Telerobot

I'm need to design and construct a Micro controller based robot which can be operated remotely. The robot must be designed to obey a few command like walk, pick, place and stop. It must be an independent system without PC. Note that walk means using legs(not wheels).

Anyone have better idea here?

ongjie said:

I'm need to design and construct a Micro controller based robot which can be operated remotely. The robot must be designed to obey a few command like walk, pick, place and stop. It must be an independent system without PC. Note that walk means using legs(not wheels).

Anyone have better idea here?

Click to expand...

Could you elaborate? You first want to use a microcontroller, then you say it must be an independent system without PC. These two concepts seem to be in conflict. You seem to have chosen the processing engine for your project without clearly defining the requirements. We have no idea of the mechanical scale. Walk, pick, place, and stop don't really tell us very much. The final solution may involve a microcontroller, multiple micro controllers, a PC, a network of PC's. How can we help you if you can't clearly define the requirements.

Agree, i really want to help and participate, but i am not quite sure what do you mean...

Can you afford to control it with an IR remote control?

Or you want somthing like voice recognintion system?

Make an initial design, then Try to ask specific questions so we can help..

good luck.

I plan to design a spider like robot which has more than 6 legs. Operate remotely with RF module. This design will use PIC16F877A microcontroller. It must be able to carry or pick something using servo gripper. How many legs it will be more stable to carry something?

I'd guess the intent is to have a self-powered, locally controlled robot with high level commands being sent over RF.

Look around for "hexapods" - that's the usual 6 legged configuration with 2 servo's per leg. The gait is pretty standard - lift 3 legs, move, switch legs. Using more legs probably won't help too much.

THe more legs the more stability, but 6 or 8 is the number that is used for stability because it costs too much and is too power hungry to get more to get more. 8 is already pushing it. THe only practical reason you might use 8 legs is if you wanted to combine the front two legs to also be arms or grippers. Then when carrying something you would have a stable 6 legged robot. If you turned a 6 legged robot's front two legs to also be arms , you would only have a 4 legged robot when carrying something.

Is this a class project? Because walking robots aren't cheap...

IR remote control is cheap though. YOu can get those keychain things that have a few buttons on it with an IR PCB+header for $30 or less. A header pin on the PCB just goes from low to high whenever you click a certain button on the keychain.
**broken link removed**

EDIT: THis isn't IR, but I bet it works better anyways since you don't need line of sight. You could also use RF transmitter used for remote control planes or helicopters. Have you seen the book Insectronics? It's pretty cheap and would help you out a lot.

I agree, using more legs wont help, using 6 or 8 is a standard and is very sufficient.

how ever i made you this schematic that may help, this is a simple mechanical system to transform rotaional motion into the motion of the legs of a spider, the schematic actually shows one leg only.

All you have to do is try to imagine it running, you will see that in case the GREEN gear is turning, the end of the RED leg will move also in a circle, which - if well sychronised with the other legs - will get your robot mooving..

I hope somebody understand this because lately on this forum, when i try to help someone, people tell me they didn't undersatand what i meant!

here is the same idea in a roboic magazine, this picture show a full working system with 8 legs. sorry if the pic quality is poor, i don't have a scanner.

The other variation of ikalogic's concept is to have the pivot for the leg be a slotted groove. The result is that rather than the leg extending outwards from the body, it remains in the same vertical plane. (More like a dog, cat, cow, deer, camel, or goat than like a spider or ant).

(Just think of the wheels of a steam locomotive and the bars attaching those wheels together).

I'd stay away from servos if it's a school project...servos cost money and you need many of them for a walking robot. You can pull it off with 2 servos actually- 1 servo attached to the beam that forms the middle legs. When the servo turns it pushes down on the right/left and raises the left/right and allows the robot to lift one side off the ground. The second servo (or paralleled servos) controls the front and back legs using a series of beams connected to be a parallelogram. The result is that (front and back legs only, not the middle legs) the left legs will push forward when the right legs push back and vice versa. You can a somewhat shuffling motion.

Look at the robot on he cover of this book:
https://www.amazon.com/gp/reader/0071412417/ref=sib_dp_pt/104-7199346-0205531#reader-link

The two visible servos are paralled together (for greater strength) but they move the front and back legs. The parallelogram set of jointed beams moves the left front-back legs forward and the right front-back legs producing the push forward to walk. There is a servo in the middle of the beam forming the two middle legs and those lift alternating sides of the robot off the ground to raise the front-back legs on a particular side.

Every other design requires at least 2 servos per leg. 6 legs = 12 servos...at least. Add 6 more if you want knee joints.