|
|
Post by donburch on Aug 8, 2006 22:10:44 GMT 10
I have noticed quite a few websites describing autonomous wheeled robot bases advocate using servos modified for continuous turning.
In my very limited understanding, a servo is a motor with electronics (to accept digital (PWM) pulses, and turn them into fairly precise amounts of motor movement) and mechanical stops to prevent more than 180degrees movement.
So then we modify the servo by removing the mechanical stops and disabling the feedback part of the electronics ... turning it effectively into a stepper motor.
I get the impression that H-bridge is the electronic circuit to control a plain motor. I haven't looked into this yet, so please excuse my flawed knowledge.
So my question is ... for an autonomous robot which will be controlled by sensors (so presumably doesn't require precise motor control), is a modified servo, stepper, or H-bridge and motor going to be more economical ? Is there any other significant advantage to the modified servo (or other method) ?
|
|
|
|
Post by Bones on Aug 9, 2006 0:34:02 GMT 10
Hi Don,
Thje only problem with a modified servo is that it's flatout forward or flatout backward or stop. But good if your using a Serial Servo Controller to contol other parts of the robot.
A H bridge is basicly 4 transistors that can supply either a + or a - to the wiring on a motor. So you can go back, forward, stop and short the terminals on the motor for Dynamic braking.
Micro controller can have 2 PWM channels. Run them into H bridges then you can slowdown or speedup and change direction.
You just send the speed number in (say 0-255) and it hold that speed. Fully adjustable.
Bones
|
|
|
|
Post by Dingo on Aug 9, 2006 18:41:44 GMT 10
If you are going to use modified servos, I'd probably go for servos with strong gears than normal (those Hi-tec brass ones over the std nylon) as they aren't designed for continuous use. Some servos are only rated for tens of hours of operation but they should go longer than that.
A modified servo is not like a stepper motor. A stepper motor's internal magnet configuration means it can only move in discrete steps. A modified servo's internal motor is just a DC brushed motor and so is not limited to discrete steps - it is the feedback circuitry that makes a servo move to a specific point, not the internal configiuation of the motor.
An H bridge is to control normal DC motors, not stepper motors.
A "normal" DC motor is a good idea for a bot, the best idea is to use an encoder wheel of some kind (usually IR) to measure the actual revolutions it has turned to then know if the wheels are actually turning when power is applied..... and how far.
|
|
|
|
Post by donburch on Aug 11, 2006 7:43:37 GMT 10
Yes, it was the comments that modified servos are not designed for continuous operation and so good for only about 100 hours which stared me wondering why they are advocated so often. The Parallax BoE can connect 4 servos, so they use servos; whereas the Arrick ARobot's controller board provides 4 servos and 2 "powerful output"s so they use a motor (and a servo for steering). I guess it comes down to BoE being a general purpose educational controller (with BoE-bot being but one of many "experiments") vs. ARobot is designed with heavy duty components specifically as a mobile robot. I do however prefer the idea of controlling 2 drive motors directly, rather than 1 drive motor and one steering servo, because I expect that it gives more control over the unit with more accurate movements and minimal turning circle. At RoboCup Junior competition yesterday I noticed some of the line following robots jerking wildly back and forth; and wondered if it was (a) lack of fine motor control, (b) using only one sensor, or (c) inexperienced programming. OK, c is a given for primary school students ;-) You may gather that I'm seriously thinking of the ARobot to get mobile - but I just can't budget A$459 for what is essentially just a sheet of metal, a motor, a servo and some sensors ... and if Abe Howell can build a robot in a CD case, then can I ? |
|
|
|
Post by emzed on Aug 28, 2006 21:48:48 GMT 10
|
|
|
|
Post by donburch on Aug 28, 2006 22:15:54 GMT 10
It was the Carpet Rover and Pololu base which pushed me into making my own chassis. My Parallax BoE controller will drive modified servos (A$23 each), so I really just want a chassis for attaching the components. I think Carpet Rover looks cool, but what is it ? A sheet of Acrylic plastic with rounded corners, and a couple of right-angle mounting brackets for the motors. And the Pololu base is sold for A$55 for just a flat pre-cut acrylic base without motors. So expensive for something so simple ! So first step was to use a CD as a base, and next I'll get some Acrylic and a Dremmel and make my own bigger chassis. |
|
|
|
Post by SDFSD on Aug 14, 2008 17:59:46 GMT 10
|
|
|
|
Post by FSDFS on Aug 14, 2008 18:01:17 GMT 10
|
|
|
|
Post by FSDDSF on Aug 14, 2008 18:03:04 GMT 10
|
|
|
|
Post by SDFSDF on Aug 14, 2008 18:04:29 GMT 10
I also know that when that decision is made you want to be on the road to a training program as soon as possible. free online games.Just remember that choosing a Training program for yourself and the dog is as important as choosing. play war games.a college. free online war games.You will need to know a lot about. online games.the program and what you really need from a specific Guide Dog school. |
|
