Advanced Multi-Agent-System for Security applications

brainystitchAI and Robotics

Nov 14, 2013 (3 years and 8 months ago)

47 views

Advanced Multi
-
Agent
-
System
for Security applications


Dr. Reuven Granot

Faculty of Science and Scientific Education

University of Haifa, Israel

rgranot@smile.net.il

June
19
,
2006

RISE
2006

2

Robotic activities at University of
Haifa


The new
Faculty of Science and Scientific Education’s

mission is focused toward
interdisciplinary research

and
education.



The robotic activities have their background in the initiative of the
Research & Technology Unit

at MAFAT Israel MoD were I served in
the last decade as Scientific Deputy.


We have concentrated interest and research in

Multi


Agent
Supervised Autonomous Systems (Tele robotics),
while
continuing steady support of the Manual Remote operations in
different combat environments.

June
19
,
2006

RISE
2006

3

Overview


The Tele
-
robotics paradigm.


The Control Agent

as the implementation of the
relevant behavior.


Human Robot Interaction.


JAUS and Real time Control System
Architectures.


Evaluation of concepts using Small Size Scaled
Model.


Video demonstration.

June
19
,
2006

RISE
2006

4

The Need of Unmanned Systems


DDD



Dull


Dirty


Dangerous


Distant



at different scale


Macro: space,


Micro: telesurgery, micro and
nano devices

Regarding Defense and Security the
need is well recognized to perform
tasks that are:

All these applications require an
effective interface

between the
machine and a human in charge of operating/ commanding the
machine.

June
19
,
2006

RISE
2006

5

The Tele
-
robotics paradigm

Telerobotics

is a form of
Supervised Autonomous Control.

A machine can be distantly operated by:



continuous control
: the HO is responsible to
continuously supply the robot all the needed
control commands.



a
coherent cooperation

between man and
machine, which is known to be
a hard task
.

Supervision and intervention by a human
would provide the
advantages

of on
-
line
fault correction and debugging
, and
would relax the amount of structure needed in the
environment,
since a human supervisor could anticipate and
account for many unexpected situations
.

June
19
,
2006

RISE
2006

6

Remote Controlled vehicles in combat environment



RC is still preferred by designers

o

Simple,
but not practical

for combat environment because
the human operator:


is very much dependent upon the controlled process



needs long readjustment time to switch between the
controlled and the local (combat) environment.


The needed control metaphor:
Human Supervised Autonomous



The state of the art of the current technology
has not yet
solved

the problem of controlling
complex tasks

autonomously
in
unexpected contingent environments
.

o

dealing with
unexpected

contingent events remains to be a
major problem of robotics.



Consequence:
A
human

operator
should be able to interfere:
remains

at least in the supervisory loop.

June
19
,
2006

RISE
2006

7

Why Security Systems should make use of
the Telerobotic paradigm


Require


Reduced number

of human operators.


HO should control simultaneously

several systems.


High
flexibility

and factor of
surprise.



HO should be capable to deal with other duties in somehow
relaxed mode of operation.


Means:


Distributed systems.


Coherent collaboration

of human intelligence with machine
superior capabilities.


Make the machine an
agent

in human operator’s
service
.

June
19
,
2006

RISE
2006

8

The spectrum of control modes.



Solid line= major loops are closed through computer, minor loops through human.


traded control:

control is
or

at
operator
or

at the
autonomous sub
-
system.


shared control
: the
instructions given by
HO and by the robot
are combined
.


strict supervisory
control
: the HO
instructs

the robot,
then
observes

its
autonomous actions.

A
telerobot

can use:

June
19
,
2006

RISE
2006

9

Human Robot Interaction


In
supervised autonomously controlled equipment,
a human operator generates
tasks
, and a computer
autonomously closes some

of the controlled loops
.


Control bandwidth


Robot SW:
high


Human response:
slow

June
19
,
2006

RISE
2006

10

The Agent


An agent is a computer system capable of
autonomous

action in some environments.


A general way in which
the term agent is used

is to denote
a hardware or software
-
based computer system

that enjoys
the following properties:


autonomy
: agents operate without the direct intervention of
humans or others, and have some kind of control over their actions
and internal state;


social ability
: agents interact with other agents (and possibly
humans) via some kind of
agent
-
communication language
;


reactivity
: agents perceive their environment, (which may be the
physical world, a user via a graphical user interface, or a collection
of other agents), and respond in a timely fashion to changes that
occur in it;


pro
-
activeness
: agents do not simply act in response to their
environment; they are able to exhibit goal
-
directed behavior by
taking the initiative
.

June
19
,
2006

RISE
2006

11

Agents are not Objects


Differ from
Objects


autonomous, reactive and pro
-
active



encapsulate some state,


are more than expert systems



are situated in their environment

and
take action

instead of
just advising

to do so.




Agents may act inside the robot software to implement
behaviors:



Feedback controllers



Control subassemblies



Perform Local Goals/ tasks

June
19
,
2006

RISE
2006

12

The Control Agent


The agent is a control
subassembly
.


It may be built upon a
primitive task

or composed
of an
assembly

of subordinate agents.


The agent
hierarchy

for a specific task is
pre
-
planned

or defined by the human operator as
part
of the preparation

for execution of the task.


The final sequence of operation is deducted from
the hierarchy or

negotiated

between agents in the
hierarchy.

June
19
,
2006

RISE
2006

13

Agent control loop


agent starts in some initial
internal state
i
0

.


observes its environment
state
e
, and generates a
percept
see(e)
.



internal state of the agent is then updated via
next
function, becoming
next_(i
0
,
see(e))
.



the action selected by agent is

action (
next(i
0
,
see(e))
))


This action is then performed
.


Goto (
2
).

June
19
,
2006

RISE
2006

14

Human Operator


Monitors the
activities

and the
performance

of the assembly of
agents
.


Responsible for the
completion of the major task

(global goal)



may interfere by sending
change orders
.



emergent (executed immediately)


“as is ordered” or


normal


checked by the interface agent


which
negotiates execution

with other agents in order to
optimize execution performance



Conflict resolution algorithm


defined as
default
, or


defined by the
human operator

in its change order or


suggested to

the operator by a simplified
decision support
algorithm.

June
19
,
2006

RISE
2006

15

Man Machine Interface

is
still one of the most
recognized technology
gaps/ challenges of semi
autonomous systems.

Intelligent Control

will be achieved using Intelligent Agents.

June
19
,
2006

RISE
2006

16

Interface Agent


A software entity, which is capable to
represent

the human in the computer SW environment.


It
acts on behalf

of the human


Follows rules

and has a well defined expected
attitude/ action.


May be instructed
on the fly

and may receive
during mission

updated commands from the
human operator.


We need to build agents in order to carry out the
tasks,
without the need to tell the agents how to
perform these tasks.

June
19
,
2006

RISE
2006

17

Task
-
level supervisory control system block diagram
.


Controlling agent

Task level

controller

Robot hardware

desired
tasks

formatted
outputs

control
signals

raw
robot
outputs



An agent can be considered as a
control subassembly,
also called
behavior
.



The
feedback is given to the agent

in both
processed and raw

form.

June
19
,
2006

RISE
2006

18


RCS

Embeds a
hierarchy of agents

within
a hierarchy of organizational units:
Intelligent Nodes

or
RCS_Nodes
.

Squad
Commander
Squad
Commander
Squad
Commander
Platoon
Commander
Vehicle
Commander
Vehicle
Commander
Vehicle
Commander
Vehicle
Commander
Squad
Commander
JAUS

From
M. W. Torrie


A hierarchy of Commanders


different resolution in space and time

June
19
,
2006

RISE
2006

19

RCS_Node

Value
Judgment

Sensory
Processing

World
Modeling

Behavior
Generation

Knowledge
Database

Update

Plan

State

Predicted
Input

Observed
Input

Perceived
Objects &
Events

Commanded
Actions
(Subgoals)

Commanded
Task (Goal)

Plan
Results

Situation
Evaluation

June
19
,
2006

RISE
2006

20

Agents in Behavior
Generation hierarchy


Tasks are decomposed and
assigned in a command
chain.


Actions are coordinated



Resources are allocated as
plan approved.



Tasks achievements are
monitored (VJ)


Execution in parallel

June
19
,
2006

RISE
2006

21

Evaluation of concept


As an emerging scientific field, the field of
robotics (like AI) lacks the
metrics

and
quantifiable measures

of performance.


Evaluation is done
against
common sense

and
qualitative

experimental results
.


the
legitimacy
of transfer of conclusions
over different scale

applications or different
implementations remains to be decided by
specific designs.

June
19
,
2006

RISE
2006

22

Small Size Scaled Model


The implementation differs by mechanical,
perceptual and control elements from the full scale
application.


It still may help to identify
unusual situations
which
the software agent must be capable to deal
with.


Full scale machines may be tested only at field
ranges, which are
time consuming

and
very

expensive
.


A small scale model
may be tested in office

environment, enabling the software developers to
shorten test cycles

by
orders of magnitude.


June
19
,
2006

RISE
2006

23

D
9
Bulldozer




The operator has very limited
information

about his surroundings
or machine performance.


A
good starting project
:


earthmoving tasks are loosely coupled with
locomotion tasks.


earthmoving tasks are not really simple and


locomotion tasks are not really complicated.

June
19
,
2006

RISE
2006

24

Expected situations



The bulldozer moves forward placing the blade too low


The human decides: the blade should be placed higher



Command issued: “lift the blade”.


experiencing too much power to enable earth moving
forward


the human operator would prefer to
withdraw

and
attack the soil from a new position behind


the human operator is
distant


the bulldozer
is “close” to the ditch
;

>

a better practice would be to first complete the maneuver
.




Bulldozer using Fuzzy Control decides to perform the
better practice and withdraws
only after the maneuver is
completed
.

June
19
,
2006

RISE
2006

25

The Model

June
19
,
2006

RISE
2006

26

Drawbacks


DC motors are of
relatively weak power

and
small
dimensions



which reduce our choice of suitable sensors.


therefore, we implemented



simulated beacon


CMUcam placed above

-

is a simulation of the
"Flying Eye" concept of FCS


We
were unable

to control the
speed
of the vehicle.



We had to restrict

our testing to control


the vehicle
rotation

around a perpendicular axis


to manipulate the
raising of the blade
.


June
19
,
2006

RISE
2006

27

robot.WMV
\
bulldozer
-
autonomous

robot.mpg
\
bulldozer
-
autonomous

4
min

3
min

June
19
,
2006

RISE
2006

28

June
19
,
2006

RISE
2006

29

June
19
,
2006

RISE
2006

30

Conclusions


Security systems should use the advantages of the
Telerobotic paradigm

in order to perform complex
tasks with few operators.


Agents

are implementations of behaviors.


Behavior based
Architectures

are better
implemented using the
Multi Agent technology
.


Human Machine Interaction

is better implemented
through the
Interface Agent
.


Machine Intelligence

may be achieved
implementing agents into the JAUS/ RCS Model
Architecture.

June
19
,
2006

RISE
2006

31

Some
References


NATO Core Group in Robotics

(members)


2005
: Bridging the Gap in
military Robotics (to be published as NATO document)
www.fgan.de/~natoeuro/EuropeanRobotics
-
Publication.pdf



Sheridan,

T.B., Telerobotics, Automation, and Human Supervisory Control,
MIT Press,
1992



Granot R,
Agent based Human Robot Interaction.

at
IPMM
2005
,
Monterey, California,
19
-
25
July
2005


Granot, R., Feldman, M.,
2004
: "Agent based Human Robot Interaction of a
combat bulldozer." Unmanned Ground Vehicle Technology IV, at SPIE
Defense & Security Symposium
2004
(formerly
AeroSense
)
12
-
16
April
2004
, Gaylord Palms Resort and Convention Center Orlando, Florida USA,
paper number
5422
-
25



Granot, R.,
2002
: "Architecture for Human Supervised Autonomously
Controlled Off
-
road Equipment.


Automation Technology for Off
-
road
Equipment",
ASAE, Chicago, Il, USA, July
26
-
28
,
2002
, p
24


Meystael M. A. and Albus, S. J.

"Intelligent Systems. Architecture, Design,
and Control", John Wiley & Sons Inc.,
2002



Michael Wooldridge, "Intelligent Agents: Theory and Practice"
http://www.csc.liv.ac.uk/~mjw/pubs/ker
95
/


June
19
,
2006

RISE
2006

32

Contact

Dr. Reuven Granot


rgranot@smile.net.il


granot@math.haifa.ac.il

University of Haifa


Faculty of Science and Scientific Education

Mount Carmel Haifa

31905
ISRAEL



Office


+
972 4
-
828
-
8422
cellular +
972 52 341
-
0193


http://math.haifa.ac.il/robotics


This presentation is downloadable from

http://math.haifa.ac.il/robotics/Projects/MyPapers/RISE
2006
.ppt