Copyright B. Williams
16.412J/6.834J, Fall 03
Introduction To Cognitive Robots
Prof. Brian Williams
Prof. Nick Roy
Wednesday, February 4
th
, 2004
Copyright B. Williams
16.412J/6.834J, Fall 03
Outline
? Course Objectives
? Student Introductions and Goals
? Example: Robots as Explorers
? Example: Human Robot Interaction
Copyright B. Williams
16.412J/6.834J, Fall 03
Course Objective 1
To understand the main types of cognitive robots and their
driving requirements:
? Robots That Navigate
– Hallway robots, Field robots, Underwater explorers, stunt air vehicles
? Engineering and “Immobile” Robots
– Intelligent spaces
– Robust space probes
? Cooperating Robots
– Cooperative Space/Air/Land/Underwater vehicles, distributed traffic networks,
smart dust.
Accomplished by:
? Case studies during lectures
? Supports course final project (Objective 4).
Copyright B. Williams
16.412J/6.834J, Fall 02
Immobile Robots in Space
Image courtesy of NASA.
Copyright B. Williams
16.412J/6.834J, Fall 03
Autonomous Systems use Models to
Anticipate or Detect Subtle Failures
600 700 800 900 1000 1100 1200 1300 1400
400
500
600
700
800
900
1000
1100
1200
time (minutes)
CO
2
con
cen
tr
ati
on
(
pp
m)
crew requests entry to
plant growth chamber
crew enters chamber
lighting fault
crew leaves
chamber
NASA Mars Habitat
Airlock
Plant Growth Chamber
Crew Chamb
e
r
CO
2
tank
lighting system
chamber
control
flow regulator 2
pulse injection valves
CO
2
flow regulator 1
Copyright B. Williams
16.412J/6.834J, Fall 03
courtesy NASA Ames
Image courtesy of NASA.
Copyright B. Williams
16.412J/6.834J, Fall 03
Course Objective 2
? To understand fundamental methods for creating
the major capabilities of cognitive robots.
Accomplished by:
?~ Lectures on core methods
?~ 3-4 problem sets to exercise basic understanding
of methods.
Plan
Execute
Monitor &
Diagnosis
Locate in
World
Navigate
Map
Copyright B. Williams
16.412J/6.834J, Fall 03
Topics On Cognitive Robot Capabilities
? Robots that Plan and Act in the World
– Robots that Deftly Navigate
– Planning and Executing Complex Missions
? Robots that Are State-Aware
– Robots that Find Their Way In The World
– Robots that Deduce Their Internal State
? Robots that Preplan For An Uncertain
Future
– Theoretic Planning in a Hidden World
– State and Fault Aware Systems
Copyright B. Williams
16.412J/6.834J, Fall 03
Course Objective 3
? To dive into the recent literature, and collectively
synthesize, clearly explain and evaluate the state
of the art in intelligent embedded systems.
Accomplished by:
?Group lectures on advance topic
?40 minute or 80 minute lectures
?tutorial article on ~2 methods, to support lectures.
?Groups of size ~2.
Copyright B. Williams
16.412J/6.834J, Fall 03
Course Objective 4
? To apply one or more core reasoning methods to create a
simple agent that is driven by Goals or Rewards
Accomplished by:
? Final project during last third of course
? Implement and demonstrate one or more reasoning methods in a
simple cognitive robot scenario (simulated or hardware).
? Final project report.
? Short project demonstration.
Plan
Execute
Monitor &
Diagnosis
Locate in
World
Navigate
Map
Copyright B. Williams
16.412J/6.834J, Fall 03
Outline
? Course Objectives
? Student Introductions and Goals
? Example: Robots as Explorers
? Example: Human Robot Interaction
Copyright B. Williams
16.412J/6.834J, Fall 03
Outline
? Course Objectives
? Student Introductions and Goals
? Example: Robots as Explorers
? Example: Human Robot Interaction
Robotic Space Explorers:
To Boldly Go Where No AI System
Has Gone Before
A Story of Survival
16.412J/6.834J
September 19, 2001
Copyright B. Williams
16.412J/6.834J, Fall 03
Readings and Assignment
Readings:
? Remote Agent: to Boldy Go Where No AI System Has
Gone Before,
N.Muscettola, P. Nayak, B. Pell and B. Williams, Artificial
Intelligence 103 (1998) 5-47.
Copyright B. Williams
16.412J/6.834J, Fall 03
Outline
? Motivation
? Model-based autonomous systems
? Remote Agent Example
Cryobot & Hydrobot
courtesy JPL
Europa
``Our vision in NASA is to open the Space Frontier . . . We must
establish a virtual presence, in space, on planets, in aircraft and
spacecraft.’’
- Daniel S. Goldin, NASA Administrator, May 29, 1996
courtesy JPL
Distributed Spacecraft Interferometers
search for Earth-like Planets Around Other Stars
Copyright B. Williams
16.412J/6.834J, Fall 03
Miscommanded:
? Mars Climate Orbiter
? Clementine
courtesy of JPL
Spacecraft should watch out for their own survival.
Cassini Maps Titan
courtesy JPL
? 7 year cruise
? ~ 150 - 300
ground operators
?~ 1 billion $
? 7 years to build
A Capable Robotic Explorer: Cassini
?150 million $
?2 year build
? 0 ground ops
Faster, Better, Cheaper
courtesy JPL
Mars Pathfinder and Sojourner
Copyright B. Williams
16.412J/6.834J, Fall 03
Four launches in 7 months
Mars Climate Orbiter: 12/11/98
Mars Polar Lander: 1/3/99
Stardust: 2/7/99
QuickSCAT: 6/19/98
courtesy of JPL
Copyright B. Williams
16.412J/6.834J, Fall 03
Objective: Support programmers with
embedded languages that avoid these
mistakes, by reasoning about hidden
state automatically.
Leading Diagnosis:
? Legs deployed during descent.
? Noise spike on leg sensors
latched by software monitors.
? Laser altimeter registers 40m.
? Begins polling leg monitors to
determine touch down.
? Latched noise spike read as
touchdown.
? Engine shutdown at ~40m.
Reactive Model-based
Programming Language (RMPL)
Mars Polar Lander Failure
Programmers often make
commonsense mistakes when
reasoning about hidden state.
Image courtesy of JPL.
Copyright B. Williams
16.412J/6.834J, Fall 03
Traditional spacecraft commanding
GS,SITURN,490UA,BOTH,96-355/03:42:00.000;
CMD,7GYON, 490UA412A4A,BOTH, 96-355/03:47:00:000, ON;
CMD,7MODE, 490UA412A4B,BOTH, 96-355/03:47:02:000, INT;
CMD,6SVPM, 490UA412A6A,BOTH, 96-355/03:48:30:000, 2;
CMD,7ALRT, 490UA412A4C,BOTH, 96-355/03:50:32:000, 6;
CMD,7SAFE, 490UA412A4D,BOTH, 96-355/03:52:00:000, UNSTOW;
CMD,6ASSAN,490UA412A6B,BOTH, 96-355/03:56:08:000, GV,153,IMM,231,
GV,153;
CMD,7VECT, 490UA412A4E,BOTH, 96-355/03:56:10.000, 0,191.5,6.5,
0.0,0.0,0.0,
96-350/
00:00:00.000,MVR;
SEB,SCTEST,490UA412A23A,BOTH, 96-355/03:56:12.000, SYS1,NPERR;
CMD,7TURN, 490UA412A4F,BOTH, 96-355/03:56:14.000, 1,MVR;
MISC,NOTE, 490UA412A99A,, 96-355/04:00:00.000, ,START OF TURN;,
CMD,7STAR, 490UA412A406A4A,BOTH 96-355/04:00:02.000, 7,1701,
278.813999,38.74;
CMD,7STAR, 490UA412A406A4B,BOTH,96-355/04:00:04.000, 8,350,120.455999,
-39.8612;
CMD,7STAR, 490UA412A406A4C,BOTH,96-355/04:00:06.000, 9,875,114.162,
5.341;
CMD,7STAR, 490UA412A406A4D,BOTH,96-355/04:00:08.000, 10,159,27.239,
89.028999;
CMD,7STAR, 490UA412A406A4E,BOTH,96-355/04:00:10.000, 11,0,0.0,0.0;
CMD,7STAR, 490UA412A406A4F,BOTH,96-355/04:00:12.000, 21,0,0.0,0.0;
What Makes this Difficult:
Cassini Case Study
courtesy JPL
Reasoning through interactions is complex
Houston, We have a problem ...
courtesy of NASA
? Quintuple fault occurs
(three shorts, tank-line and
pressure jacket burst, panel
flies off).
? Mattingly works in ground
simulator to identify new
sequence handling severe
power limitations.
? Mattingly identifies novel
reconfiguration, exploiting
LEM batteries for power.
? Swaggert & Lovell follow
novel procedure to repair
Apollo 13 lithium hydroxide
unit.
Survival can require replanning
the complete mission on the fly.
Helium tank
Fuel tankOxidizer tank
Main
Engines
Flow
1
= zero
Pressure
1
= nominal
Pressure
2
= nominal
Acceleration = zero
Copyright B. Williams
16.412J/6.834J, Fall 03
Challenge: Thinking Through Interactions
Programmers must reason through system-wide
interactions to generate codes for:
? command confirmation
? goal tracking
? detecting anomalies
? isolating faults
? diagnosing causes
? hardware reconfig
? fault recovery
? safing
? fault avoidance
? control coordination
Equally problematic at mission operations level
Copyright B. Williams
16.412J/6.834J, Fall 03
Outline
? Motivation
? Model-based autonomous systems
? Remote Agent Example
Copyright B. Williams
16.412J/6.834J, Fall 03
Course Objective 2
? To understand fundamental methods for creating
the major components of intelligent embedded
systems.
Plan
Execute
Monitor &
Diagnosis
Copyright B. Williams
16.412J/6.834J, Fall 03
Programmers generate breadth of functions from
commonsense models in light of mission goals.
? Model-based Programming
? Program by specifying commonsense, compositional
declarative models.
? Model-based Planning, Execution and Monitoring
? Provide services that reason through each type of
system interaction from models.
? on the fly reasoning requires significant search &
deduction within the reactive control loop.
Model-based Autonomy
courtesy of NASA
? Quintuple fault occurs
(three shorts, tank-line and
pressure jacket burst, panel
flies off).
? Mattingly works in ground
simulator to identify new
sequence handling severe
power limitations.
? Mattingly identifies novel
reconfiguration, exploiting
LEM batteries for power.
? Swaggert & Lovell work on
Apollo 13 emergency rig
lithium hydroxide unit.
Styles of Thinking Through Interactions
? Quintuple fault occurs
(three shorts, tank-line and
pressure jacket burst, panel
flies off).
? Mattingly works in ground
simulator to identify new
sequence handling severe
power limitations.
? Mattingly identifies novel
reconfiguration, exploiting
LEM batteries for power.
? Swaggert & Lovell work on
Apollo 13 emergency rig
lithium hydroxide unit.
? Multiple fault diagnosis of
unexperienced failures.
? Mission planning and
scheduling
? Hardware reconfiguration
? Scripted execution
Styles of Thinking Through Interactions
Copyright B. Williams
16.412J/6.834J, Fall 03
Example of a Model-based Robot:
? Goal-directed
? First time correct
? projective
? reactive
? Commonsense models
? Heavily deductive
Scripts
component models
Goals
Diagnosis
& Repair
Mission
Manager
Executive
Planner/
Scheduler
Remote Agent
Mission-level
actions &
resources
Copyright B. Williams
16.412J/6.834J, Fall 03
Conventional Wisdom: Reservations
about Intelligent Embedded Systems
? “[For reactive systems] proving theorems is out of the question”
[Agre & Chapman 87]
Copyright B. Williams
16.412J/6.834J, Fall 03
Many problems aren’t so hard
Copyright B. Williams
16.412J/6.834J, Fall 03
Generate
Non-conflicting
Successor
Generate
Non-conflicting
Successor
Candidates with decreasing
likelihood or value
SAT
SAT
Generalization
of Conflicts
Generalization
of Conflicts
Developed RISC-like,
deductive kernel (OPSAT)
How can general deduction achieve
reactive time scales?
Solutions
When you have eliminated
the impossible, whatever
remains, however improbable
[costly], must be the truth.
- Sherlock Holmes.
The Sign of the Four.
((Image removed due to copyright considerations.)
Copyright B. Williams
16.412J/6.834J, Fall 03
Transition Systems +
Constraints + Probabilities
Closed
Valve
Open
Stuck
open
Stuck
closed
Open Close
0. 01
0. 01
0.01
0.01
inflow = outflow = 0
Can model-based agents perform many different types
of reasoning from a common model?
VDECU Op_State
On
Engine Op State
Burn_Ignition
Burn
Burn_Termination
Shut_down
Off
Early_Prep Wait
Late_Prep
Copyright B. Williams
16.412J/6.834J, Fall 03
Outline
? Motivation
? Model-based autonomous systems
? Remote Agent Example
Copyright B. Williams
16.412J/6.834J, Fall 03
Real-Time
Execution
Real-Time
Execution
Flight
H/W
Flight
Fault
Monitors
Fault
Monitors
Planning Experts
(incl. Navigation)
Planning Experts
(incl. Navigation)
Remote Agent Architecture
Ground
System
Ground
System
RAX Manager
Diagnosis
& Repair
Mission
Manager
Executive
Planner/
Scheduler
Remote Agent
RAX_START
Copyright B. Williams
16.412J/6.834J, Fall 03
Real-Time
Execution
Real-Time
Execution
Flight
H/W
Flight
Fault
Monitors
Fault
Monitors
Planning Experts
(incl. Navigation)
Planning Experts
(incl. Navigation)
Ground
System
Ground
System
RAX Manager
Diagnosis
& Repair
Mission
Manager
Executive
Planner/
Scheduler
Remote Agent
RAX_START
Executive requests plan
Copyright B. Williams
16.412J/6.834J, Fall 03
Real-Time
Execution
Real-Time
Execution
Flight
H/W
Flight
Fault
Monitors
Fault
Monitors
Planning Experts
(incl. Navigation)
Planning Experts
(incl. Navigation)
Ground
System
Ground
System
RAX Manager
Diagnosis
& Repair
Mission
Manager
Executive
Planner/
Scheduler
Remote Agent
RAX_START
Remote Agent Architecture
Copyright B. Williams
16.412J/6.834J, Fall 03
Real-Time
Execution
Real-Time
Execution
Flight
H/W
Flight
Fault
Monitors
Fault
Monitors
Planning Experts
(incl. Navigation)
Planning Experts
(incl. Navigation)
Ground
System
Ground
System
RAX Manager
Diagnosis
& Repair
Mission
Manager
Executive
Planner/
Scheduler
Remote Agent
RAX_START
Mission manager establishes goals,
planner generates plan
Copyright B. Williams
16.412J/6.834J, Fall 03
Real-Time
Execution
Real-Time
Execution
Flight
H/W
Flight
Fault
Monitors
Fault
Monitors
Planning Experts
(incl. Navigation)
Planning Experts
(incl. Navigation)
Ground
System
Ground
System
RAX Manager
Diagnosis
& Repair
Mission
Manager
Executive
Planner/
Scheduler
Remote Agent
RAX_START
Executive executes plan
Copyright B. Williams
16.412J/6.834J, Fall 03
Real-Time
Execution
Real-Time
Execution
Flight
H/W
Flight
Fault
Monitors
Fault
Monitors
Planning Experts
(incl. Navigation)
Planning Experts
(incl. Navigation)
Ground
System
Ground
System
RAX Manager
Diagnosis
& Repair
Mission
Manager
Executive
Planner/
Scheduler
Remote Agent
RAX_START
Diagnosis system monitors and repairs
Walk Through of
Cassini Saturn Orbital Insertion
Copyright B. Williams
16.412J/6.834J, Fall 03
Real-Time
Execution
Real-Time
Execution
Flight
H/W
Flight
Fault
Monitors
Fault
Monitors
Planning Experts
(incl. Navigation)
Planning Experts
(incl. Navigation)
Ground
System
Ground
System
RAX Manager
Diagnosis
& Repair
Mission
Manager
Executive
Planner/
Scheduler
Remote Agent
RAX_START
Plan for Next Time Horizon
Image courtesy of JPL.
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Engine
Power
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(a)
Engine OffOff
Delta_V(direction=b, magnitude=200)
Power
Mission Manager Sets Goals
over Horizon
Copyright B. Williams
16.412J/6.834J, Fall 03
Planner Repeatedly Applies
Library of Operational Constraints
Thrust
Goals
Engine Thrust (b, 200)
Delta_V(direction=b, magnitude=200)
contains
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(b)
Engine Thrust (b, 200) Off
Delta_V(direction=b, magnitude=200)
Power
Warm Up
meets
met_by
contained_by
contained_by
equals
Planner Repeatedly Applies
Library of Operational Constraints
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(a)
Engine OffOff
Delta_V(direction=b, magnitude=200)
Power
Planner Starts
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(a)
Engine Thrust (b, 200) OffOff
Delta_V(direction=b, magnitude=200)
Power
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(a)
Engine
Delta_V(direction=b, magnitude=200)
Power
Thrust (b, 200) OffOff
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(a)
Engine
Delta_V(direction=b, magnitude=200)
Power
Thrust (b, 200) OffOff
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(a)
Engine
Delta_V(direction=b, magnitude=200)
Power
OffOff Thrust (b, 200)
Copyright B. Williams
16.412J/6.834J, Fall 03
Point(b)
Thrust
Goals
Attitude
Point(a)
Engine Off
Delta_V(direction=b, magnitude=200)
Power
Thrust (b, 200) Off
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(a) Point(b)
Engine Thrust (b, 200)Off
Delta_V(direction=b, magnitude=200)
Power
Off
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(a) Point(b)
Engine Thrust (b, 200)Warm Up OffOff
Delta_V(direction=b, magnitude=200)
Power
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(a) Point(b) Turn(b,a)
Engine Thrust (b, 200)Warm Up OffOff
Delta_V(direction=b, magnitude=200)
Power
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude
Point(a) Point(b) Turn(b,a)
Engine Thrust (b, 200)Warm Up OffOff
Delta_V(direction=b, magnitude=200)
Power
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude Turn(a,b)
Point(a) Point(b) Turn(b,a)
Engine Thrust (b, 200) OffOff
Delta_V(direction=b, magnitude=200)
Power
Warm Up
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude Turn(a,b)
Point(a) Point(b) Turn(b,a)
Engine Thrust (b, 200) OffOff
Delta_V(direction=b, magnitude=200)
Power
Warm Up
Copyright B. Williams
16.412J/6.834J, Fall 03
Thrust
Goals
Attitude Turn(a,b)
Point(a) Point(b) Turn(b,a)
Engine Thrust (b, 200) OffOff
Delta_V(direction=b, magnitude=200)
Power
Warm Up
Plan Completed!
Copyright B. Williams
16.412J/6.834J, Fall 03
Plans Allow Temporal Flexibility
Through Least Committment
[1035, 1035]
[130,170]
<0, 0>
[0, 300]
[0, + ∞ ]
[0, + ∞]
[0, 0]
Copyright B. Williams
16.412J/6.834J, Fall 03
Real-Time
Execution
Real-Time
Execution
Flight
H/W
Flight
Fault
Monitors
Fault
Monitors
Planning Experts
(incl. Navigation)
Planning Experts
(incl. Navigation)
Ground
System
Ground
System
RAX Manager
Diagnosis
& Repair
Mission
Manager
Scripted
Executive
Planner/
Scheduler
Remote Agent
RAX_START
The executive dynamically
schedules and dispatches tasks
Copyright B. Williams
16.412J/6.834J, Fall 03
Executing Temporal Plans
[130,170]]
<0, 0>
[0, 300]
[0, + ]
[0, + ]
[0, 0]
? Propagate temporal constraints
? Select enabled events
? Terminate preceding activities
? Run next activities
Copyright B. Williams
16.412J/6.834J, Fall 03
Propagating Timing Constraints
Can Be Costly
EXECUTIVE
CONTROLLED SYSTEM
Copyright B. Williams
16.412J/6.834J, Fall 03
EXECUTIVE
CONTROLLED SYSTEM
Propagating Timing Constraints
Can Be Costly
Copyright B. Williams
16.412J/6.834J, Fall 03
EXECUTIVE
CONTROLLED SYSTEM
Propagating Timing Constraints
Can Be Costly
Copyright B. Williams
16.412J/6.834J, Fall 03
EXECUTIVE
CONTROLLED SYSTEM
Propagating Timing Constraints
Can Be Costly
Copyright B. Williams
16.412J/6.834J, Fall 03
EXECUTIVE
CONTROLLED SYSTEM
Propagating Timing Constraints
Can Be Costly
Copyright B. Williams
16.412J/6.834J, Fall 03
EXECUTIVE
CONTROLLED SYSTEM
Propagating Timing Constraints
Can Be Costly
Copyright B. Williams
16.412J/6.834J, Fall 03
EXECUTIVE
CONTROLLED SYSTEM
Solution: Compile Temporal
Constraints to an Efficient Network
Copyright B. Williams
16.412J/6.834J, Fall 03
EXECUTIVE
CONTROLLED SYSTEM
Solution: Compile Temporal
Constraints to an Efficient Network
Copyright B. Williams
16.412J/6.834J, Fall 03
EXECUTIVE
CONTROLLED SYSTEM
Solution: Compile Temporal
Constraints to an Efficient Network
Copyright B. Williams
16.412J/6.834J, Fall 03
Real-Time
Execution
Real-Time
Execution
Flight
H/W
Flight
Fault
Monitors
Fault
Monitors
Planning Experts
(incl. Navigation)
Planning Experts
(incl. Navigation)
Ground
System
Ground
System
RAX Manager
Diagnosis
& Repair
Mission
Manager
Scripted
Executive
Planner/
Scheduler
Remote Agent
RAX_START
Execution and Fault Recovery involves
monitoring and commanding hidden states
Copyright B. Williams
16.412J/6.834J, Fall 03
Programmers and operators must reason through
system-wide interactions to generate codes for:
Reconfiguring ModesEstimating Modes
? monitoring
? tracking goals
? confirming commands
? detecting anomalies
? diagnosing faults
? reconfiguring hardware
? coordinating control
policies
? recovering from faults
? avoiding failures
Model-based Execution of Activities
Copyright B. Williams
16.412J/6.834J, Fall 03
Model-based Execution as
Stochastic Optimal Control
Controller
Plant
mode
Estimation
mode
reconfiguration
s’(t)
μ(t)
f
s(t)
g
o(t)
Model
Livingstone
Goals
Copyright B. Williams
16.412J/6.834J, Fall 03
Closed
Open
Stuck
open
Stuck
closed
Open
Close
Cost 5
Prob .9
Models
? modes engage physical processes
? encoded as finite domain constraints
? probabilistic automata for dynamics
? Concurrency to model multiple processes
Vlv = closed =>
Outflow = 0;
vlv=open =>
Outflow = M
z
+
(inflow);
vlv=stuck open =>
Outflow = M
z
+
(inflow);
vlv=stuck closed=>
Outflow = 0;
Fuel tankOxidizer tank
Copyright B. Williams
16.412J/6.834J, Fall 03
Reconfiguring for a Failed Engine
Copyright B. Williams
16.412J/6.834J, Fall 03
Reconfiguring for a Failed Engine
Open four
valves
Fuel tankOxidizer tank
Copyright B. Williams
16.412J/6.834J, Fall 03
Reconfiguring for a Failed Engine
Valve fails
stuck closed
Open four
valves
Fuel tankOxidizer tank
Copyright B. Williams
16.412J/6.834J, Fall 03
Reconfiguring for a Failed Engine
Fire backup
engine
Valve fails
stuck closed
Open four
valves
Fuel tankOxidizer tank
Copyright B. Williams
16.412J/6.834J, Fall 03
Possible Behaviors
Visualized by a Trellis Diagram
S T
X
0
X
1
X
N-1
X
N
?Assigns a value to each
variable.
?Consistent with all state
constraints.
?A set of concurrent
transitions, one per automata.
?Previous & Next states
consistent with source &
target of transitions
Copyright B. Williams
16.412J/6.834J, Fall 03
Model-based Execution as
Stochastic Optimal Control
Controller
Plant
mode
Estimation
mode
reconfiguration
s’(t)
μ(t)
f
s(t)
g
o(t)
Model
Livingstone
Goals
Fire backup
engine
Valve fails
stuck closed
S T
X0 X1 XN-1 XN
S T
X0 X1 XN-1 XN
least cost reachable
goal state
First ActionCurrent Belief State
Control Sequencer
Deductive Controller
System Model
Commands
Observations
Control Program
Plant
Titan Model-based ExecutiveRMPL Model-based Program
State goalsState estimates
Control Sequencer:
Generates goal states
conditioned on state estimates
Mode
Estimation:
Tracks likely
States
Mode
Reconfiguration:
Tracks least-cost
state goals
z Executes concurrently
z Preempts
z Asserts and queries states
z Chooses based on reward
Fire backup
engine
Valve fails
stuck closed
S T
X0 X1 XN-1 XN
S T
X0 X1 XN-1 XN
least cost reachable
goal state
First ActionCurrent Belief State
Copyright B. Williams
16.412J/6.834J, Fall 03
Mode Estimation and Diagnosis
Observe
“no thrust”
Find most likely reachable states
consistent with observations.
Copyright B. Williams
16.412J/6.834J, Fall 03
Mode Reconfiguration and Repair
Goal: Achieve Thrust
Copyright B. Williams
16.412J/6.834J, Fall 03
Goal: Achieve Thrust
Mode Reconfiguration and Repair
Copyright B. Williams
16.412J/6.834J, Fall 03
Goal: Achieve Thrust
Mode Reconfiguration and Repair
courtesy JPL
Ames-JPL NewMaap Demonstration:
New Millennium Advanced Autonomy Prototype
July - November, 1995
Copyright B. Williams
16.412J/6.834J, Fall 03
Remote Agent on Deep Space 1
Started: January 1996
Launch: Fall 1998
Image courtesy of JPL.
Image courtesy of JPL.
Copyright B. Williams
16.412J/6.834J, Fall 03
Remote Agent Experiment
May 17-18th experiment
? Generate plan for course correction and thrust
? Diagnose camera as stuck on
– Power constraints violated, abort current plan and replan
? Perform optical navigation
? Perform ion propulsion thrust
May 21th experiment.
? Diagnose faulty device and
– Repair by issuing reset.
? Diagnose switch sensor failure.
– Determine harmless, and continue plan.
? Diagnose thruster stuck closed and
– Repair by switching to alternate method of thrusting.
? Back to back planning
See rax.arc.nasa.gov
Copyright B. Williams
16.412J/6.834J, Fall 03
Mars Exploration Rovers – Jan. 2004
Mission Objectives:
? Learn about ancient water and climate on Mars.
? For each rover, analyze a total of 6-12 targets
– Targets = natural rocks, abraded rocks, and soil
? Drive 200-1000 meters per rover
? Take 1-3 panoramas both with Pancam and mini-TES
? Take 5-15 daytime and 1-3 nightime sky observations with
mini-TES
Mini-TES
Pancam
Navcam
Rock Abrasion Tool
Microscopic Imager
Mossbauer spectrometer
APXS
Image courtesy of JPL.
Copyright B. Williams
16.412J/6.834J, Fall 03
Mars Exploration Rover
Surface Operations Scenario
Target
Day 4
During the Day
Science Activities
Day 1
Long-Distance Traverse
(<20-50 meters)
Day 2
Initial Position;
Followed by
“Close Approach”
During the Day
Autonomous On-
Board Navigation
Changes, as needed
Day 2 Traverse Estimated
Error Circle
Day 3
Science Prep
(if Required)
Day 2 Traverse
Estimated Error Circle
Activity Name
Durati
on
101121314151617181920212230123456789
101121314151617181920212230123456789
DTE
4.50
0.75
DTE period DFE
Night Time Rover Operations
16.97
Night Time Rover OperationsSleep
Wakeup
Pre-Comm Session Sequence Plan Review
Current Sol Sequence Plan Review
1.50
1.50
Current Sol Sequence Plan Review
Prior Sol Sequence Plan Review
2.00
Prior Sol Sequence Plan Review
Real-TIme Monitoring
4.50
0.75
Real-TIme Monitoring Real-TIme Monitoring
Downlink Product Generation...
2.75
Downlink Product Generation
Tactical Science
Assessment/Observation Planning
5.00
Tactical Science Assessment/Observation Planning
Science DL Assessment Meeting
1.00
Science DL Assessment Meeting
Payload DL/UL Handoffs
0.50
Payload DL/UL Handoffs
Tactical End-of-Sol Engr. Assessment &
Planning
5.50
Tactical End-of-Sol Engr. Assessment & Planning
DL/UL Handover Meeting
0.50
DL/UL Handover Meeting
Skeleton Activity Plan Update
2.50
Skeleton Activity Plan Update
SOWG Meeting
2.00
SOWG Meeting
Uplink Kickoff Meeting
0.25
Uplink Kickoff Meeting
Activity Plan Integration & Validation
1.75
Activity Plan Integration & Validation
Activity Plan Approval Meeting
0.50
Activity Plan Approval Meeting
Build & Validate Sequences
2.25
Build & Validate Sequences
UL1/UL2 Handover
1.00
UL1/UL2 Handover
Complete/Rework Sequences
2.50
Complete/Rework Sequences
Margin 1
0.75
Margin 1
Command & Radiation Approval
0.50
Command & Radiation Ap
Margin 2
1.25
Margin 2
Radiation
0.50
Radiation
MCT Team
7.00
4.00
One day in the life of a Mars rover
Courtesy: Jim Erickson
Downlink Assessment Science Planning Sequence Build/Validation Uplink
Copyright B. Williams
16.412J/6.834J, Fall 03
EUROPA
Automated
Planning System
EUROPA
Automated
Planning System
Science
Navigation
Engineering
Resource
Constraints
DSN/Telcom
Flight Rules
Science Team
Sequence
Build
MAPGEN: Automated
Science Planning for MER
Planning Lead: Kanna Rajan (ARC)
Copyright B. Williams
16.412J/6.834J, Fall 03
One Project Challenge
? What would it be like to operate MER if it
was fully autonomous?
Course project:
? Demonstrate an autonomous MER mission
in simulation, and in the MIT rover testbed.
Copyright B. Williams
16.412J/6.834J, Fall 03
Next Challenge: Mars Smart Lander (2009)
Mission Duration: 1000 days
Total Traverse: 3000-69000 meters
Meters/Day: 230-450
Science Mission: 7 instruments, sub-surface science
package (drill, radar), in-situ sample “lab”
Technology Demonstration:
(2005).
Image courtesy of JPL.