Newsgroups: comp.robotics
Path: brunix!sgiblab!spool.mu.edu!agate!doc.ic.ac.uk!warwick!nott-cs!rsx
From: rsx@cs.nott.ac.uk (Richard Seldon)
Subject: Mobile Robots Project
Message-ID: <1993Aug26.130753.12986@cs.nott.ac.uk>
Organization: Nottingham University
Date: Thu, 26 Aug 93 13:07:53 GMT
Lines: 2169

I recently sent out a few items to the bulletin board asking for info'
on Mobile Robots.  I have come to the end of the project, so I am posting
all the items I received.  I've taken out the rubbish (such as 15 line 
addresses!!!!).  I have included one of the letters I wrote so the replies
will be a bit more meaningful to you and I have also included the questionaire
I sent out.  If you are interested in this article you may find getting a
copy of the questionaire on paper useful, since I have only listed the 
answers not the questions they answer! Does that make sense? I hope so!!!!!


WITHOUT THE HELP OF EVERYONE WHO REPLIED MY DISSERTATION WOULD HAVE
BEEN HARDER AND NOT HALF AS ENJOYABLE AS IT HAS BEEN.

		THANKS FOR TAKING THE TIME TO REPLY

				RICH.


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

The first letter I sent out.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Dear world

I am currently  doing an MSc in Information Technology.
To complete it I am investigating the current state of Mobile Robots, a
subject I covered in my Mechanical Engineering degree.

Since the major research areas appear to be America, Japan, France
and Australia the bulletin board appears to be a good place to start
my enquiries.

What I am after is:

1)   Any information relating to Mobile Robots.
2)   Any names (and e-mail addresses where available)
     of people I could get in touch with.

All replies will be very much appreciated, either on the bulletin board
or direct to my e-mail adress, as above.


		with thanks

				Richard SELDON
					University of NOTTINGHAM
					UK.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

This is a copy of the questionaire I sent out.  The answers given later
on relate to the questions asked below.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
		      ==============================	
		      /	MOBILE ROBOTS QUESTIONAIRE \
		      \		   BY		   /
		      /     RICHARD L SELDON       \
		      ==============================

	      ================================================
	      *	A)	ABOUT THIS QUESTIONAIRE:	     *
	      *						     *
	      *	This is aimed at anyone who is involved with *
	      *	Mobile Robots and can also spare the time to *
	      *	answer a few short questions.		     *
	      ================================================

	      ================================================
	      *	B)       ABOUT MYSELF:			     *
	      *						     *	
	      *	Richard L SELDON			     *
	      *	Honours Degree Mechanical Engineering 1989   *
	      *   Brookes University, Oxford		     *
	      * CURRENTLY				     *
	      * MSc in Information Technology, expected 1993 *
	      *   Nottingham University			     *		
	      ================================================	
	
	      ================================================
	      * C)      ABOUT THE PROJECT:		     *
	      *						     *
	      * The course supervisor, Dr P Edwards, who is  *
	      *	also my project supervisor has a keen        *
	      * interest in Robotics.			     *
	      *  					     *
	      * The purpose of this project is to assess the *
	      * current state of Mobile Robots worldwide.    *
	      ================================================



	 **********************************************************
	 ** I would like to thank everyone who has already made  **
	 ** contributions to this project including information, **
	 ** e-mail addresses and references.			 **
	 **********************************************************

	
    |********************************************************************
    |
    | Question 1:
    | -----------
    | How does your robot move itself? (Legs, wheels, tracks etc)
    |
    | Answer 1:
    | ---------
    |
    |
    | 
    | Question 2:
    | -----------
    | What speeds and/or gaits are possible?
    |
    | Answer 2:
    | ---------
    |
    |
    |
    |
    | Question 3:
    | -----------
    | How is it operated? (eg teleoperated)
    | 
    | Answer 3:
    | ---------
    |
    |
    |
    | Question 4:
    | -----------
    | What does it do?
    |
    | Answer 4:
    | ---------
    |
    |
    | Question 5:
    | -----------
    | How succesful has it been?
    |
    | Answer 5:
    | ---------
    |
    |
    |
    |
    | Question 6:
    | -----------
    | Are there any ideas for current/future applications
    | other than research?
    |
    | Answer 6:
    | ---------
    |
    | 
    |
    | Question 7:
    | -----------
    | What language is used for programming it?
    |
    | Answer 7:
    | ---------
    |
    | 
    |
    |
    | Question 8:
    | -----------
    | Any extra details you would like to include:
    |
    | Answer 8:
    | --------- 
    |
    |
    |      
    |									  
    |									  
    |									  
    |									  
    | Question 9:							  
    | -----------							  
    | Journals where further details of your robot have been published    
    | (if applicable) or any references you feel would be useful.         
    |									  
    | Answer 9:								  
    | ---------								  
    |									  
    |									  
    |									  
    |*********************************************************************
 


NOTE 1:  If you can reply via e-mail I would be grateful if the
         SUBJECT heading could be set to qreplies:- This will
	 make filing as they arrive more efficient.

NOTE 2:  I would like to include photographs, where possible, of
	 as many of the robots mentioned in my project.  If you
	 reply to this and have a photograph(s) I would be pleased
         to receive a copy of it by ordinary post. (My address is
	 included at the bottom.)  I have access to a GIF viewer
	 if anyone wishes to send photos that way.

	 I would be happy to send stamps (or stamped, addressed
	 envelope) to cover mailing costs.  Please indicate if
         you wish me to do so.



	 TTTTT   H   H   AAAAA   NN    N   K   K   SSSSS
	   T     H   H   A   A   N N   N   K  K    S
           T     HHHHH   AAAAA   N  N  N   KKK     SSSSS
           T     H   H   A   A   N   N N   K  K        S
           T     H   H   A   A   N    NN   K   K   SSSSS

	+++++++++++++++++++++++++++++++++++++++++++++++++
        +++++++++++++++++++++++++++++++++++++++++++++++++	






Date: Mon, 12 Jul 93 14:47:00 GMT
From: kenneth r lipka <ferret@engin.umich.edu>
Message-Id: <9307121447.AA12995@raven.engin.umich.edu>
To: Richard Seldon <rsx@Cs.Nott.AC.UK>
Subject: Re: Mobile Robots

	Richard,

	I think I might be able to help you with your request for info on 
studies in Mobile Robotics.

	This summer I am working in the Advanced Technology Lab (ATL) here on
the main campus of the University of Michigan (Ann Arbor).  The Mechanical
Engineering department maintains a small mobile robotics group here.  The main
person to contact (I suppose) would be Johann Borenstein, a research scientist.
His e-mail address is:  JOHANN_BORENSTEIN@UM.CC.UMICH.EDU.  The other main ME
Robotics person here is Prof. Yorem Koren.  His address is:  YOREM_KOREN@UM.CC.
UMICH.EDU.

	Hopefully these people should be able to help you.  Feel free to e-mail
me with any other questions - I can only try to answer them.  (Something I just
remembered.  Prof. Koren is on sabbatical until the end of the summer; so any
e-mail might not reach him for a little while.)

   ________)
  (    /      Ken Lipka
    __/_)     ferret@engin.umich.edu 
   ( /        Graduate Student - Mechanical Engineering 
 (__/         University of Michigan:  Ann Arbor, Michigan, USA 

Disclaimer:  Any errors are the fault of the American Education System.


Date: 12 Jul 93 09:04:00 MST
From: NICKGILL@batman.bmd.trw.com
Subject: Re: Mobile Robots
To: rsx <rsx@Cs.Nott.AC.UK>

Hello Richard,

I am working on a Army R&D mobil robot. It uses a cybermotion platform and
is equipted with various sensors to detect intrusion, theft, etc.

Nick


To: Richard Seldon <rsx@Cs.Nott.AC.UK>
Organization: Computer Science, Trinity College Dublin
From: Gerard Lacey <Gerard.Lacey@cs.tcd.ie>
Sender: Gerard.Lacey@cs.tcd.ie

Would you like to narrow you request a little and I'll tell you 
what I know ? ie is it the mechanical, electronic, computer 
architectures, sensors and sensor fusion, or tasks that you are 
interested in.  ( I have dived into the field fairly recently)

ps have you read the robotics FAQ ?

Gerard Lacey email: gerard.lacey@cs.tcd.ie
Computer vision Lab
Trinity College Dublin
Ireland

Date: Mon, 12 Jul 1993 13:39:30 -0700
From: Michael Sahota <sahota@cs.ubc.ca>
To: Richard Seldon <rsx@Cs.Nott.AC.UK>
Subject: Re: Mobile Robots

Your best bet is to look at conference proceedings.
Try: AAAI , and Robotics & Automation.

Also, your post did not make it clear whether you 
were interested in robot architecture (i.e. control and programming)
or mechanical issues.

Cheers,
	Michael

Date: Tue, 13 Jul 1993 17:02:02 +0100 (MET)
From: Ravenous Max <IANIGRO@astrba.ba.cnr.it>
Subject: Re: Mobile Robots
To: rsx@Cs.Nott.AC.UK

Hello Richard, my name is Massimo Ianigro. I am a graduate student and i work
with the italian National Research. I work on Mobile Robots, especially on
ultrasonic-based navigation and mobile robot location using vision techinques.
In my institute there are also other researchers involved in projects with
mobile robots.

For further information, feel you free to contact me.
	ciao
		massimo

p.s. I apologize for my bad English.  :-)

 -----------------------------------------------------------------------------
| Massimo Ianigro         |                                                   |
| CNR  -  IESI            |           E-write at :  IANIGRO@ASTRBA.BA.CNR.IT  |
| Via Amendola, 166/5     |           Speak at   :  39 80 481969              |
| 70126  BARI   ITALY     |           Fax at     :  39 80 484311              |
 ----------------------------------------------------------------------------

To: rsx@Cs.Nott.AC.UK
Subject: Re: Mobots
Date: Wed, 14 Jul 1993 15:28:59 +0100
From: "Gerard Lacey (tel. +353-1-7021136)" <Gerard.Lacey@cs.tcd.ie>

Most mobile robots are built with a particular niche in mind.  We
have not reached the stage of real flexibility with mobiles.
Because of this the architecture/electronics tends to shape the
capabilities of a robot to a lesser or greater extent.

There are DTI standards in the UK for speed (top speed that is)
I can't remember the figure, for some reason 6m/s comes to mind
but it sounds a little fast

Lisp and C seem to be the most popular.
If you can get your hands on the IFAC proceedings from a workshop
on Intelegent Autonomus Vehicles held in University of Southampton 
in March 93 you will be well on the way.  Prof Chris Harris organised
it. It would have answers to many of your questions.  The EC have
a project running called PANORAMA, lots of partners, lots of mobiles
they would probably through some info your way.

Good Luck

Gerry

From: Nasser Esmaili <nasser@cs.unsw.oz.au>
To: rsx@Cs.Nott.AC.UK
Date: Thu, 15 Jul 93 17:52:23 +1000
Subject: Re: Mobile Robots

Hello there,

   I am a PhD student in school of Computer Sc. and Eng. at Universitty of
New South Wales in Australia. My PhD thesis (Fully by Research) is based on
Machine learning technics and I had to design and build a Mobile Robot as a
plattform for my research.

   So, if I can be a help, please don't hesitate to contact me.

   Cheers,
   Nasser.

Date: Thu, 15 Jul 93 10:40:51 -0400
From: Ramiro Liscano <liscano@watnow.uwaterloo.ca>
To: "rsx@Cs.Nott.AC.UK" <rsx@Cs.Nott.AC.UK>
Subject: Mobots

Richard,

I received your email message concerning a survey into what tasks
mobile robots can do. What I would be more interested in is what they
can do effectively and efficiently. I have worked in this area for
several years and have contacted several companies. Of course most of
my interest has been in Canada. It is not clear what type of
technology is needed by these companies. Several problems occur. 1)
They are not very open in discussing their needs, 2) In some cases
aadvanced technology is not warranted. For example, there is a company
here that produces tele-operated bomb and waste disposal mobile
robots. They do well in the international market without any advanced
equipment. So far none of their customers have demanded any. Also look
at Denning, it is in receivership, was their technology not adequate
or was it a matter of economics?.To get to the point, The problem is a
match of what can they do and what is needed.

Ramiro Liscano <liscano@watnow.uwaterloo.ca>
Systems Design Engineering
University of Waterloo
200 University Av.
Waterloo, Ontario, Canada N2L 3G1
Phone:		(519) 885-1211 x6086
Fax:		(519) 746-4791

Date: Thu, 15 Jul 1993 10:42:08 -0400
From: "Caglan M. Aras" <aras@ecerl2.ncsu.edu>
To: rsx@Cs.Nott.AC.UK
Subject: Help?

Hi Richard,

Can you be more specific? A fellow here, named Steven
Goodridge is into the task side of Mobile Robots. You can
reach him at sggoodri@eos.ncsu.edu or sggoodri@ecerl2.ece.ncsu.edu.
I would suggest you send to both accounts since one is 
an account that may be closed for the summer.

Good luck,

Caglan Aras

Date: 15 Jul 93 08:34:00 MST
From: NICKGILL@batman.bmd.trw.com
Subject: Re: Mobile Robots
To: rsx <rsx@Cs.Nott.AC.UK>

Hello Richard,

Well the Cybermotion mobile platform uses three Z80's for collision avoidance,
and general movement. The sensors will use a STD32 bus with microwave, infrared,and some video equipment to detect intrusion. This may be used to protect all 
Army warehouses in the future.

Nick

Date: Thu, 15 Jul 1993 13:39:13 -0500
To: rsx@Cs.Nott.AC.UK
From: Rody Biggert <biggert@coral.bucknell.edu>
Subject: Mobile Robots

Dear Richard:

        I saw your request for information on mobile robots.  I am
currently working on a six-legged hexapod here at Bucknell University in
Pennsylvania and I was wondering if you could send me any of the
information you've received.  I'm looking mainly for information concerning
small hexapods so anything you could send me would be very helpful.

                                        Thanks,

                                                Rody

"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""

Rody Biggert

biggert@mail.bucknell.edu

"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""

From: "Robert W. Albrecht eeb537 5-1600" <bob@maxwell.ee.washington.edu>
Subject: Re: Mobots WHAT DO YOU WANT?
To: rsx@Cs.Nott.AC.UK
Date: Thu, 15 Jul 93 11:14:04 PDT

TO RICHARD SELDON:

WHAT KIND OF "help" DO YOU WANT?

OUR WORK IS CONCERNED WITH MOBILE ROBOTS IN BUILT ENVIRONMENTS
(indoors and in built interbuilding spaces).

Robert W. Albrecht
Professor

Date: Thu, 15 Jul 93 16:57:10 EDT
From: Robotics Club <rbc@engr.ucf.edu>
To: rsx@Cs.Nott.AC.UK

I am afraid that purpose has never been a concern of our group,
being fervent technology nerds, we believe that robots are
destined to exist, and that we should all be willing to sacrifice
our grades, education , and personal lives to that goal.

Hope you excuse the flippant nature of this response, however
it is in fact, factual. :-)

eric wampner
wampner@acme.ucf.edu
rbc@engr.ucf.edu

Date: Fri, 16 Jul 93 10:26:56 +0200
From: Sjur Jonas Vestli <vestli@ifr.ethz.ch>
To: rsx@Cs.Nott.AC.UK
Subject: Re: Mobots

Richard

What do you want to know? And please don't use the term Mobots.
And maybe you should contact more people in the UK. Try
GILLIAN@advanced-robotics-research-centre.salford.ac.uk, the 
secretary of Advanced Robotics Research Ltd. who should be able
to tell you who at ARRL are working on mobile robots right now.

Sjur

Date: Fri, 16 Jul 93 17:29:23 +0200
From: Hannu.Lehtinen@vtt.fi
Subject: Re: Update to mobots survey
To: rsx@Cs.Nott.AC.UK

Richard,  

I thought, it would help.

Our work is not on the list. I am working on a 1200 kg walking machine (tele-
operated by radio). My colleagues are working in the ESRIT PANORAMA, where
autonomous workin machines is made.

Send me your postal address if interested.

    Yours     Hannu

From: "Maja J. Mataric" <maja@ai.mit.edu>
Date: Fri, 16 Jul 93 18:14:17 EDT
To: rsx@Cs.Nott.AC.UK
Subject: Mobots

I couldn't quite tell what your request is about, but in case you are
looking for some papers about control architectures for mobile robots,
I can send you a few, if you send me an address.  Cheers,

<>Maja Mataric

Date: Mon, 19 Jul 93 01:43 PDT
From: "P. Douglas Reeder" <reeder@reed.edu>
To: rsx@Cs.Nott.AC.UK
Subject: Re: Mobile Robots

In article <1993Jul12.122337.15595@cs.nott.ac.uk> you write:
:
:2)   Any names (and e-mail addresses where possible) of researchers
:     that I could contact.
:

Have you tried searching indexes such as the Science Citations Index?

One place to look is the proceedings of the latest IEEE robotics conferece.

See also IEEE Trans. on System ,Man and Cybernetics, v 20 num 6, 
Sepecial Issue on NUnmanned vehicle and intelligent Robotics Systems.

Int. J. Robotics Research v9 n 2, april 1990 special issue on legged
locomotion.

Doug Reeder                              Internet: reeder@reed.edu
Div, Grad & Curl                         USENET:   ...!tektronix!reed!reeder
programming & derivative work 
I am actively seeking scientific programming contracts.

Date: Mon, 19 Jul 93 10:07:28 EDT
From: Johann.Borenstein@um.cc.umich.edu
To: rsx@Cs.Nott.AC.UK
Subject: Mobots

I am not sure if I replied to your message already or not.  If you have 
specific questions about mobile robots, maybe I can direct you towards the
appropriate answers. 
Johann

From: Xavier C <xavice@essex.ac.uk>
Date: Tue, 20 Jul 93 12:03:43 BST
To: rsx@Cs.Nott.AC.UK
Subject: Mobile Robots

Hi Richard,

My name is Cedric and I'm doing an M.Sc in Computer Science here at Essex 
University. For my dissertation, I'm also working on mobile robots (IAVs). 
I'm especially concerned with building up a software paradigm for control 
of these IAVs, within certain constraints. I can definitely furnish you with 
more details if you want it. I'm also quite interested in hearing about what 
you are working on. Hopefully, we may be able to exchange some ideas. Cheers !

Cedric Xavier
University of Essex

Date: Thu, 22 Jul 1993 09:28:57 +0100 (MET)
From: Ravenous Max <IANIGRO@astrba.ba.cnr.it>
Subject: Re: Mobile Robots
To: rsx@Cs.Nott.AC.UK

hi Richard,

I can supply all the info you are requesting. I work on a Transition Research
Corporation LABmate mobile robot. I can provide all the info plus a photo of
our robot as it has been modified according to our needs.

If you want i can send to you also technical info provided by TRC.

If you provide a fax number, i can send them ASAP.

Let me know,

		Massimo

 -----------------------------------------------------------------------------
| Massimo Ianigro         |                                                   |
| CNR  -  IESI            |           E-write at :  IANIGRO@ASTRBA.BA.CNR.IT  |
| Via Amendola, 166/5     |           Speak at   :  39 80 481969              |
| 70126  BARI   ITALY     |           Fax at     :  39 80 484311              |
 ----------------------------------------------------------------------------
From: Xavier C <xavice@essex.ac.uk>
Date: Sun, 25 Jul 93 14:44:24 BST
To: rsx@Cs.Nott.AC.UK
Subject: Mobile Robots

Hi, I'm a post-graduate student here at the University of Essex. I am currently researching to formalise the software paradigm we are going to implement for the mobile robots we have just built. If possible, I would like to find out more about your area of work. If you require, I could also furnish you with more details regarding my work, as well as the mobile robots we have here at the University. Regards

Cedric Xavier

Department of Computer Science
University of Essex
Wivenhoe Park
Colchester CO4 3SQ
England

From: Johann.Borenstein@um.cc.umich.edu
To: rsx@Cs.Nott.AC.UK
Subject: Mobile Robots

Sorry, heavy work load doesn't permit me to spend the time on writing up
1-2 pages of specific info. You may have to use the hard way (read through
the literature). 
Johann :)

Date: Mon, 26 Jul 93 11:41:40 EDT
From: Sanjiv Singh <ssingh@frc2.frc.ri.cmu.edu>
To: Richard Seldon <rsx@Cs.Nott.AC.UK>
Subject: MOBILE ROBOTS

There are lots of group at CMU working with mobile robots.  you'll
have to be more specific.

Sanjiv

To: rsx@Cs.Nott.AC.UK
Subject: MOBILE ROBOTS - comp.robotics #6425
From: Richard Goodwin <rich+@cs.cmu.edu>
Date: Mon, 26 Jul 93 11:55:20 -0400
Sender: Richard_T_Goodwin@gs8.sp.cs.cmu.edu

Richard,

What type of mobile robots are you interested in?  Or more basically,
what type of interest do you have?  

If you let me know, I can probably give you a few names and email
addresses.

Richard
___\---

Date: Mon, 26 Jul 93 13:41:51 EDT
From: Ben.Brown@ius4.ius.cs.cmu.edu
To: rsx@Cs.Nott.AC.UK
Subject: mobile robots

Richard,

There are a number of researchers at CMU, my self included, who are interested
in, and/or working on mobile robots. If you can tell me more specifically
what you need, and what you intend to do with the information, I'll be happy
to help (assuming your intentions are honorable:).

	Ben

Date: Mon, 26 Jul 1993 23:07:06 -0400
From: Karl Kluge <kckluge@eecs.umich.edu>
To: Richard Seldon <rsx@Cs.Nott.AC.UK>
Subject: MOBILE ROBOTS

At Carnegie Mellon, Chuck Thorpe (cet@cs.cmu.edu) is head of the Unmanned
Ground Vehicle Project. Dean Pomerleau (pomerlea@cs.cmu.edu) created the
ALVINN neural network road follower. Tony Stentz (axs@cs.cmu.edu) works
on cross-country navigation, Martial Hebert works on segmenting laser
range finder images.

Karl

Date: Tue, 27 Jul 93 00:13 PDT
From: "P. Douglas Reeder" <reeder@reed.edu>
To: rsx@Cs.Nott.AC.UK
Subject: Re: MOBILE ROBOTS

If you find out who's doing it at Ohio State, please tell me!  I've been
trying to find out myself.


Doug Reeder                              Internet: reeder@reed.edu
Div, Grad & Curl                         USENET:   ...!tektronix!reed!reeder
programming & derivative work 
I am actively seeking scientific programming contracts.

From: Xavier C <xavice@essex.ac.uk>
Date: Tue, 27 Jul 93 16:43:57 BST
To: rsx@Cs.Nott.AC.UK
Subject: Mobile Robots at Essex

Hi Richard,

First to answer your question : there really is no connection between my course
and robots ! I had to choose a dissertation topic and since Essex was building
some simple mobile robots to be used as a teaching tool, I figured I may as 
well apply the knowledge I have (supposedly) to these robots.

I could definitely mail you a few photographs of our robots - no problem. 
Except that we only have one prototype at the moment, and even that has just 
been taken to pieces again. It should be ready in about 2 weeks time.

Included below are some parts of my progress report which I submitted about
2 months ago. I hope it might be of interest.

********************************************************************************

Essex is setting up a new systems laboratory with a robotics theme. It has 
been proposed that the most effective programming paradigm for the new lab is 
to utilise libraries of C functions directed at the differing application 
levels which range from low level hardware control to high level AI work. A 
major reason for this is that the principal development environment (VxWorks) 
supports C programming and that the VxWorks target (a robot) has an 
interpretive C shell which allows functions and commands to be used in a 
seemingly transparent manner.

This project will :

1) review other robot programming paradigms, and highlight any advantages or 
shortcomings in the preferred approach,

2) seek to define a well specified framework approach that will cater for 
inter robot/computer command messaging, AI and low level hardware programming

The objectives have to be realised within the constraints that they must work 
with the existing principal laboratory components. These are :

1) Intelligent Autonomous Vehicle (IAV)

In the case of the Essex lab, IAVs are essentially VME racks mounted on 
simple mobile platforms, named MOVERS (MObile VmE RackS). They are small,
modular, lightweight and with a low power consumption. These racks may contain
cards for processor control, communications, stepper motor drivers and digital
I/O. The racks would not be directly connected to any workstation but would 
communicate via the internet protocols. Wireless internet commuunications may 
be provided via radio Ethernet or Serial Line Interface Protocol (SLIP).

The IAV has 3 wheels, 2 of them differentially driven by stepper motors, while
the third is a castor. There will also be 4 ultrasound scanners, 4 bumber 
boards and an IR detector on the IAV. 3 IR transmitters will be placed around 
the room.

2) Real Time OS

The Essex lab will employ the Wind River VxWorks OS which is a high 
performance, real time, multithreaded, fixed priority OS that can be 
downloaded, programmed and debugged from UNIX workstations via Ethernet.

3) Programming Considerations

Low level systems work requires students to build hardware boards and then 
write some code to drive them. For each system they design, they will start off
with a fairly basic program and refine it as they progress with their design 
and testing. Eventually, these modular systems may combine to form a complete 
robot which will in turn produce a set of functions (or libraries) to drive 
the various sub-systems.

********************************************************************************

I would be very interested if you have any references or material on software 
paradigms used on other mobile robots. It has already been decided that our 
IAVs will be programmed in C, but I am having trouble trying to formalise it. 
I might try to model it along the lines of RCCL or Brook's Subsumption 
Architecture.

Anyway, do include your snail mail address when you reply so I can mail those 
photographs or any other material you might require. Cheers !

Cedric Xavier
University Of Essex

Date: Tue, 27 Jul 93 19:21:31 +0200
From: Hannu.Lehtinen@vtt.fi
Subject: Mobile Robots
To: rsx@Cs.Nott.AC.UK

Richard,

Short answers concerning the walking machine MECANT I of HUT.

1:   200 kg
2:   0.5 m/s
3:   ?
4:   Tele-operated
5:   6 legs
6:   Periodic, wave, free gait under devoplement
7:   All gaits possible, free gait to soft soil
8:   To walk on soft soil like forest
9:   C

Some papers coming. Out of photos now. 

    Greetings    Hannu (from Finland)

To: rsx@Cs.Nott.AC.UK
Subject: Mobile Robot Summary
Date: Tue, 27 Jul 93 15:17:57 -0400
From: Gregg_Podnar@globe.edrc.cmu.edu

Rich,
	EPRI (the Electric Power Research Institute) in the U.S. did
	at least one summary book of all the commercially available
	and research mobile robots.  I wouldn't be surprised if they've
	done more (I think this one was about 1988).  If you have no
	success finding it, I'll try to come up with a better pointer.
								Gregg

From: sggoodri@eos.ncsu.edu
To: rsx@Cs.Nott.AC.UK
Subject: Re:  Mobile Robots
Date: Mon, 02 Aug 93 11:06:15 EDT

Dear Richard,

   I would be glad to send you photos, papers, etc. let me know specifics. I
will try to answer your questions here.

Our robot, MARGE, recently won event three of the 1993 AAAI/IAAI autonomous mobile
robot competition in Washington DC three weeks ago. There are as many
different opinions about artificial intelligence as there are about politics,
but this event helped us clarify that our robot is competent in accomplishing 
a task in an a priori unknown environment.

MARGE is based on a Cybermotion Corp. indoor mobile robot. We added a payload
that included cameras, on-board VME bus, ultrasonic ragefinders, etc. Our
robot is completely autonomous, and uses a colony of fuzzy controllers to 
provide behavior-based control inspired by the intelligence of insects.

> The sort of information I am interested in is:
> 
> 		1)  Payloads
                     the Cybermotion base weighs about 300 lbs, to which we
                     added about 150 lbs of payload. I wish I had metric 
                     conversion here with me, I actually hate U.S. units, sorry.
> 		2)  Speed
                     MARGe top speed is approx 800 CM/sec, but this is
                     dangerous around people indoors.
> 		3)  Repeatability
                     We are interested in simulating the robust behavior of
                     living things, not the precise incompetence of other
                     robots. Behavioral mutations make our robot more
                     competent, so that it does not endlessly loop by making
                     the same mistakes. We use a complex blend of motivational
                     states that borders on chaos. However, the robot always
                     performs its task if at all possible, given its insect
                     brain. Repeatability is not quantifiable.
> 		4)  Accuracy
                     MARGE has wheel encoders, if that's what you mean. they
                     allow the robot to return to a position within 10 cm
                     after touring the room with dead reckoning alone. But we
                     never use just one sense to dominate the vehicle.
> 		5)  Method of movement (legs, wheels, tracks etc.)
                     MARGE has three wheels which always point in the
                     direction of travel.
> 		6)  Types of gait
> 		7)  How succesful they are
                    MARGE adapts to unknown and/or dynamic environments.
                    We won a major event in a national competition.
                    We feel that an autonomous mobile robot is more than a
                    computer on wheels; it is a CREATURE and must be designed
                    with this in mind. the Brain and body affect each other 
                    in an intimate manner. Our techniques can be applied to
                    any type of locomotion because they are inspired by the
                    distributed, qualitative intelligence of animals rather
                    than a fragile world model.
> 		8)  Planned uses if other than research
                    I intend to take my ideas all the way.
> 		9)  Programming language used.
                    C, OS-9 operating system and my own fuzzy control
                    description script. 
> 
> I really want to see how advanced mobile robots are and the different
> types available.  I would really like to include a photo album with
> this project showing all the robots mentioned.
> 
> If you can help along these lines, would you be willing to take and post
> a few piccies if I supplied a stamped, addressed envelope? (I may
> be able to get a fax address if that's simpler.)
> 

Please send envelope, thanks.

                                    
Steven Goodridge
Dept. Electrical & Computer Engineering
Room 232 Daniels Hall
NCSU, Raleigh, NC USA 27695-7911                                     

To: rsx@Cs.Nott.AC.UK
Subject: Re: Mobile Robots 
From: Richard Goodwin <rich+@cs.cmu.edu>
Date: Mon, 02 Aug 93 13:33:17 -0400
Sender: Richard_T_Goodwin@gs8.sp.cs.cmu.edu


Rich,

There are many mobile robots under developement here at CMU.  Most are
used for research purposes.  What I would suggest is that you try to get
a copy of the annual report from the Robotics Institute.  This should
list the robots under developement and give the names of people
involved.  (The Field Robot Center may also have one.)  Try sending mail
to reports@cs.cmu.edu.  They should be able to send you a copy or tell
you where you can get one.

I hope this helps.

Richard
___\---


Date: Mon, 16 Aug 1993 17:04-EDT
From: Mary.Jo.Dowling@ius5.ius.cs.cmu.edu
To: rsx@Cs.Nott.AC.UK
Subject: Re: Mobile Robots - Photos

Well will be happy to supply you with color slide duplicates of the
robots you requested at no charge. We only charge when people
specifically request color or b&W prints of the machines. We give
duplicate slides to people so that they may convert the image to
whatever medium they choose, or have your publisher scan them directly
into your final document for printing. Slides give the best overall
quality for printing, especially in color.
 I will get the slides off to you airmail within the next couple of days.
 Do you need captions for the photos or project descriptions of the projects?
Let me know.
mj
mjd+@cs.cmu.edu

Date: Sat, 14 Aug 93 10:27:02 +0200
From: jim crowley <jlc@castor.imag.fr>
Message-Id: <9308140827.AA00549@castor.imag.fr>
To: rsx@Cs.Nott.AC.UK
Subject: Mobil Robotics at CMU


Richard L. Seldon
  Having been one of the first to work on mobile robots at CMU
I am intrigued by your post to comp.robotics. 
  Could you explain what you are looking for?
    Jim Crowley

Organization: School of Computer Science, Carnegie Mellon
Sender:  gwp@gan.cimds.ri.cmu.edu

I have built a few mobile robots here.  You also will want to contact Kevin 
Dowling (nivek@cmu.edu).					Gregg
Date: Wed, 28 Jul 93 10:43:31 EDT
From: gwp@GAN.CIMDS.RI.CMU.EDU
Sender: gwp@GAN.CIMDS.RI.CMU.EDU



From: Rob Fraser <fraser@aeronautics.southampton.ac.uk>
Date: Tue, 24 Aug 93 14:24:54 BST
To: rsx@Cs.Nott.AC.UK
Subject: Mobile Robotics

                  ADVANCED SYSTEMS RESEARCH GROUP
             Department of Aeronautics and Astronautics
                     University of Southampton
                        SOUTHAMPTON SO9 5NH

Dear Richard,

Chris Harris passed on your letter to me for handling. I'll cover your
points in order:

1. IAV Workshop
---------------

Yes, we organised the workshop last April, and it was very successful.
I'm afraid we donot have any copies of the proceedings available, but
Pergamon Press is publishing it, and enquiries should be directed to 
them:

		IFAC Publications,
		Pergammon Press,
		Headington Hill Hall,
		Oxford,
		OX3 0BW.

The proceedings are entitled "Intelligent Autonomous Vehicles 93", April
19th-21st, 1993, University of Southampton, UK.

2. If you would care to email me a copy of your questionaire, I'll fill
it in. I am clearly not on the bulletin board you mention... which one
are you using?

3. PANORAMA PROJECT
-------------------

I worked on the PANORAMA project for the last four years. It terminates in
about 2 months time, after a total of five years effort. It is a European
Esprit 2 project, involving a great many companies from around Europe: 
SAGEM, BAe, EASAMS, VTT (Finland), Helsinki University, Madrid University,
Southampton University, LNETI, CEA (France) and many others.

The PANORAMA project is developing perception and navigation systems 
for autonomous land vehicle applications, through an incremental development
and demonstration  program. A Mercedes 4 wheel drive vehicle REMI, owned by
SAGEM, is to be the first system testbed, integrating subsystems developed in
the laboratories of the consortium partners. The final demonstrators are to
exhibit autonomous and semi-autonomous capability in unstructured outdoor
environments. Two testbeds are to be developed -- a forestry vehicle which
is to operate in a Finnish forest environment, and a tracked Drilling Machine,
which is required to navigate accurately within  a quarry. The mission 
management processes form an integral part of the Intelligent Systems 
Controller module, being developed by British Aerospace, EASAMS, CEA and 
Southampton University.

Final testbed integration is now underway for demonstration in Finland at
the end of September.

Obtaining information on Panorama is quite hard by our own admission!! The 
best source is the IAV-93 proceedings, were there are about seven papers
by PANORAMA consortium members. In particular, the one by Thierry van Den
Bogaert of SAGEM provides an outline of the Perception and Navigation System,
which is at the heart of the mobile robot system.

I'll aslo see what I can dig out here in the way of project brochures or
papers...

For information, the definitive public history of the PANORAMA project will
be available in the form of a book which I am editting which should be 
available after Christmas (not much use to you immediatley though). It 
should be published by World Scientific Press, but is only in the very
early stages of preparation.

I hope this has been of some use. I'll see what I can dig out of the files
for you on PANORAMA. As a group, we now have in excess of twenty researchers
working on a variety of Uk and European projects on Intelligent Control and
Systems Control. We are also getting involved with Systems Integration
Projects now, and in particular the use of Object Technology in Open
Command and Control Systems (PROTEUS project).

Good luck with the survey. If you have any queries, please use the email 
address below.

Best Wishes,

Rob Fraser.

| R.J.C.Fraser             | ADVANCED SYSTEMS RESEARCH GROUP
| PROTEUS Project          | Department of Aeronautics and Astronautics
                           | University of Southampton
                           | SOUTHAMPTON SO9 5NH
                           | Tel: (0703) 592854
                           | Fax: (0703) 593058
                           | Email: fraser@uk.ac.southampton.aero

Date: Sun, 22 Aug 93 12:41:08 EDT
From: Kevin Dowling <nivek@frc2.frc.ri.cmu.edu>
To: rsx@Cs.Nott.AC.UK
Subject: Tessellator
Reply-To: Kevin Dowling <nivek@cmu.edu>

Richard,

Thanks for your message. I'll send a photo of the system in it's
current mobile base configuration. The manipulator is not yet aboard,
but should be on within several weeks.

My plan is to finish it this fall and send it to NASA for final
certification. Things are going slow but well.

BTW, I've been putting a study of European space robotics together and
a group of space robotics people from the US will be traveling to
Europe in the late fall.

I have Nottingham Trent University as one site to visit with an S.X.
Godber as the contact. Do you know of relevant work in this area both
at Nottingham as well in the rest of Britain?

Other UK groups I have include:
	AI Applications Institute (Edinburgh)
	Advanced Robotics Research Ltd (ARRL)
	British Aerospace PLC 
	BA Sowerby Research Center 
	BA Space Division AI Group (Bristol)
	BA Space Systems (Herfordshire)
	British Robot Association (Birmingham)
	DRA Aerospace Division (Farnborough)
	JET (Abington)
	Kings College (London)
	Logica Cambridge Ltd. (Cambridge)
	National Advanced Robotics Research Centre (Salford)
	Oxford
	Rutherford Appleton Labs (Surrey)
	University of Edinburgh
	University of Reading
	University of Salford
	Vega Space Systems (Herts)

Any input would be appreciated.

				nivek


Date: Tue, 24 Aug 93 10:26:02 EDT
From: Kevin Dowling <nivek@frc2.frc.ri.cmu.edu>
Message-Id: <9308241426.AA09747@FRC2.FRC.RI.CMU.EDU>
Received: by scythe.frc.ri.cmu.edu (4.1/SMI-4.0)
	id AA11015; Tue, 24 Aug 93 10:30:07 EDT
To: rsx@Cs.Nott.AC.UK
In-Reply-To: <199308241121.HAA14069@cmu2.cs.cmu.edu> (rsx@Cs.Nott.AC.UK)
Subject: Re: European Space Trip
Reply-To: Kevin Dowling <nivek@cmu.edu>

Here it is:

aka: Kevin Dowling			Carnegie Mellon University
tel: (412) 268-8830			The Robotics Institute
adr: nivek@ri.cmu.edu			Pittsburgh, PA 15213

thanks,

					nivek


To: rsx@Cs.Nott.AC.UK
Subject: qreplies
Date: Fri, 30 Jul 93 13:30:37 -0400
From: Todd_Jochem@ius4.ius.cs.cmu.edu

1.) Wheels
2.) Up to 55mph
3.) Autonomously using a wide variety of road following and cross
country systems.
4.) Drive on roads or rough terrain; avoids obstacles.
5.) We like to think it has been pretty successful at what it does.
6.) Yes, In the US there is a program called Intelligent 
Vehicle/Highways systems which we a targeting as a possible 
commercial market.
7.) C and C++
8.) The researchers in the group heare at CMU are Chuck Thorpe, Dean
Pomerlea, Martial Hebert, and Tony Stentz. They supervize about 12-15
graduate students including myself.

--Todd

Date: Fri, 30 Jul 93 12:45:06 PDT
From: Chuck McManis <Chuck.McManis@eng.sun.com>
To: rsx@Cs.Nott.AC.UK
Subject: Re: qreplies - Mobile Robots
>    | Answer 1:
>    | ---------
>    | Tracks.
>    | Answer 2:
>    | ---------
>    | PWM speed control with 16 "speeds" although only 8 produce enough
>    | torque to move the robot.
>    |
>    | Answer 3:
>    | ---------
>    | An onboard multiprocessor control system.
>    |
>    | Answer 4:
>    | ---------
>    | Turn shiny new batterys into hard to dispose waste material.
>    |
>    | Answer 5:
>    | ---------
>    | Quite, I believe I have converted anywhere from 10 to 20 kilos
>    | of batteries.
>    |
>    | Answer 6:
>    | ---------
>    | The goal is to have a device capable for fetching a beverage on
>    | demand.
>    |
>    | Answer 7:
>    | ---------
>    | Mostly assembler, some C. 
>    |
>    | Answer 8:
>    | --------- 
>    | This is the fifth incarnation of the "Clyde" series, the previous
>    | version used wheels and a castor and had less sophisticatd logic.
>    | The primary mechanical problem is constructing custom machines without
>    | access to a machine shop. The primary sensor problem is vision sensors
>    | The primary logic problem is navigating amongst my 3 year olds toys.
>    |									  
>    | Answer 9:
>    | ---------
>    |									  
>    | None yet.
>    |									  
>    |*********************************************************************
-- 
--Chuck McManis			    Mr. NIS+			    Sunsoft
These opinions are my own and no one elses, but you knew that didn't you.

Date: Sat, 31 Jul 93 21:22 BST
From: Rich Walker <grlw1@central-unix-service.cambridge.ac.uk>
To: rsx@Cs.Nott.AC.UK
Subject: qreplies - Mobile Robots - comp.robotics #3876

As the only member of the Shadow project who's on-line:

In article <1993Jul30.155413.5612@cs.nott.ac.uk>, you ask:

(Ahem: Answer 1:)
 Main robot: biped.

 Others: wheeled (tho' we did knonk up a hexapod one day for fun)

|>     | Answer 2:
|>     | ---------
Biped: Slow lumber; when balancing works, should get slow dance even. Come and 
see it at Turing...
|>     | 
|>     | Answer 3:
|>     | ---------
|>     |
Currently, almost all our robots have off-board Archimedes.
|>     |
|>     | Answer 4:
|>     | ---------
|>     |
Well, the biped walks. The cart we're doing should do a shakey. We've also done
various cup-collectors, navigators, &c
|>     |
|>     | Answer 5:
|>     | ---------
Well, biped walks with overhead beam; should balance soon. Others: if they're
finished, they work.
|>     |
|>     | Answer 6:
|>     | ---------
Domestic robot.
|>     | 
|>     | Answer 7:
|>     | ---------
Archimedes BASIC, now some Modula-3
|>     |
|>     | Answer 8:
|>     | --------- 
Everything necessary for (6) I'm afraid
-- 
--- Rich! Walker (grlw1@cus.cam.ac.uk)
"From what book is the phrase 'Dragon Murmerings' taken?"
"I don't know, but all who hear it die..."
"So far..."

Further info:


The Shadow Project, a Brief Synopsis. (0.2)

[Disclaimer: Any insulting references are entirely due to lack of knowledge;
corrections gratefully received]
[Second Disclaimer: No-one else in the Shadow Project has verified this lot;
it's all off the top of my head, and so may well be incorrect]
[Note: this text is (c) Rich Walker 26 Jul 1993. redistribute freely
electronically but by other media only with permission]

The origins of the Shadow Project are murky, and lost in the depths of history.
Many things are muttered of, and few things are known for sure. Yes, I wasn't
around then. From what I can gather, two people came together, Richard
Greenhill and David Buckley; the first organised, and the second designed. As
time passed, others agglomerated, and there formed a definite group, wherein
the whole was more than the sum of its parts.

What was the plan of the Shadow Project? To build a proper domestic robot [1].

And how was this to be achieved? What was required? Much thought followed, and
eventually, the First Law Of Domestic Robotics was deduced:
     Human houses are designed for humans to use them.
From this, it was further deduced that a cart would be of limited utility. [2]
And so, the Plan arose: to construct the Android Domestic Robot...

The state of robotics was glanced at briefly, and these things were noticed:
1. There was a ridiculous amount of work on industrial robots.
2. Work on useful manipulating robots had almost stopped after Shakey[3].
3. There was, however, a lot of work on carts moving in environments.
4. Vision was becoming well-explored.
5. Natural Language Processing was one of those things that someone else was
   going to do, because of the sheer amount of other knowledge it required.
   Anyway, building NLP systems wasn't building robots.
6. No-one (almost) was working on bipeds. Probably because they're a complete
   sod to build.

And so, the Shadow Project set forth on its long and lonely road (violin music)
to build a useful android. In the traditional human model.

That's the history over and done with. Now, the current status.

We have a biped, and it walks. It doesn't balance (yet: we're working on that
this summer) but it stands up on its own, and it walks. It's ripped-off from
the best biped model we know of: that of the human being; it has an
almost-standard musculature straight out of Gray.

To do muscles: we produced muscles... Splendid things, running on compressed
air; the resulting hissing and clanking would nearly impress the BBC Sound
Effects guys who did Marvin...

We also did some assorted carts.
For example: one, with software running on a BBC micro (that's a 2MHz 6502,
32k RAM to you lot) could do face-recognition.

Another wanders around an arena, finding & collecting cups, and avoiding walls.
And it retrains itself to avoid them better if it bumps into them...

Then there's a plethora of light-following variants.

And we've got a couple of people building us interesting things; we should get
some hands soon, based on our muscles.

We actually independently invented light-striping: one of those classic `Hey
I've got a really good idea' ... putter putter putter ... `Oh, did you know xxx
did that in yyy?'

Software: We're quite into neural nets as control mechanisms, so we've got a
neural net compiler (net description-> BASIC program[4]...) that we use as
standard. We've got some pretty good vision systems running at various times,
and we've got an abominable problem with old code we can't reuse...

AI: Well, we're working on bits of this. We have an interesting idea for the
knowledge base, and we've got some stuff for that going. There's a planner
coming up, and this should also be running on a cart we're doing for
experimentation by the end of the summer (Shakey II - The Next Generation, or
something like that)

We expect to be able to do things like NLP fairly easily, because by the time
we really need them, they'll be more mature technology...

Other stuff: We also sit around and ponder implications and complexities quite
a lot. We're probably going to end up producing some papers on the implications
of domestic robots, because the set of implications is huge [5]. We're also
pondering things like: training them, getting them to be _safe_, (and we mean
seriously safe, not just complying to any of the current sets of safety
regulations) getting people to accept them, and so forth...

Long-term thoughts: This is a long-term project. We talk about 20 years.
We currently have no funding, because
1. it's so long term
2. we're nobody, in an academic or industrial sense...
This is, to a point, an advantage: it forces us to think, and it means we have
no externally-imposed deadlines.
We're open. If you want to join us, you can; we're happy to have anyone and
everyone aboard. We're a little bit paranoid, and a little bit mad, but that's
not important. Because we're not connected to anyone, anyone can work with
us...
We're getting there. Slowly.

Interested in joining? If you're in the London area, the Shadow Project can be
contacted via Richard Greenhill (071 609 8522 after 6pm, 071 700 2487 before)
We tend to meet on Wednesday evenings, to putter at things and generally make
slow progress.

Sadly, we're on paper-mail only at the moment:
  The Shadow Project
  357 Liverpool Road
  London
  N1 1NL
should do the trick...


-------------------------------------------------------------------------------
Footnotes:

1: Robotics terminology is shoddy. Herein:
Domestic robot: something that replaces butler, cook, maid, valet, and a spare
    pair of hands.
Industrial robot: something that performs a precise sequence of actions, and
    tends to require reprogramming between tasks.
Android: something designed to closely resemble a human being.
Cart: a robot that is not an android.

2: According to the 1LoDR, houses have stairs. If you've seen books on building
   robots for houses, this is where things get very messy. Often referred to
   as the Dalek Problem: the Daleks could conquer the Universe, but they
   couldn't chase the Doctor up a flight of stairs... Eventually, they managed
   it: a mystical blue glow came from their base, and they floated up...

3: Shakey the Robot: first interesting Domestic Robot. Took several days to
   construct a plan for moving some blocks in a slightly intricate manner, and
   to execute this plan. Although, it did cope with a change in the world
   during the plan. (Putting others work down? Who, me?)

4: At the moment, we tend to use Acorn Archimedes machines. These run PDF
   anyway, and the BASIC on them runs outrageously fast. Partly because it was
   written by a mad hacker, who had a hand in the design of the CPU... partly
   because it uses 40-bit floating point, and chucks IEEE out of the window...
   Thus, it's practical to use BASIC for nets: an example, 2048-pixel input, 3
   layer, 7-output net to control the cup-collecter mentioned. Averaging 2
   cycles a second, w'out optimisation; a back-propagate halved that speed...

5: See Appendix H for further details.
   (ie well, I'd like to tell you, but I'm not sure you're the right sort of
   person :->)
   More seriously, the implications _are_ vast, and we really don't like to
   talk about them much, because either you alarm someone, or you fall flat on
   your face when someone points out a flaw in your wonderful dream... Neither
   is very pleasant.

Date: Sun, 1 Aug 93 00:12:49 -0700
To: Richard Seldon <rsx@Cs.Nott.AC.UK>
From: Chuck Rice <carice@netcom.com>
Subject: Re: qreplies - Mobile Robots

>    | Answer 1:
>    | ---------
       Two RC Car wheels powered by 2 surplus 12 volt motors 
>    |
>    | Answer 2:
>    | ---------
        It is only partially completed, but using an 8volt lead-acid 
        battery, it will move 6 to 8 inches a second. 

>    | Answer 3:
>    | ---------
        So far, only tethered, but I have a mini-board now, and am 
        getting a 6.270 and plan to make it atonomous.
>    |
>    | Answer 4:
>    | ---------
        Nothing so far, but teach me.
>    |
>    | Answer 5:
>    | ---------
        I have only been working on it for a month or so, so it 
        is a little early to say, but if you include the enjoyment
        derived, it has been a great success!
>    |
>    | Answer 6:
>    | ---------
        I have many, some more marketable than others. But I want
        to try them befor I give them to others. 
>    |
>    | Answer 7:
>    | ---------
        So far, assembler. (I am an IBM 370 assembler programmer by 
        trade) but I may give interactive C a try since I need to 
        learn C.

>    | Answer 8:
>    | --------- 
        I have started my son (10 years old) with the LEGO TECHNIC stuff,
        and have just ordered a dacta set to add to his collection. He has
        already built the example single moter car and a motorized gripper 
        arm. I plan on giving the mini-board to him. He attended the
        Homebrew Robotics Club meeting with me and was WOWed by it. 

Chuck Rice                                            carice@netcom.com

From: Greg Kulosa <gkulosa@oberon.com>
Date: Mon, 2 Aug 93 14:35:49 EDT
To: Richard Seldon <rsx@Cs.Nott.AC.UK>
Subject: Re: qreplies - Mobile Robots

|>     | Answer 1:
|>     | ---------
2 wheels, one on each side near the front.  There is a caster wheel in
the back center.  It steers tank-style.

|>     | Answer 2:
|>     | ---------

About 2 inches/second

|>     | Answer 3:
|>     | ---------
On board CPU (68HC11)

|>     | Answer 4:
|>     | ---------
Wanders around the room, bumping into things and backing away.  He
will also head for the brightest light around (follows a flashlight).

|>     | Answer 5:
|>     | ---------

Very.  He is very successful at avoiding objects and wamdering around
:-)

|>     | Answer 6:
|>     | ---------
None.  He is my first working robot.  

|>     | Answer 7:
|>     | ---------
He is programmed in C.  More specifically, I use the "Interactive-C"
language written by Randy Sargent and Fred Martin of MIT.

|>     | Answer 8:
|>     | --------- 
Body is constructed of LEGO.  Brain is an MIT 6.270 board.

-- 
Greg A. Kulosa		|  "If we are to be damned, let's be damned for what
Systems Administrator	|   we really are"    - Jean-Luc Picard

Date: Wed, 4 Aug 93 09:06:12 +0200
From: Sjur Jonas Vestli <vestli@ifr.ethz.ch>
To: rsx@Cs.Nott.AC.UK
Subject: qreplies


> |>     | Answer 1:
> |>     | ---------
Three wheels in, one of them (thre front wheel) being steered and driven,
the others passive. The wheels have independant supension and air filled
rubber tires. (Robin Reliant configuration)

> |>     | Answer 2:
> |>     | ---------
Maximum of about 1 m/s.

> |>     | Answer 3:
> |>     | ---------
Autonomous operation with occasional command input.
> |>     |
> |>     | Answer 4:
> |>     | ---------

Intended to distribute internal mail in a building of the Swiss Federal
Institute of Technology. Currently serving as a R&D platform.
> |>     |
> |>     | Answer 5:
> |>     | ---------
We are not that far yet.
> |>     |
> |>     | Answer 6:
> |>     | ---------
See question 4.

> |>     | Answer 7:
> |>     | ---------
Modula-2, in the near future we will port to Oberon-2

> |>     | Answer 9:								IEEE Conference on Robotics and Automation, Atlanta, GA, USA,
2-7/5/93

IEEE Conference on Neural Networks, San Fransisco, CA, USA,
28/3-1/4/93


Date: Wed, 4 Aug 1993 12:12:30 -0500
From: Karl Altenburg <altenbur@plains.nodak.edu>
To: Richard Seldon <rsx@Cs.Nott.AC.UK>
Subject: Re: qreplies - Mobile Robots
:     |
:     | Answer 1:
:     | ---------
:     |	Wheels, differential drive, two wheels powered by the a gear train,
:     |		two wheels tethered by a pulley, (skid turn)
:     | 
:     | Answer 2:
:     | ---------
:     | Approx. 18cm/sec max.

:     | Answer 3:
:     | ---------
:     |	Autonomous, MiniBoard (6811 based microcontroller)
:     |
:     | Answer 4:
:     | ---------
:     |	Used in a cooperative, multiple, mobile, robot system.  The robots
:     | (currently using six of them) are capable of foraging for targets,
	retreiving them and returning them to a home base, they are also
	able to communcate with each other using a recruitment signal.
:     |
:     | Answer 5:
:     | ---------
:     |	So far thing have been pretty successful.  
:     |
:     | Answer 6:
:     | ---------
:     |	Possible applications: havesting of small produce, hazardous waste
:     | clean up, sparse deposit mining
:     |
:     | Answer 7:
:     | ---------
:     | C, Dave Dunfield's Micro-C Cross Compiler for the 6811
:     |
:     | Answer 8:
:     | --------- 
:     | The robots are small, approx. 20cm on a side, contructed of LEGO
:     | Technics, and are equipt with synchronized light proximity sensors,
:     | Modulated IR receivers, a forklift like mechanism with a break beam     
:     | sensor between the forks, photo-sensors for detecting targets and 
:     | IR transmitters and lights used for communication/recruitment.
:     |									  
:     | Answer 9:							  
:     | ---------							  
:     |	Nothing published yet, however, a description of the work will appear 
:     | in the working notes of the IJCAI workshop on Dynamically, Interacting 
:     |	Robot. 
--
Karl R Altenburg				altenbur@plains.NoDak.edu
North Dakota State University, Fargo, ND  58105

All things are artificial, for nature is the art of God.  SIR THOMAS BROWNE

Date: Thu, 5 Aug 93 02:09 PDT
From: "P. Douglas Reeder" <reeder@reed.edu>
To: rsx@Cs.Nott.AC.UK
Subject: qreplies: - Mobile Robots

I have two robots, the walker and Bugdozer

:    | Answer 1:
:    | ---------
Walker: six legs

Bugdozer: differential drive wheels (two powered wheels & a caster)

:    | Answer 2:
:    | ---------
:    |
Walker: statically stable gaits, includings crawls and alternating tripod

Bugdozer: roughly 1m/s
:    | 
:    | Answer 3:
:    | ---------
Walker: offboard microcontroller

Bugdozer onboard microcontrolller

:    | Answer 4:
:    | ---------
Walker: walks

Bugdozer: seeks out & pushes appropriately colored objects toward
humans.

:    | Answer 5:
:    | ---------
Walker: walks very slowly and energy inefficiently

Bugdozer: not yet completed

:    | Answer 6:
:    | ---------
Walker: no

Bugdozer: no, other than as a toy

:    | Answer 7:
:    | ---------
Walker: 68HC11 assembler

Bugdozer: 68HC11 assembler

:    | Answer 8:
:    | --------- 
Walker: no

Bugdozer: no
:    |									  
:    | Answer 9:

reference: "Mobile Robots: Inspiration to Implementation" by Joseph
Jones & Anita M. Flynn.


-- 
Doug Reeder                              Internet: reeder@reed.edu
Div, Grad & Curl                         USENET:   ...!tektronix!reed!reeder
programming & derivative work 
I am actively seeking scientific programming contracts.

Date: Thu, 5 Aug 93 19:08:05 EDT
From: dulimart@cps.msu.edu
To: Richard Seldon <rsx@Cs.Nott.AC.UK>
Subject: qreplies:-

Could you also post your findings to the newsgroup after you compile it.

Thanks,
  Hans.

       | Answer 1: wheels
       | ---------

       | Answer 2: max 1000 mm/sec
       | ---------
       |
       | Answer 3: autonomous
       | ---------
       |
       | Answer 4: indoor navigation, mapping
       | ---------

       | Question 5:
       | -----------
       | How succesful has it been?
       |
       | Answer 5: move between two given locations in a long hallway.
       | ---------
       |
       | Answer 7: C, C++
       | ---------
       |

Date: Fri, 6 Aug 93 14:42:47 -0400
From: Dave Novick <dkn@cimar.me.ufl.edu>
To: rsx@Cs.Nott.AC.UK
Subject: qreplies: Mobile Robot Questionaire

--------------------------------------------------------------
    /|     |___|     |\       
   / |     |   |     | \     
  /  |     |   |     |  \    David Novick
 /    \___/     \___/    \   300 MEB, CIMAR
|                         |  Univ of Florida
|         DAVEMAN         |  Gainesville, Fl 32611
 \     _ __     __ _     /
  \   / |  |   |  | \   /    e-mail: dkn@cimar.me.ufl.edu
   \ |      \ /      | /
    \|       |       |/      Center for Intelligent
                             Machines and Robotics
--------------------------------------------------------------
Date: Fri, 6 Aug 93 14:50:18 -0400
From: Dave Novick <dkn@cimar.me.ufl.edu>
To: rsx@Cs.Nott.AC.UK
Subject: qreplies: Mobile Robot Questionaire

    | Answer 1:
    | ---------
    |  The robot uses wheels for locomotion.
    |
    | Answer 2:
    | ---------
    |  20 mph under manual operation
    |   7 mph under autoonmous operation
    |
    | Answer 3:
    | ---------
    |  Teleoperated, autonomous operation, or  radio controlled
    |
    | Answer 4:
    | ---------
    |  It is a navigation test vehicle.  It's currently plans paths, follows
    |  paths, estimates its current position, and proforms obstacle avoidance.
    |
    | Answer 5:
    | ---------
    |  Can reach specified goal position, accurate to within 1 meter.
    |
    | Answer 6:
    | ---------
    |  The purpose of this project, funded by the U.S. Air Force, is to develop
    |  a navigation system for an excavator that is capable of autonomously
    |  detectiong and removing unexploded buried  munitions.
    |
    | Answer 7:
    | ---------
    |  C is used for the programs that run on the onboard computer.
    |  C and C++ are used for the simulation programs.
    | 
    | Answer 8:
    | --------- 
    |  Ongoing research includes adding vision, radar and possiably infrared
    |  sensors to the test vehicle.  Currently the onboard sensors include
    |  16 sonar sensors, and Inertial Navigation System (INS) and a Global
    |  Positioning Satalites (GPS)

    | Answer 9:								  
    | ---------								  
    |  Crane, C.D. and Nease, A.  "Navigation of an Autonomous Vehicle,"
    |  Proceedings of the Fifth Topical Meeting on Robotics and Remote Systems,
    |  Americal Nuclear Society, Knoxville, TN April 1993
    |									  
    |*********************************************************************
 
P.S.  When your compilation is complete, I would like a copy of it.  Thanks!

--------------------------------------------------------------
    /|     |___|     |\       
   / |     |   |     | \     
  /  |     |   |     |  \    David Novick
 /    \___/     \___/    \   300 MEB, CIMAR
|                         |  Univ of Florida
|         DAVEMAN         |  Gainesville, Fl 32611
 \     _ __     __ _     /
  \   / |  |   |  | \   /    e-mail: dkn@cimar.me.ufl.edu
   \ |      \ /      | /
    \|       |       |/      Center for Intelligent
                             Machines and Robotics
--------------------------------------------------------------

To: Richard Seldon <rsx@Cs.Nott.AC.UK>
Subject: Re: qreplies - Mobile Robots
Cc: callv@essex.ac.uk


Answer 1: Wheels

Answer 2: up to 1m/sec

Answer 3: 4 modes: autonomous, collaborative (DAI), central  
supervisor, semi-autonomous (ie telepresence)

Answer 4: Distributed AI work

Answer 5: It has meet our expectations

Answer 6: Yes

Answer 7: Our own software architecture based on developments on C  
and VxWorks

Answer 8: no (but maybe I don't really understand this question)

published (if applicable) or any references you feel would be useful.         

Answer 9: I will try to advise you later								  

If you email Paul Chernett (Internet: cherp@essex.ac.uk) he can  
arrange to send you some GIF (assuming this is your preferred format)  
images of our robots. Perhaps you might want to wait until you are  
closer to writing your report as we may have a greater variety of  
photos/images by then.

Good luck with your project,

Vic Callaghan

PS You may be also interested to know we run a small UK ftp robot  
archive from Essex. You should be able to access it at the following  
address ftp.essex.ac.uk (in directory pub/robots). We also have a  
non-public archive we with various internal docs (mostly plans of the  
robot lab here at Essex, its role in teaching research etc). If  
required, we might be able to pass you information from this.


Date: Fri, 13 Aug 1993 10:42:07 +0100
From: Philippe DARCHE <darche@masi.ibp.fr>
To: rsx@Cs.Nott.AC.UK
Subject: qreplies

Mobile robots questionaire

Hello,

I send you some answers for your questionnaire. In the same time, I send you 
by post a copy of a paper which describes the project. I have made today some 
photographs and when I have them, I send you someone.

answer 1 :
It's a mobile robot

answer 2 :
speed : 33 cm/s
wheight : 10-12 Kg
diameter : 33 cm
height : 17 cm

answer 3 :
It's an autonomous robot from the point of view of communication and energy. 
The first step is to control the robot with the master/slave scheme. After 
some experimentations, we want a total autonomy.

answer 4 :
This robot is a device of a network of robotic actors which is a tool for 
learning the notions of parallelism for such kinds of audience as scholls, 
colleges and training centers for adults.

answer 5 : 
Mechanics are made, electronics devices is in testbed and the development of 
software will begin in september. It's a turtle like robot. You have a pencil 
in the geometrical centre and six collision sensors (Hall effect).

answer 6 :
We want to go towards IAD with reactive agents (like Brooks) and one 
application is the surveillance of rooms.

answer 7 :
We are using transputers and the development language is Meta-Vlisp with 
assembler, perhaps we shall use Occam.

answer 9 :
Bibliography :

[Darche 91] Ph. Darche. "Reseau d'Acteurs Robotiques pour l'Apprentissage du 
Parallelisme". Proceedings of the 3rd international Mexico conference on 
pedagogic roboMexico, 1991.

[Darche - Nowak 92] Ph. Darche, G. Nowak. "ActNet : A Network of Actors based 
on Transputers for Apprenticeship of Parallelism". Proceedings of Transputers'
92 (Advanced Research and Industrial Applicatpages 293-306. IOS Press, ISBN 90
5199 081 2. Arc-et-Senans, 1992.

[Darche 93]  Ph. Darche - "L'acteur robotique - Etude et proposition 
dinterfacegenerique de controle". Actes de la 4e conference internationale 
sur la robotique pedagogique. L1993.

[Darche - Nowak 93] Ph. Darche, G. Nowak. "ActNet : A Heterogeneous Network of 
Actors for Learning of Parallelism, Communication and Synchronization". NATO 
Series F : Computer and Systems Sciences - Springer-Verlag. Liege, 1993. A 
paraitre.

best Regards.

Philippe Darche


            _\\|//_
           (' O-O ')
----------ooO-(_)-Ooo--------------------------------------------------
Universite Paris VI
Institut Blaise Pascal
Laboratoire MASI
M. Darche Philippe
couloir 65/66 - bureau 206
4, Place Jussieu                  
75252 Paris cedex 05              
FRANCE
                           
Email           : darche@masi.ibp.fr | 
Telephone       : 33-1-44-27-34-23   | "excuse for my french english !"
Fax             : 33-1-44-27-62-86   |       me.
-----------------------------------------------------------------------

Date: Thu, 12 Aug 93 00:30:29 EDT
From: Kevin Dowling <nivek@frc2.frc.ri.cmu.edu>
To: rsx@Cs.Nott.AC.UK
Subject: survey

I have responded with a ahlf-dozen of our mobile robots. We have
another half-dozen or so.

					nivek

    | Answer 1:
    | ---------
	Ambler - Legs (hexapod, orthogonal leg, circulating gait)
	Tessellator - Wheels (four mecanum)
	Daedalus - Legs (framewalker, orthogonal leg)
	Terregator - Wheels (six-wheeled skid steer)
	Navlabs I and II - Wheels - (four wheel ackerman steering)


	Ambler - 1m/min over very rugged terrain
	Tessellator - 10m/min flat floor
	Daedalus - 5m/min (benign terrain)
	Terregator - 10m/min (outdoors)
	Navlabs I and II, 500m/min and 1600m/min respectively (highways)

    | Question 3:
    | -----------
	Ambler - autonomous
	Tessellator - supervised autonomy
	Daedalus - autonomous
	Terregator - autonomous
	Navlabs I and II - autonomous

    |
    | Question 4:
    | -----------

	Ambler - rougn barren terrain exploration
	Tessellator - service space shuttle
	Daedalus - long duration rough terrain exploration
	Terregator - testbed for early outdoor and mine navigation
	Navlabs I and II - high speed unmanned navigation


    | Question 5:
    | -----------
	Ambler - longest rough terrain navigation 0.5km
	Tessellator - not in service yet
	Daedalus - under construction
	Terregator - work directly spawned large and long reaching
	programs in unmanned vehicles including commerical projects
	Navlabs - cornerstone of ARPA navigation work. highly visible
and sucessful runs - 90mi at average speeds over 100km/hr

    | Question 6:
    | -----------
	Very much so. Mining, Agriculture, automoous aircraft landing,
haulage, construction, storage tank inspection, exploration of
planetary surfaces, nuclear and hazmat remediation, and many
others. Some have already occurred and several are in process.
Spin-off company has brought a number of machines into commercial
application.

    | Question 7:
    | -----------
	All - C, C++ and small amount of Lisp

    | Question 9:							  
    | -----------							  
	See IEEE R&A, PAMI, IJRA and many conference proceedings.
 

For slides of robots send email to mjd@ri.cmu.edu. Photos are supplied
for nominal cost.

						nivek

Date: Sun, 15 Aug 93 16:19:11 +0200
From: jim crowley <jlc@castor.imag.fr>
To: rsx@Cs.Nott.AC.UK
Subject: qreplies - Mobile Robots
Organization: Institut Imag, Grenoble, France

>    | Answer 1:
>    | ---------
	wheels (4 of them, 2 power, 2 "casters")
>    | 
>    | Answer 2:
>    | ---------
	0.01 m/s = 2 m/sec

>    | Answer 3:
>    | ---------
	Autonomous navigation

>    | Answer 4:
>    | ---------
	navigates autonomously (for surveilance)

>    | Answer 5:
>    | ---------
	reasonably. (read our papers)
>    |
>    | Answer 6:
>    | ---------
	security guard, factory transport, hospital transport, 
	floor cleaning
>    |
>    | Answer 7:
>    | ---------
	C (high level is CLIPS 5.1)

>    | Answer 9:								  
>    | ---------								  
	IEEE Trans on Robotics and Automation
	Journal of Robotics and Autonomous Systems

I will see what I can send by email. Post Script is easy enough.

Date: Mon, 16 Aug 1993 16:37:17 +0200
From: Martin Frischherz <frisch@zams.kapsch.co.at>
To: Richard Seldon <rsx@Cs.Nott.AC.UK>
Subject: Re: qreplies - Mobile Robots

:     | Answer 1:
:     | ---------
wheels. it is based on a rc model car

:     | Answer 2:
:     | ---------
depends on the modelcar
:     | 
:     | Answer 3:
:     | ---------
it is fully autonomous
:     |
:     | Answer 4:
:     | ---------
it  is searching a glass of jam located anywhere in the room

:     | Answer 5:
:     | ---------
it worked fine after one week

:     | Answer 6:
:     | ---------
i build it to have fun during my last sick-leave two weeks ago
:     |
:     | Answer 7:
:     | ---------
8051 assembler with my self written single-chip-mode-multitasking-system
:     |
:     | Answer 8:
:     | --------- 
the jam-glass is prepared with an IR-transmitter. The car is using a
SIEMENS IR-detector with extra focussing lens to get about 5 degrees
viewing angle and long distance capability. 
This is mounted on the top of a model-servo to be turned
continuously left and right searching the jam. Three bumpers are
available for sensing wall/chair... contact. I tried to implement some
kind of brooks subsumption architecture as described in _Mobile Robots_.
My opinion: It looks very technical
My wifes opinion: It looks very ugly
My girls would like to eat the jam :-)
Total costs: about 80$

:     | Answer 9:								  
None

If you are really interested in such a useless robot, feel free to send
me a note for further info/pictures.

| Martin Frischherz   :)-|-|-< -----------------------------------------------
| email: martin.frischherz@kapsch.co.at             | phone: ... hated ringing
| PSImail: PSI%(0232)281001141::martin.frischherz   |   fax: ... wasted  paper
|                           "I mean it must be HIGH or LOW" - John Lennon

Date:     Mon, 23 Aug 93 20:07:25 BST
From:     Tom Scutt <tws@psychology.nottingham.ac.uk>
To:       rsx@Cs.Nott.AC.UK
Subject:  qreplies

Answers to questionnaire:

1. Wheels (2 powered, 1 'free' wheel - That's 3 altogether)
2. Not sure... about 1 m/sec tops.
3. Connected by wire to external power source. Everything else is onboard
4. The robot either seeks or avoids light depending on the setting of onboard
   switches. During this process it avoids objects in its path.
5. Very succesful at proving a point - that complex behaviour can be produced
   by very simple architectures. The robot cost under 50 pounds (over 40 pounds
   of this was for the body, the electronics cost less than ten pounds), and 
   yet has outperformed a robot costing > 3000 pounds at avoiding objects.
6. Yes. Firstly the cheapness lends itself to building robots for the toy 
   market that have interesting behaviours. Secondly, the simple architecture 
   could be used to provide "thinking time" for a more classical planning 
   robot.
7. The robot is hard-wired from simple components. The only way in which the
   architecture can be changed is by two switches which determine whether
   connections between the light sensors and the motors are crossed or straight
   and whether they excite or inhibit the motors.
8. The simple analogue circuit can be thought of as a four neuron network - two
   sensory input neurons and two motor output neurons. We are working on 
   connecting the robot up to a PC-based neural network program.
9. I have not published anything on the robot, but do have some write ups on it
   which I can supply if you are interested. I also have a large amount of 
   literature on mobile robots (mainly from Brooks and various research groups
   at Edinburgh) which I can lend you if you want - time is too short to list
   them here. The robot we are using was inspired by V. Braitenberg's (1984)
   book "Vehicles".

Cheers, Tom.




















-- 
|-------------------------------------------------------------------|
|  Richard L. Seldon		     |  e-mail: rsx@cs.nott.ac.uk   |
|  I.T.C.R.			     ===============================|
|  Cripps Computer Centre (North)				    |
