Robot walk with a specific behaviour - Electronic Circuits Projects Diagrams Free

watzmann · 22nd September 2007, 12:21 AM

Hi All ,

i'm workign in my robot now.....but i have an i dea need to know if it is possible to make or not .

i need my robot to walk in specifc path, this path is a square with length of 1 meter.

i don't like the idea of line tracking , i need to make it without this tracking ground .

so please tell me how to let my robot do that , i'm using PIC 16f877

M.ALi

**Nigel Goodwin** · 22nd September 2007, 09:06 AM

You need some method of tracking it, line following is br FAR the easiest method.

If the circuit is smooth, flat, level, and has a constant friction factor, then you could use stepper motors to give a repeatable path.

watzmann · 22nd September 2007, 03:13 PM

what about using encoder and counter

**Nigel Goodwin** · 22nd September 2007, 03:32 PM

As good as a stepper, and subject to the same terrain limitations - in either case if a tyre slides on the surface at all it won't run straight and accurate.

simrantogether · 24th September 2007, 07:01 AM

the answer is time delay..

for say 3 seconds let the robot move in straight path.. after that it turns right.. again after 3 seconds it turns right.. to move in clockwise direction..

I advice you to choose time yourself so that it accurately travels the side length of 1 meter..

Regards,

Simran..

**Nigel Goodwin** · 24th September 2007, 09:07 AM

Quote:

Originally Posted by simrantogether

the answer is time delay..

for say 3 seconds let the robot move in straight path.. after that it turns right.. again after 3 seconds it turns right.. to move in clockwise direction..

I advice you to choose time yourself so that it accurately travels the side length of 1 meter..

I presume you've never actually done this?.

First problem, it won't always move the same distance during the same time period, second problem, it won't move in a straight line.

Using stepper motors will help greatly with both problems, but even then if a tyre ever slips at all it's then off course.

simrantogether · 24th September 2007, 09:14 AM

Quote:

Originally Posted by Nigel Goodwin

First problem, it won't always move the same distance during the same time period..

yes ... but the error distance will be very small

Quote:

Originally Posted by Nigel Goodwin

second problem, it won't move in a straight line..

it will move in a straight line.. because.. we are using three strepper motors two for tyres on the rear ... and one is mounted on the front.. to change the direction..

Quote:

Originally Posted by Nigel Goodwin

Using stepper motors will help greatly with both problems, but even then if a tyre ever slips at all it's then off course.

we all have seen a wheel chair used in hospitals.. its tyres are available in market easily.. they cannot slip because of bigger size.. when compated with stepper motor..

Regards,

Simran..

**Nigel Goodwin** · 24th September 2007, 09:24 AM

Quote:

Originally Posted by simrantogether

yes ... but the error distance will be very small

Using DC motors it will be considerable, with steppers it will be better, but still might not be good enough - unless it's over a VERY small distance.

Quote:

it will move in a straight line.. because.. we are using three strepper motors two for tyres on the rear ... and one is mounted on the front.. to change the direction..

Again, only over a VERY short distance.

Quote:

we all have seen a wheel chair used in hospitals.. its tyres are available in market easily.. they cannot slip because of bigger size.. when compated with stepper motor..

A wheelchair has feedback, the 'driver' controls it and keeps it straight.

Try locking the controls of an empty wheelchair and send it down a straight corridor - try it multiple times - then see how many times it doesn't hit the walls!

simrantogether · 24th September 2007, 09:35 AM

Quote:

Originally Posted by Nigel Goodwin

Using DC motors it will be considerable, with steppers it will be better, but still might not be good enough - unless it's over a VERY small distance.

yes... it is effective in a small distance about 7-10 metres... but if in a larger distance.. we should use.. servo .. i feel..

Quote:

Originally Posted by Nigel Goodwin

A wheelchair has feedback, the 'driver' controls it and keeps it straight.

Try locking the controls of an empty wheelchair and send it down a straight corridor - try it multiple times - then see how many times it doesn't hit the walls!

i am talking about the small tyres used in a wheelchair.. that you are seeing in front.. can be used with stepper motor..

Regards,

Simran..

**Nigel Goodwin** · 24th September 2007, 09:40 AM

Quote:

Originally Posted by simrantogether

yes... it is effective in a small distance about 7-10 metres... but if in a larger distance.. we should use.. servo .. i feel..

I would call that a LONG distance - even your body isn't that good - close your eyes and try and walk a straight line for 7 - 10 metres!. Make sure you have a friend there to shout and stop you, as you're likely to fall over things, walk across roads etc.

Quote:

i am talking about the small tyres used in a wheelchair.. that you are seeing in front.. can be used with stepper motor..

Those don't go straight even when you push them from behind!

simrantogether · 24th September 2007, 09:50 AM

Quote:

Originally Posted by Nigel Goodwin

I would call that a LONG distance - even your body isn't that good - close your eyes and try and walk a straight line for 7 - 10 metres!. Make sure you have a friend there to shout and stop you, as you're likely to fall over things, walk across roads etc.

ok i agree your point.. but what if tyres are big...

you obtain accuracy.. because less rotations per small distance..

Quote:

Those don't go straight even when you push them from behind!

my senior made it and presented in IIT - DELHI.. he got award..

Ambient · 24th September 2007, 02:15 PM

You could get a compass module, but even those are not 100% accurate, and they are expensive.

The only other thing I can think of is to design a servo system that will lift the front end and move it 90 degrees each time. But you will have a VERY hard time eliminating drifting over time, unless you do a line follower.

If the square is set up in one location then maybe you could get 4 lasers on fixed mounts that the robot can use for a referrence. That might work out pretty good with a stepper motor to help keep the distances consistant.

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar
Title	Starter	Forum	Replies	Latest
laser seeking robot	Syntax	Electronic Projects Design/Ideas/Reviews	6	28th February 2009 08:56 PM
Wireless robot (with a camera)	bloody-orc	Robotics Chat	30	3rd September 2008 12:03 AM
Robot wars!University of Cape town Mechatronics project	riki4eva	Electronic Projects Design/Ideas/Reviews	7	28th June 2008 06:53 AM
Motor Sizing For Moving Robots	dknguyen	Electronic Theory	6	25th April 2007 11:44 PM
Anyway to make my robot know where it is?	NewCurrent	Robotics Chat	3	20th September 2003 07:24 AM

22nd September 2007, 12:21 AM	#1
watzmann	Robot walk with a specific behaviour Hi All , i'm workign in my robot now.....but i have an i dea need to know if it is possible to make or not . i need my robot to walk in specifc path, this path is a square with length of 1 meter. i don't like the idea of line tracking , i need to make it without this tracking ground . so please tell me how to let my robot do that , i'm using PIC 16f877 M.ALi

22nd September 2007, 09:06 AM	#2
Nigel Goodwin	You need some method of tracking it, line following is br FAR the easiest method. If the circuit is smooth, flat, level, and has a constant friction factor, then you could use stepper motors to give a repeatable path. __________________ PIC programmer software, and PIC Tutorials at: http://www.winpicprog.co.uk

22nd September 2007, 03:13 PM	#3
watzmann	what about using encoder and counter

22nd September 2007, 03:32 PM	#4
Nigel Goodwin	As good as a stepper, and subject to the same terrain limitations - in either case if a tyre slides on the surface at all it won't run straight and accurate. __________________ PIC programmer software, and PIC Tutorials at: http://www.winpicprog.co.uk

24th September 2007, 07:01 AM	#5
simrantogether	well.. the answer is time delay.. for say 3 seconds let the robot move in straight path.. after that it turns right.. again after 3 seconds it turns right.. to move in clockwise direction.. I advice you to choose time yourself so that it accurately travels the side length of 1 meter.. Regards, Simran.. __________________ Simran.. 8051 Specialist.. Last edited by simrantogether; 24th September 2007 at 07:05 AM.

24th September 2007, 02:15 PM	#12
Ambient	You could get a compass module, but even those are not 100% accurate, and they are expensive. The only other thing I can think of is to design a servo system that will lift the front end and move it 90 degrees each time. But you will have a VERY hard time eliminating drifting over time, unless you do a line follower. If the square is set up in one location then maybe you could get 4 lasers on fixed mounts that the robot can use for a referrence. That might work out pretty good with a stepper motor to help keep the distances consistant.