Lecture 3:
Immobile Robots
and Space Explorers
Prof. Brian Williams
Rm 33-418
Wednesday, September 11th, 2002
Copyright B. Williams
16.412J/6.834J, Fall 02
Course Objective 1
To understand the main types of intelligent embedded systems
and their driving requirements:
• Agile Robots
– Hallway robots, Field robots, Underwater explorers, stunt air vehicles
• “Immobile” Robots
– Intelligent spaces
– Robust space probes
• Cooperating Agents
– 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
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.
• Immobile Robots: AI in the New Millennium,
B. Williams and P. Nayak , AI Magazine, Fall (1996).
Problem Set 1:
• Distributed in Class:
Wednesday, September 11th, 2002
• Due in Class:
Wednesday, September 18th, 2002
Copyright B. Williams
16.412J/6.834J, Fall 02
Immobile Robots
Xerox PARC Ubiquitous Computing Project:
• Shift computation from the desktop, in to the walls and everyday devices.
Responsive Environment Project (circa 1992):
• Create building environments that anticipate and adapt to user needs.
• Models self and its occupants
• Learns physics of building
• Learns models of user activity (e.g., office occupancy)
• Acts in order to anticipate and meet needs.
• Sets energy goals based on user’s anticipated needs.
• Regulates by distributed auction.
•Synthesizes distributed, optimal controllers to save energy.
Copyright B. Williams
16.412J/6.834J, Fall 02
Immobile Robots in Space
Copyright B. Williams
16.412J/6.834J, Fall 02
courtesy NASA
The Russian Mir Failure
Copyright
Williams
courtesy B.
NASA
Ames
16.412J/6.834J, Fall 02
MIT
Spheres
flies
in
Intl
Space
station
2003
Copyright B. Williams
courtesy Prof. Dave Miller, MIT Space Systems
Laboratory
16.412J/6.834J,
Fall 02
Autonomous Systems use Models to
Anticipate or Detect Subtle Failures
concentration
(ppm)
1200
NASA Mars Habitat
lightingsystem
A
irlock
CO2
pulseinjectionvalves
crew requests entry to
plan t grow th ch am ber
1100
1000
900
800
crew en ters cham b er
700
crew leav es
cham ber
lightin g fault
600
500
PlantG
row
thC
ham
ber
CrewChamber
400
flowregulator1
C
O
2
C
O
2
tank
600
700
800
900
1000
1100
time (minutes)
1200
1300
1400
flowregulator2
cham
ber
control
Copyright B. Williams
16.412J/6.834J, Fall 02
Robotic Space Explorers:
To Boldly Go Where No AI System
Has Gone Before
A Story of Survival
16.412J/6.834J
September 19, 2001
Outline
• Motivation
• Model-based autonomous systems
• Remote Agent Example
Copyright B. Williams
16.412J/6.834J, Fall 02
``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
Cryobot & Hydrobot
Europa
courtesy JPL
courtesy JPL
Distributed Spacecraft Interferometers
search for Earth-like Planets Around Other Stars
A Capable Robotic Explorer: Cassini
• 7 year cruise
Faster, Better, Cheaper
• ~ 150 - 300
ground operators
•~ 1 billion $
• 7 years to build
•150 million $
•2 year build
• 0 ground ops
Cassini Maps Titan
courtesy JPL
courtesy JPL
Mars Pathfinder and Sojourner
Four launches in 7 months
Mars Climate Orbiter: 12/11/98
Stardust: 2/7/99
Copyright B. Williams
Mars Polar Lander: 1/3/99
QuickSCAT: 6/19/98
courtesy of JPL
16.412J/6.834J, Fall 02
Miscommanded:
• Mars Climate Orbiter
• Clementine
courtesy of JPL
Spacecraft should watch out for their own survival.
Copyright B. Williams
16.412J/6.834J, Fall 02
Mars Polar Lander Failure
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.
Programmers often make
commonsense mistakes when
reasoning about hidden state.
Copyright B. Williams
Objective: Support programmers with
embedded languages that avoid these
mistakes, by reasoning about hidden
state automatically.
Reactive Model-based
Programming Language (RMPL)
16.412J/6.834J, Fall 02
Traditional spacecraft commanding
GS,SITURN,490UA,BOTH,96-355/03:42:00.000;
CMD,7GYON,
CMD,7MODE,
CMD,6SVPM,
CMD,7ALRT,
CMD,7SAFE,
CMD,6ASSAN,
490UA412A4A,BOTH,
490UA412A4B,BOTH,
490UA412A6A,BOTH,
490UA412A4C,BOTH,
490UA412A4D,BOTH,
490UA412A6B,BOTH,
96-355/03:47:00:000,
96-355/03:47:02:000,
96-355/03:48:30:000,
96-355/03:50:32:000,
96-355/03:52:00:000,
96-355/03:56:08:000,
CMD,7VECT,
490UA412A4E,BOTH,
96-355/03:56:10.000,
SEB,SCTEST,
CMD,7TURN,
MISC,NOTE,
CMD,7STAR,
490UA412A23A,BOTH,
490UA412A4F,BOTH,
490UA412A99A,,
490UA412A406A4A,BOTH
96-355/03:56:12.000,
96-355/03:56:14.000,
96-355/04:00:00.000,
96-355/04:00:02.000,
CMD,7STAR,
490UA412A406A4B,BOTH, 96-355/04:00:04.000,
CMD,7STAR,
490UA412A406A4C,BOTH, 96-355/04:00:06.000,
CMD,7STAR,
490UA412A406A4D,BOTH, 96-355/04:00:08.000,
CMD,7STAR,
CMD,7STAR,
490UA412A406A4E,BOTH, 96-355/04:00:10.000,
490UA412A406A4F,BOTH, 96-355/04:00:12.000,
Copyright B. Williams
ON;
INT;
2;
6;
UNSTOW;
GV,153,IMM,231,
GV,153;
0,191.5,6.5,
0.0,0.0,0.0,
96-350/
00:00:00.000,MVR;
SYS1,NPERR;
1,MVR;
,START OF TURN;,
7,1701,
278.813999,38.74;
8,350,120.455999,
-39.8612;
9,875,114.162,
5.341;
10,159,27.239,
89.028999;
11,0,0.0,0.0;
21,0,0.0,0.0;
16.412J/6.834J, Fall 02
What Makes this Difficult:
Cassini Case Study
courtesy JPL
Reasoning through interactions is complex
Reconfiguring for a Failed Engine
Oxidizer tank
Copyright B. Williams
Fuel tank
16.412J/6.834J, Fall 02
Reconfiguring for a Failed Engine
Oxidizer tank
Fuel tank
Open four
valves
Copyright B. Williams
16.412J/6.834J, Fall 02
Reconfiguring for a Failed Engine
Oxidizer tank
Fuel tank
Open four
valves
Valve fails
stuck closed
Copyright B. Williams
16.412J/6.834J, Fall 02
Reconfiguring for a Failed Engine
Oxidizer tank
Fuel tank
Open four
valves
Valve fails
stuck closed
Fire backup
engine
Copyright B. Williams
16.412J/6.834J, Fall 02
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 02
Houston, We have a problem ...
• 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
courtesy of NASA
novel procedure to repair
Apollo 13 lithium hydroxide
Survival can require replanning unit.
the complete mission on the fly.
Outline
• Motivation
• Model-based autonomous systems
• Remote Agent Example
Copyright B. Williams
16.412J/6.834J, Fall 02
Course Objective 2
• To understand fundamental methods for creating
the major components of intelligent embedded
systems.
Plan
Monitor &
Diagnosis
Copyright B. Williams
Execute
16.412J/6.834J, Fall 02
Model-based Autonomy
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.
Copyright B. Williams
16.412J/6.834J, Fall 02
Styles of Thinking Through Interactions
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
• Multiple fault diagnosis of
unexperienced failures.
• Mission planning and
scheduling
• Hardware reconfiguration
• Scripted execution
• 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.
Example of a Model-based Agent:
Goals
• Goal-directed
• First time correct
• projective
• reactive
• Commonsense models
• Heavily deductive
Copyright B. Williams
Scripts
Remote Agent
Mission
Manager
Planner/
Scheduler
Mission-level
actions &
resources
Executive
Diagnosis
& Repair
component models
16.412J/6.834J, Fall 02
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 02
Many problems aren’t so hard
Copyright B. Williams
16.412J/6.834J, Fall 02
How can general deduction achieve
reactive time scales?
Candidates with decreasing
likelihood or value
SAT
Generate
Non-conflicting
Successor
Generalization
of Conflicts
Developed RISC-like,
deductive kernel (OPSAT)
Copyright B. Williams
Solutions
When you have eliminated
the impossible, whatever
remains, however improbable
[costly], must be the truth.
- Sherlock Holmes.
The Sign of the Four.
16.412J/6.834J, Fall 02
Can model-based agents perform many different types
of reasoning from a common model?
Engine Op State
Burn_Termination
Burn
Valve
0.01
Open
0. 01
Open
Close
0. 01
Closed
Stuck
open
0.01
Shut_down
Burn_Ignition
Off
Late_Prep
Early_Prep
Stuck
closed
inflow = outflow = 0
Wait
On
VDECU Op_State
Copyright B. Williams
Transition Systems +
Constraints + Probabilities
16.412J/6.834J, Fall 02
Outline
• Motivation
• Model-based autonomous systems
• Remote Agent Example
Copyright B. Williams
16.412J/6.834J, Fall 02
Remote Agent Architecture
Remote Agent
Mission
Manager
Executive
Planner/
Scheduler
Diagnosis
& Repair
RAX_START
Ground
System
Real-Time
Execution
RAX Manager
Planning Experts
(incl. Navigation)
Copyright B. Williams
Fault
Monitors
Flight
H/W
16.412J/6.834J, Fall 02
Executive requests plan
Remote Agent
Mission
Manager
Executive
Planner/
Scheduler
Diagnosis
& Repair
RAX_START
Ground
System
Real-Time
Execution
RAX Manager
Planning Experts
(incl. Navigation)
Copyright B. Williams
Fault
Monitors
Flight
H/W
16.412J/6.834J, Fall 02
Mission manager establishes goals,
planner generates plan
Remote Agent
Mission
Manager
Executive
Planner/
Scheduler
Diagnosis
& Repair
RAX_START
Ground
System
Real-Time
Execution
RAX Manager
Planning Experts
(incl. Navigation)
Copyright B. Williams
Fault
Monitors
Flight
H/W
16.412J/6.834J, Fall 02
Executive executes plan
Remote Agent
Mission
Manager
Executive
Planner/
Scheduler
Diagnosis
& Repair
RAX_START
Ground
System
Real-Time
Execution
RAX Manager
Planning Experts
(incl. Navigation)
Copyright B. Williams
Fault
Monitors
Flight
H/W
16.412J/6.834J, Fall 02
Diagnosis system monitors and repairs
Remote Agent
Mission
Manager
Executive
Planner/
Scheduler
Diagnosis
& Repair
RAX_START
Ground
System
Real-Time
Execution
RAX Manager
Planning Experts
(incl. Navigation)
Copyright B. Williams
Fault
Monitors
Flight
H/W
16.412J/6.834J, Fall 02
Walk Through of
Cassini Saturn Orbital
Insertion
courtesy JPL
Plan for Next Time Horizon
Remote Agent
Mission
Manager
Executive
Planner/
Scheduler
Diagnosis
& Repair
RAX_START
Ground
System
Real-Time
Execution
RAX Manager
Planning Experts
(incl. Navigation)
Copyright B. Williams
Fault
Monitors
Flight
H/W
16.412J/6.834J, Fall 02
Thrust
Goals
Power
Attitude
Engine
Copyright B. Williams
16.412J/6.834J, Fall 02
Mission Manager Sets Goals
over Horizon
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Off
16.412J/6.834J, Fall 02
Planner Repeatedly Applies
Library of Operational Constraints
Thrust
Goals
Delta_V(direction=b, magnitude=200)
contains
Engine
Copyright B. Williams
Thrust (b, 200)
16.412J/6.834J, Fall 02
Planner Repeatedly Applies
Library of Operational Constraints
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
contained_by
Point(b)
Attitude
Engine
Copyright B. Williams
equals
meets
contained_by
Warm Up
Thrust (b, 200)
met_by
Off
16.412J/6.834J, Fall 02
Planner Starts
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Thrust (b, 200)
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Thrust (b, 200)
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Thrust (b, 200)
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Thrust (b, 200)
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Point(b)
Thrust (b, 200)
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Point(b)
Thrust (b, 200)
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Warm Up
Point(b)
Thrust (b, 200)
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Warm Up
Point(b)
Thrust (b, 200)
Turn(b,a)
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a)
Engine
Off
Copyright B. Williams
Warm Up
Point(b)
Thrust (b, 200)
Turn(b,a)
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a) Turn(a,b)
Engine
Off
Copyright B. Williams
Warm Up
Point(b)
Thrust (b, 200)
Turn(b,a)
Off
16.412J/6.834J, Fall 02
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a) Turn(a,b)
Engine
Off
Copyright B. Williams
Warm Up
Point(b)
Thrust (b, 200)
Turn(b,a)
Off
16.412J/6.834J, Fall 02
Plan Completed!
Thrust
Goals
Delta_V(direction=b, magnitude=200)
Power
Attitude
Point(a) Turn(a,b)
Engine
Off
Copyright B. Williams
Warm Up
Point(b)
Thrust (b, 200)
Turn(b,a)
Off
16.412J/6.834J, Fall 02
Plans Allow Temporal Flexibility
Through Least Committment







Copyright B. Williams
16.412J/6.834J, Fall 02
The executive dynamically
schedules and dispatches tasks
Remote Agent
Mission
Manager
Scripted
Executive
Planner/
Scheduler
Diagnosis
& Repair
RAX_START
Ground
System
Real-Time
Execution
RAX Manager
Planning Experts
(incl. Navigation)
Copyright B. Williams
Fault
Monitors
Flight
H/W
16.412J/6.834J, Fall 02
Executing Temporal Plans
[0, 300]


<0, 0>

[130,170]]
• Propagate temporal constraints
• Select enabled events
• Terminate preceding activities
• Run next activities
Copyright B. Williams
16.412J/6.834J, Fall 02
Propagating Timing Constraints
Can Be Costly
EXECUTIVE
CONTROLLED SYSTEM
Copyright B. Williams
16.412J/6.834J, Fall 02
Propagating Timing Constraints
Can Be Costly
EXECUTIVE
CONTROLLED SYSTEM
Copyright B. Williams
16.412J/6.834J, Fall 02
Propagating Timing Constraints
Can Be Costly
EXECUTIVE
CONTROLLED SYSTEM
Copyright B. Williams
16.412J/6.834J, Fall 02
Propagating Timing Constraints
Can Be Costly
EXECUTIVE
CONTROLLED SYSTEM
Copyright B. Williams
16.412J/6.834J, Fall 02
Propagating Timing Constraints
Can Be Costly
EXECUTIVE
CONTROLLED SYSTEM
Copyright B. Williams
16.412J/6.834J, Fall 02
Propagating Timing Constraints
Can Be Costly
EXECUTIVE
CONTROLLED SYSTEM
Copyright B. Williams
16.412J/6.834J, Fall 02
Solution: Compile Temporal
Constraints to an Efficient Network
EXECUTIVE
CONTROLLED SYSTEM
Copyright B. Williams
16.412J/6.834J, Fall 02
Solution: Compile Temporal
Constraints to an Efficient Network
EXECUTIVE
CONTROLLED SYSTEM
Copyright B. Williams
16.412J/6.834J, Fall 02
Solution: Compile Temporal
Constraints to an Efficient Network
EXECUTIVE
CONTROLLED SYSTEM
Copyright B. Williams
16.412J/6.834J, Fall 02
Execution and Fault Recovery involves
monitoring and commanding hidden states
Remote Agent
Mission
Manager
Scripted
Executive
Planner/
Scheduler
Diagnosis
& Repair
RAX_START
Ground
System
Real-Time
Execution
RAX Manager
Planning Experts
(incl. Navigation)
Copyright B. Williams
Fault
Monitors
Flight
H/W
16.412J/6.834J, Fall 02
Model-based Execution of Activities
Programmers and operators must reason through
system-wide interactions to generate codes for:
•
•
•
•
•
monitoring
tracking goals
confirming commands
detecting anomalies
diagnosing faults
Estimating Modes
Copyright B. Williams
• reconfiguring hardware
• coordinating control
policies
• recovering from faults
• avoiding failures
Reconfiguring Modes
16.412J/6.834J, Fall 02
Model-based Execution as
Stochastic Optimal Control
Goals
Model
Controller
Plant
mode
Estimation
o(t)
s’(t)
mode
reconfiguration
(t)
s (t)
g
f
Livingstone
Copyright B. Williams
16.412J/6.834J, Fall 02
Models
• modes engage physical processes
• encoded as finite domain constraints
• probabilistic automata for dynamics
• Concurrency to model multiple processes
vlv=stuck open =>
Outflow = Mz+(inflow);
vlv=open =>
Outflow = Mz+(inflow);
Stuck
open
Open
Open
Cost 5
Prob .9
Closed
Vlv = closed =>
Outflow = 0;
Copyright B. Williams
Close
Stuck
closed
vlv=stuck closed=>
Outflow = 0;
16.412J/6.834J, Fall 02
Possible Behaviors
Visualized by a Trellis Diagram
X0
X1
XN-1
XN
S
T
•Assigns a value to each
variable.
•A set of concurrent
transitions, one per automata.
•Consistent with all state
constraints.
•Previous & Next states
consistent with source &
target of transitions
Copyright B. Williams
16.412J/6.834J, Fall 02
Model-based Execution as
Stochastic Optimal Control
Goals
Model
Controller
mode
Estimation
o(t)
Valve fails
stuck closed
Plant
X0
X1
S
Livingstone
Current Belief State
Copyright B. Williams
s’(t)
mode
reconfiguration
(t)
s (t)Fire backup
XN-1
engine
g
XN
T
X0
f
X1
S
First Action
XN-1
XN
T
least cost reachable
goal state
16.412J/6.834J, Fall 02
RMPL Model-based Program
Control Program
Titan Model-based Executive
Control Sequencer:
Generates
goal states
Control
Sequencer
conditioned on state estimates
Executes concurrently
 Preempts
 Asserts and queries states
 Chooses based on reward

State estimates
System Model
State goals
Mode
Estimation:
Tracks likely
States
Mode
Reconfiguration:
Tracks least-cost
state goals
Deductive Controller
Commands
Observations
Plant
Valve fails
stuck closed
X0
Fire backup
engine
X1
S
Current Belief State
XN-1
XN
X0
T
X1
S
First Action
XN-1
XN
T
least cost reachable
goal state
Mode Estimation and Diagnosis
Observe
“no thrust”
Find most likely reachable states
consistent
with
observations.
Copyright B. Williams
16.412J/6.834J, Fall 02
Mode Reconfiguration and Repair
Goal: Achieve Thrust
Copyright B. Williams
16.412J/6.834J, Fall 02
Mode Reconfiguration and Repair
Goal: Achieve Thrust
Copyright B. Williams
16.412J/6.834J, Fall 02
Mode Reconfiguration and Repair
Goal: Achieve Thrust
Copyright B. Williams
16.412J/6.834J, Fall 02
Demonstration:
New Millennium Advanced
Autonomy Prototype
July - November, 1995
courtesy JPL
Remote Agent on Deep Space 1
Started: January 1996
Launch: Fall 1998
Copyright B. Williams
16.412J/6.834J, Fall 02
Remote Agent Experiment
See rax.arc.nasa.gov
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
Copyright B. Williams
16.412J/6.834J, Fall 02
Beyond: Cooperative Exploration
Model-based Embedded
and Robotic Systems Group, MIT
Copyright B. Williams
Distributed Planning Group, JPL
16.412J/6.834J, Fall 02
“With autonomy we declare that no
sphere is off limits. We will send
our spacecraft to search beyond the
horizon, accepting that we cannot
directly control them, and relying on
them to tell the tale.”
Bob Rasmussen
Architect
JPL Mission
Data System
Descargar

IJCAI Introduction