An Adaptive Architecture
for Physical Agents
Pat Langley
Computational Learning Laboratory
Center for the Study of Language and Information
Stanford University, Stanford, California USA
http://cll.stanford.edu/
Thanks to D. Choi, K. Cummings, N. Nejati, S. Rogers, S. Sage, D. Shapiro, and J. Xuan
for their contributions. This talk reports research. funded by grants from DARPA IPTO
and the US National Science Foundation, which are not responsible for its contents.
General Cognitive Systems
The original goal of artificial intelligence was to design and
implement computational artifacts that:
 combined many cognitive abilities in an integrated system;
 exhibited the same level of intelligence as humans;
 utilized its intelligence in a general way across domains.
Instead, modern AI has divided into many subfields that care
little about systems, generality, or even intelligence.
But the challenge remains and we need far more research on
general cognitive systems.
The Domain of In-City Driving
Consider driving a vehicle
in a city, which requires:
 selecting routes
 obeying traffic lights
 avoiding collisions
 being polite to others
 finding addresses
 staying in the lane
 parking safely
 stopping for pedestrians
 following other vehicles
 delivering packages
These tasks range from
low-level execution to
high-level reasoning.
The Fragmentation of AI Research
language
planning
perception

reasoning
action
learning
Newell’s Vision
In 1973, Allen Newell argued “You can’t play twenty questions
with nature and win”. Instead, he proposed that we:
 move beyond isolated phenomena and capabilities to develop
complete models of intelligent behavior;
 demonstrate our systems’ intelligence on the same range of
domains and tasks as humans can handle;
 evaluate these systems in terms of generality and flexibility
rather than success on a single application domain.
However, there are different paths toward achieving these goals.
An Architecture with Communicating Modules
language
planning
perception
reasoning
action
learning
software engineering / multi-agent systems
An Architecture with Shared Short-Term Memory
language
planning
perception
short-term
beliefs and goals
reasoning
action
learning
blackboard architectures
Architectures and Constraints
Newell’s vision for research on theories of intelligence was that:
 agent architectures should make strong theoretical assumptions
about the nature of the mind.
 architectural designs should change only gradually, as new
structures or processes are determined necessary.
 later design choices should be constrained heavily by earlier
ones, not made independently.
A successful architecture is all about mutual constraints, and it
should provide a unified theory of intelligent behavior.
He associated these aims with the idea of a cognitive architecture.
An Architecture with Shared Long-Term Memory
language
perception
planning
short-term
beliefs and goals
long-term memory
structures
action
learning
cognitive architectures
reasoning
A Constrained Cognitive Architecture
language
perception
planning
short-term
beliefs and goals
long-term memory
structures
action
learning
reasoning
The ICARUS Architecture
In this talk I will use one such framework  ICARUS  to illustrate
the advantages of cognitive architectures.
Like previous candidates, it incorporates ideas from theories of
human problem solving and reasoning.
However, ICARUS is also distinctive in its concern with:
 physical agents that operate in an external environment;
 the hierarchical structure of knowledge and its origin.
These concerns have led to different assumptions than earlier
cognitive architectures like ACT-R, Soar, and Prodigy.
Theoretical Claims of ICARUS
Our recent work on ICARUS has been guided by six principles:
1. Cognition grounded in perception and action
2. Cognitive separation of categories and skills
3. Hierarchical organization of long-term memory
4. Cumulative learning of skill/concept hierarchies
5. Correspondence of long-term/short-term structures
6. Modulation of symbolic structures with numeric content
These ideas distinguish ICARUS from most other architectures.
Architectural Commitment to Memories
A cognitive architecture makes a specific commitment to:
 long-term memories that store knowledge and procedures;
 short-term memories that store beliefs and goals;
 sensori-motor memories that hold percepts and actions.
Each memory is responsible for different content that the agent
uses in its activities.
ICARUS’ Memories
Perceptual
Buffer
Long-Term
Conceptual
Memory
Short-Term
Conceptual
Memory
Environment
Long-Term
Skill Memory
Short-Term
Goal/Skill
Memory
Motor
Buffer
Architectural Commitment to Representations
For each memory, a cognitive architecture also commits to:
 the encoding of contents in that memory;
 the organization of structures within the memory;
 the connections among structures across memories.
Most cognitive architectures rely upon formalisms similar to
predicate calculus that express relational content.
These build on the central assumption of AI that intelligence
involves the manipulation of list structures.
ICARUS’ Percepts are Objects with Attributes
(self me speed 24.0 wheel-angle 0.02 limit 25.0 road-angle 0.06)
(segment g1059 street 2 dist -5.0 latdist 15.0)
(segment g1050 street A dist -45.0 latdist nil)
(segment g1049 street A dist oor latdist nil)
(lane-line g1073 length 100.0 width 0.5 dist 35.0 angle 1.57 color white)
(lane-line g1074 length 100.0 width 0.5 dist 15.0 angle 1.57 color white)
(lane-line g1072 length 100.0 width 0.5 dist 25.0 angle 1.57 color yellow)
(lane-line g1100 length 100.0 width 0.5 dist -15.0 angle 0.0 color white)
(lane-line g1101 length 100.0 width 0.5 dist 5.0 angle 0.0 color white)
(lane-line g1099 length 100.0 width 0.5 dist -5.0 angle 0.0 color yellow)
(lane-line g1104 length 100.0 width 0.5 dist 5.0 angle 0.0 color white)
(intersection g1021 street A cross 2 dist -5.0 latdist nil)
(building g943 address 246 c1dist 43.69 c1angle -0.73 c2dist nil c2angle nil)
(building g941 address 246 c1dist 30.10 c1angle -1.30 c2dist 43.70 c2angle -0.73)
(building g939 address 197 c1dist 30.10 c1angle -1.30 c2dist 33.40 c2angle -2.10)
(building g943 address 172 c1dist 33.40 c1angle -2.09 c2dist 50.39 c2angle -2.53)
(sidewalk g975 dist 15.0 angle 0.0)
(sidewalk g978 dist 5.0 angle 1.57)
ICARUS’ Beliefs are Relations Among Objects
(not-on-street me g2980)
(not-approaching-cross-street me g2980)
(current-street me A)
(not-delivered g2980)
(in-leftmost-lane me g2533)
(fast-for-right-turn me)
(driving-in-segment me g2480 g2533)
(steering-wheel-straight me)
(aligned-with-lane me g2533)
(on-right-side-of-road me)
(buildings-on-right me g2231 g2230 g2480)
(buildings-on-right me g2231 g2222 g2480)
(buildings-on-right me g2231 g2211 g2480)
(buildings-on-right me g2230 g2222 g2480)
(buildings-on-right me g2230 g2211 g2480)
(buildings-on-right me g2222 g2211 g2480)
(buildings-on-left me g2366 g2480)
(buildings-on-left me g2370 g2480)
(currrent-building me g2222)
(not-on-cross-street me g2980)
(current-segment me g2480)
(in-U-turn-lane me g2533)
(lane-to-right me g2533)
(fast-for-U-turn me)
(at-speed-for-cruise me)
(centered-in-lane me g2533)
(in-lane me g2533)
(in-segment me g2480)
(increasing me g2231 g2230 g2480)
(increasing me g2231 g2222 g2480)
(increasing me g2231 g2211 g2480)
(increasing me g2230 g2222 g2480)
(increasing me g2230 g2211 g2480)
(increasing me g2222 g2211 g2480)
(buildings-on-left me g2368 g2480)
(buildings-on-left me g2372 g2480)
Teleoreactive Logic Programs
ICARUS encodes long-term knowledge of three general types:
 Concepts: A set of conjunctive relational inference rules;
 Primitive skills: A set of durative STRIPS operators;
 Nonprimitive skills: A set of clauses which specify:
 a head that indicates a goal the method achieves;
 a set of (possibly defined) preconditions;
 one or more ordered subskills for achieving the goal.
These teleoreactive logic programs can be executed reactively
but in a goal-directed manner (Nilsson, 1994).
ICARUS Concepts for In-City Driving
(in-segment (?self ?sg)
:percepts ((self ?self segment ?sg) (segment ?sg)))
(aligned-with-lane (?self ?lane)
:percepts ((self ?self) (lane-line ?lane angle ?angle))
:positives ((in-lane ?self ?lane))
:tests
((> ?angle -0.05) (< ?angle 0.05)) )
(on-street (?self ?packet)
:percepts ((self ?self) (packet ?packet street ?street)
(segment ?sg street ?street))
:positives ((not-delivered ?packet) (current-segment ?self ?sg)) )
(increasing-direction (?self)
:percepts ((self ?self))
:positives ((increasing ?b1 ?b2))
:negatives ((decreasing ?b3 ?b4)) )
ICARUS Skills for In-City Driving
(on-street-right-direction (?self ?packet)
:percepts ((self ?self segment ?segment direction ?dir)
(building ?landmark))
:start
((on-street-wrong-direction ?self ?packet))
:ordered
((get-in-U-turn-lane ?self) (prepare-for-U-turn ?self)
(steer-for-U-turn ?self ?landmark)) )
(get-aligned-in-segment (?self ?sg)
:percepts ((lane-line ?lane angle ?angle))
:requires ((in-lane ?self ?lane))
:effects
((aligned-with-lane ?self ?lane))
:actions
((steer (times ?angle 2))) )
(steer-for-right-turn (?self ?int ?endsg)
:percepts ((self ?self speed ?speed) (intersection ?int cross ?cross)
(segment ?endsg street ?cross angle ?angle))
:start
((ready-for-right-turn ?self ?int))
:effects
((in-segment ?self ?endsg))
:actions
((times steer 2)) )
Hierarchical Structure of Long-Term Memory
ICARUS organizes both concepts and skills in a hierarchical manner.
concepts
Each concept is defined in terms of other concepts and/or percepts.
skills
Each skill is defined in terms of other skills, concepts, and percepts.
Hierarchical Structure of Long-Term Memory
ICARUS interleaves its long-term memories for concepts and skills.
concepts
skills
For example, the skill highlighted here refers directly to
the highlighted concepts.
Architectural Commitment to Processes
In addition, a cognitive architecture makes commitments about:
 performance processes for:
 retrieval, matching, and selection
 inference and problem solving
 perception and motor control
 learning processes that:
 generate new long-term knowledge structures
 refine and modulate existing structures
In most cognitive architectures, performance and learning are
tightly intertwined.
ICARUS’ Functional Processes
Perceptual
Buffer
Long-Term
Conceptual
Memory
Long-Term
Skill Memory
Conceptual
Inference
Problem Solving
Skill Learning
Short-Term
Conceptual
Memory
Perception
Skill Retrieval
Environment
Short-Term
Goal/Skill
Memory
Skill
Execution
Motor
Buffer
The ICARUS Control Cycle
On each successive execution cycle, the ICARUS architecture:
1. places descriptions of sensed objects in the perceptual buffer;
2. infers instances of concepts implied by the current situation;
3. finds paths through the skill hierarchy from top-level goals;
4. selects one or more applicable skill paths for execution;
5. invokes the actions associated with each selected path.
Thus, ICARUS agents are examples of what Nilsson (1994) refers
to as teleoreactive systems.
Basic ICARUS Processes
ICARUS matches patterns to recognize concepts and select skills.
concepts
Concepts are matched bottom up, starting from percepts.
skills
Skill paths are matched top down, starting from intentions.
ICARUS Interleaves Execution and Problem Solving
Skill Hierarchy
Problem
Reactive
Execution
?
no
impasse?
Primitive Skills
yes
Problem
Solving
Executed plan
Interleaving Reactive Control and Problem Solving
Solve(G)
Push the goal literal G onto the empty goal stack GS.
On each cycle,
If the top goal G of the goal stack GS is satisfied,
Then pop GS.
Else if the goal stack GS does not exceed the depth limit,
Let S be the skill instances whose heads unify with G.
If any applicable skill paths start from an instance in S,
Then select one of these paths and execute it.
Else let M be the set of primitive skill instances that have not already failed in which G is an effect.
If the set M is nonempty,
Then select a skill instance Q from M.
Push the start condition C of Q onto goal stack GS.
Else if G is a complex concept with the unsatisfied subconcepts H and with satisfied subconcepts F,
Then if there is a subconcept I in H that has not yet failed,
Then push I onto the goal stack GS.
Else pop G from the goal stack GS and store information about failure with G's parent.
Else pop G from the goal stack GS.
Store information about failure with G's parent.
This is traditional means-ends analysis, with three exceptions: (1) conjunctive
goals must be defined concepts; (2) chaining occurs over both skills/operators
and concepts/axioms; and (3) selected skills are executed whenever applicable.
A Successful Problem-Solving Trace
initial state
(clear C)
(hand-empty)
(unst. C B)
(unstack C B)
(clear B)
(on C B)
goal
(unst. B A)
(ontable A T)
(holding C)
(putdown C T)
(clear A)
(unstack B A)
(hand-empty)
(on B A)
(holding B)
C
B
B
A
A
C
Architectures as Programming Languages
Cognitive architectures come with a programming language that:
 includes a syntax linked to its representational assumptions
 inputs long-term knowledge and initial short-term elements
 provides an interpreter that runs the specified program
 incorporates tracing facilities to inspect system behavior
Such programming languages ease construction and debugging
of knowledge-based systems.
For this reason, cognitive architectures support far more efficient
development of software for intelligent systems.
Programming in ICARUS
The programming language associated with ICARUS comes with:
 the syntax of teleoreactive logic programs
 the ability to load and parse such programs
 an interpreter for inference, execution, planning, and learning
 a trace package that displays system behavior over time
We have used this language to develop adaptive intelligent
agents in a variety of domains.
The Origin of Skill Hierarchies
ICARUS’ commitment to hierarchical organization raises a serious
question about the origin of its structures.
We want mechanisms which acquire these structures in ways that:
 are consistent with knowledge of human behavior;
 operate in an incremental and cumulative manner;
 satisfy constraints imposed by other ICARUS components.
This requires some source of experience from which to create
hierarchical structures.
ICARUS Learns Skills from Problem Solving
Skill Hierarchy
Problem
Reactive
Execution
?
no
impasse?
Primitive Skills
yes
Problem
Solving
Skill
Learning
Executed plan
Three Questions about Skill Learning
 What is the hierarchical structure of the network?
 The structure is determined by the subproblems that arise
in problem solving, which, because operator conditions
and goals are single literals, form a semilattice.
 What are the heads of the learned clauses/methods?
 The head of a learned clause is the goal literal that the
planner achieved for the subproblem that produced it.
 What are the conditions on the learned clauses/methods?
 If the subproblem involved skill chaining, they are the
conditions of the first subskill clause.
 If the subproblem involved concept chaining, they are the
subconcepts that held at the outset of the subproblem.
Constructing Skills from a Trace
(clear C)
skill chaining
1
(hand-empty)
(unst. C B)
(unstack C B)
(clear B)
(on C B)
(unst. B A)
(ontable A T)
(holding C)
(putdown C T)
(clear A)
(unstack B A)
(hand-empty)
(on B A)
(holding B)
C
B
B
A
A
C
Constructing Skills from a Trace
(clear C)
1
(hand-empty)
(unst. C B)
(unstack C B)
(clear B)
(on C B)
(unst. B A)
(clear A)
(unstack B A)
skill chaining
2
(ontable A T)
(holding C)
(putdown C T)
(hand-empty)
(on B A)
(holding B)
C
B
B
A
A
C
Constructing Skills from a Trace
concept chaining
(clear C)
3
1
(hand-empty)
(unst. C B)
(unstack C B)
(clear B)
(on C B)
(unst. B A)
(clear A)
(unstack B A)
2
(ontable A T)
(holding C)
(putdown C T)
(hand-empty)
(on B A)
(holding B)
C
B
B
A
A
C
Constructing Skills from a Trace
skill chaining
(clear C)
1
(hand-empty)
(unst. C B)
(unstack C B)
4
3
(clear B)
(on C B)
(unst. B A)
(clear A)
(unstack B A)
2
(ontable A T)
(holding C)
(putdown C T)
(hand-empty)
(on B A)
(holding B)
C
B
B
A
A
C
Learned Skills in the Blocks World
(clear (?C)
:percepts ((block ?D) (block ?C))
:start
(unstackable ?D ?C)
:skills
((unstack ?D ?C)))
(clear (?B)
:percepts ((block ?C) (block ?B))
:start
[(on ?C ?B) (hand-empty)]
:skills
((unstackable ?C ?B) (unstack ?C ?B)))
(unstackable (?C ?B)
:percepts ((block ?B) (block ?C))
:start
[(on ?C ?B) (hand-empty)]
:skills
((clear ?C) (hand-empty)))
(hand-empty ( )
:percepts ((block ?D) (table ?T1))
:start
(putdownable ?D ?T1)
:skills
((putdown ?D ?T1)))
Skill Clauses Learning for In-City Driving
(parked (?ME ?G1152)
:percepts ( (lane-line ?G1152) (self ?ME))
:start
( )
:skills
( (in-rightmost-lane ?ME ?G1152)
(stopped ?ME)) )
(in-rightmost-lane (?ME ?G1152)
:percepts ( (self ?ME) (lane-line ?G1152))
:start
( (last-lane ?G1152))
:skills
( (driving-in-segment ?ME ?G1101 ?G1152)) )
(driving-in-segment (?ME ?G1101 ?G1152)
:percepts ( (lane-line ?G1152) (segment ?G1101) (self ?ME))
:start
( (steering-wheel-straight ?ME))
:skills
( (in-lane ?ME ?G1152)
(centered-in-lane ?ME ?G1101 ?G1152)
(aligned-with-lane-in-segment ?ME ?G1101 ?G1152)
(steering-wheel-straight ?ME)) )
Learned Skill for Changing Lanes
We have trained the
ICARUS driving agent in
a cumulative manner.
First we present the
system with the task of
changing lanes using
only primitive skills.
The architecture calls
on problem solving to
handle the situation and
caches the solution in a
new composite skill.
This skill later lets the
agent change lanes in a
reactive fashion.
Learned Preparation for Turning
After the ICARUS agent
has mastered changing
lanes, we give it the
more complex task of
preparing for a turn.
Again the system calls
on problem solving, but
this time it can use the
skill already learned.
As before, the solution
is stored as a new skill,
this one calling on the
acquired skill for
changing lanes.
Learned Turning and Parking
Next we present the
task of turning a corner
and parking in the
rightmost lane.
As before, the system
calls the problem solver
to handle the situation
and stores additional
structures in memory.
The resulting skill
hierarchy can turn and
park from the initial
position using only
reactive control.
Intellectual Precursors
ICARUS’ design has been influenced by many previous efforts:
 earlier research on integrated cognitive architectures
 especially ACT, Soar, and Prodigy
 earlier frameworks for reactive control of agents
 research on belief-desire-intention (BDI) architectures
 planning/execution with hierarchical transition networks
 work on learning macro-operators and search-control rules
 previous work on cumulative structure learning
However, the framework combines and extends ideas from its
various predecessors in novel ways.
Directions for Future Research
Future work on ICARUS should introduce additional methods for:
 forward chaining and mental simulation of skills;
 learning expected utilities from skill execution histories;
 learning new conceptual structures in addition to skills;
 probabilistic encoding and matching of Boolean concepts;
 flexible recognition of skills executed by other agents;
 extension of short-term memory to store episodic traces.
Taken together, these features should make ICARUS a more general
and powerful cognitive architecture.
Contributions of ICARUS
ICARUS is a cognitive architecture for physical agents that:
 includes separate memories for concepts and skills;
 organizes both memories in a hierarchical fashion;
 modulates reactive execution with goal seeking;
 augments routine behavior with problem solving; and
 learns hierarchical skills in a cumulative manner.
These concerns distinguish ICARUS from other architectures, but
it makes commitments along the same design dimensions.
Some Other Cognitive Architectures
ACT
Soar
PRODIGY
EPIC
GIPS
RCS
CAPS
APEX
3T
CLARION
Dynamic
Memory
Society
of Mind
Concluding Remarks
We need more research on integrated intelligent systems that:
 are embedded within a unified cognitive architecture;
 incorporate modules that provide mutual constraints;
 demonstrate a wide range of intelligent behavior;
 are evaluated on multiple tasks in challenging testbeds.
If you do not yet use a cognitive architecture in your research on
intelligent agents, please consider it seriously.
For more information about the ICARUS architecture, see:
http://cll.stanford.edu/research/ongoing/icarus/
End of Presentation
Aspects of Cognitive Architectures
As traditionally defined and utilized, a cognitive architecture:
 specifies the infrastructure that holds constant over domains,
as opposed to knowledge, which varies.
 models behavior at the level of functional structures and
processes, not the knowledge or implementation levels.
 commits to representations and organizations of knowledge
and processes that operate on them.
 comes with a programming language for encoding knowledge
and constructing intelligent systems.
Early candidates were cast as production system architectures,
but alternatives have gradually expanded the known space.
In the Beginning . . .
AI
monobloc
The Big Bang Theory of AI
language
planning
perception

reasoning
action
learning
The Task of In-City Driving
Transfer Effects in the Blocks World
20 blocks
Learning Curves for In-City Driving
FreeCell Solitaire
FreeCell is a full-information card game that, in most cases, can
be solved by planning; it also has a highly recursive structure.
Transfer Effects in FreeCell
16 cards
An Approach to Hierarchy Learning
We have extended ICARUS to incorporate a module for means-ends
problem solving that:
 decomposes complex problems into subproblems;
 relies on heuristic search to find useful decompositions.
When ICARUS cannot execute a skill because its start condition is
unmet, this mechanism:
 chains backward off skills that would achieve the condition; or
 chains backward off definitions of the unsatisfied concept.
Traces of successful problem solving serve as the basis for new
hierarchical structures.
Evaluation of Intelligent Systems
Experimental studies of intelligent systems have lagged behind
ones for component methods because:
 they focus on more complex, multi-step behavior;
 they require more engineering to develop them;
 they rely on interaction among their components.
Together, these factors have slowed the widespread adoption of
experimental evaluation.
Repositories for Intelligent Systems
Public repositories are now common among the AI subfields,
and they offer clear advantages for research by:
 providing fast and cheap materials for experiments;
 supporting replication and standards for comparison.
However, they can also produce undesirable side effects by:
 focusing attention on a narrow class of problems;
 encouraging a ‘bake-off ’ mentality among researchers.
To support research on intelligent systems, we need testbeds
and environments designed with them in mind.
Concluding Remarks
We must also think about ways to overcome social obstacles to
pursuing this research agenda:
 conference tracks on integrated systems (e.g., AAAI)
 testbeds that evaluate general intelligence (e.g., GGP)
Both RoboCup and ICDL are in excellent positions to foster
more work along these lines.
I hope to see increased activity of this type at future meetings
of these conferences.