B,C)

Selected Operator: None

Working Memory

Diagrammatic component

Soar WM DRS

World

Working Memory:

Block (A), Block (B), Block (C), On (A,B), On (B,C)

Selected Operator: None

biSoar

Soar

Page 30: A Cognitive Map for an Artificial Agent

LTM and Learning in biSoar

• No change to LHS of LTM rules in biSoar• RHS can extract information from or modify

diagrammatic component as well.– If a and b are clear and goal is on(a,b) then

translate(a on b)

• Chunking in the bimodal case is straightforward.

Page 31: A Cognitive Map for an Artificial Agent

biSoar

• Soar + Diagrammatic Representation System (DRS)

• biSoar does not do– Any sort of image processing– Object recognition

• Assumes– a diagrammatic representation (DRS form) of

the input is available.

Page 32: A Cognitive Map for an Artificial Agent

Representing LSS in biSoar

If goal is find_next_location and curr location is x and traveling in direction Dx on route Rx, then

destination is location y, diagram is DRSx

Page 33: A Cognitive Map for an Artificial Agent

Representing LSS in biSoar

If goal is find_next_location and curr location is A and traveling in direction Dx on route Rx, then

destination is location B, diagram is DRSx

If goal is find_next_location and curr location is R2R5 and traveling Right on Route R2, then destination is P2, diagram is DRS1

R1

R2

R3

R4 R5

Page 34: A Cognitive Map for an Artificial Agent

Examples – Route-finding

• Given a map, find route from P1 to P2Route-finding Strategy• locate the starting & destination locations

in the map• make the starting location the current

location• Find the routes on which the current

location lies• For each route, find the directions of travel• for each route and direction of travel, find

the next location • calculate the Euclidean distance between

these new locations and the destinations• pick the location that is closest to the

destination and make that the current point

• repeat 3-8 until destination is reached

Page 35: A Cognitive Map for an Artificial Agent

Learning while route-finding

Example rules learned during wayfinding

Page 36: A Cognitive Map for an Artificial Agent

Within-task transfer

• Task1 – P1 to R3R5

R1

R2

R3

R4 R5

Page 37: A Cognitive Map for an Artificial Agent

Within-task transfer

• Task1 – P1 to R3R5, Task2 – P4 to P2

R1

R2

R3

R4 R5

Page 38: A Cognitive Map for an Artificial Agent

Within-task transfer

• Task1 – P1 to R3R5, Task2 – P4 to P2• New route-finding task – P4 to R3R5

R1

R2

R3

R4 R5

Page 39: A Cognitive Map for an Artificial Agent

Between-task transfer

• Geographic Recall problem• What’s the spatial relation between R1R3 and

R3R5?

Page 40: A Cognitive Map for an Artificial Agent

Finding short-cuts

R1

R4

R2

R3

If goal is find_routes and at R2R5 then there is a route r5 in the up direction.

Page 41: A Cognitive Map for an Artificial Agent

Finding short-cuts

R1

R4

R2

R3

If goal is find_routes and at R2R5 then there is a route r5 in the up direction.

Find route from P2 to P5

Page 42: A Cognitive Map for an Artificial Agent

Finding short-cuts

R1

R2R5

R3

R4

R2

R3

If goal is find_routes and at R2R5 then there is a route r5 in the up direction.

Find route from P2 to P5

Page 43: A Cognitive Map for an Artificial Agent

Conclusion

• biSoar’s CM (representation of LSS)– Is non-holistic– has metric and non-metric information– Can be used to solve a variety of tasks

involving LSS.– Supports learning and transfer of learned

information within and between tasks.

Page 44: A Cognitive Map for an Artificial Agent

Thanks!