DIY : Using Lego Kits

LEGOMindstorms ® NXT ® is a product offered by the world’s fourth largest manufacturer of toys, LEGO. It ships as a kit, which consists of the NXT or the CPU, various components such as axles, gears, and other fabricated plastic parts that make it very easy to assemble and modify your robot quickly. Creating good looking robots really easily seems essential to LEGO’s aim to create as much curiousity as possible in robotics.

However beneath the surface, there is enough to interest the more experienced enthusiast. Controlling the three servo motors is a 32-bit ARM7 Micro-controller that uses  6-wire ports to monitor its plethora of sensors: Light, Colour, Ultrasonic, Touch, and Sound.

The NXT software used to program the bot has a graphical interface, once again showing LEGO’s interest in roping in first timers. And once again, the GUI is only the surface of a powerful programming environment called LabView, developed by the Texas giant, National Instruments.

The same LabView software has been used by scientists and engineers worldwide in a variety of applications from controlling MP3 players to testing the Microsoft XBox, to helping  control the NASA Mars Pathfinder exploration.

All in all, the NXT is a unique product. There’s no denying that it was designed for the hobbyist, but those who can, have used it to make some pretty amazing things. The 280$ price tag can put you off, but if you’re looking for a good robotics kit, or even something to get you started with, then this should be on your list.

DIY : PID For Line Following

At slower speeds, line following is pretty simple – if the sensors say it is going left, steer right and if going right, steer left. This process has its limitations though, mainly when the speed is increased.  This is when a PID controller starts to shine.

PID stands for Proportional, Integral and Derivative.  A PID controller is a mathematics based procedure that processes sensor data and uses it to control the direction (and/ or speed) of a robot to keep it on course. Why does PID work better than our simple model described above? Let’s talk about how robot acts (or behaves) as it follows a line to see why.

Behaviour of a line follower

Let’s say our robot has 3 sensors- Left, Centre and Right. When the Centre sensor sees the line, the robot is programmed to go straight. When the Left (Right) sensor sees the line, the robot is programmed to turn right (left). This will typically cause the robot the wobble back and forth over the line and if going too fast, it may lose control and stop following the line altogether.

Before we go into how exactly PID works, here are some terms you need to know :

Error - The difference between the target position and the measured position the Error.

Proportional - Measures how far your robot is away from the line.

Integral - Measures the accumulated Error over time.

Derivative - Measures the rate at which the robot is moving left-to-right or right-to-left.

The “Control” variable, which helps in precise motion of the bot is essentially a combination of the P, I and D elements with coefficients Kp, Kd and Ki,

In order for your line follower to attain optimum performance, tuning is the most important last step to be followed :

• Kp = 1, Kd = Ki = 0. The goal is to get the robot to follow the line even if it is very wobbly
• Assign Kd = 1. Try increasing this value until you see lesser amount of wobbling
• Once the robot is fairly stable at following the line, assign a value of 0.5 to 1.0 to K_{i .}Since Integral is cumulative, the K_i value has a significant impact. You may end up adjusting it by .01 increments.
• Once the robot is following the line with good accuracy, you can increase the speed and see if it still is able to follow the line.

Keep in mind that you need not implement P, I and D. Only P or even PI could serve your purpose in which case don’t complicate the system further.

In Retrospect

Let us look back at ROBOTIX 2011 from the point of view of a robotics aficionado with a fetish for out – of – the – box solutions to even the minimum of robot tasks. As we dug into our archives, we found a treasure trove of  interesting robots that featured in this year’s edition of the fest. Without a lengthy preamble, lets head straight into technical glory!

RoboCop was designed to be an event that tested a bot’s ability to identify criminals from civilians . A bot actually shot marbles using a hydraulic mechanism based on air pressure at the “criminals”.  In Pirate Bay, some robots had multiple arms for digging through sand and pebble. One even used a sweeping mechanism to traverse through sand!  A stunning bot used multiple underwater propellers with Blue LEDs as illumination in R.A.F.T. Yet another used air propellers instead of water propellers.

The event “The Fugitives” broke new ground with teams of no less than two bots collaborating to corner stationary “fugitives” placed in a 10X10 grid. Wireless communication was used by several participating teams to run the team of bots. Ballista needed bots to detect a moving light source and shoot balls at it which was done with perfection by several teams.

Here are a few of the extraordinary designs which caught our (very observant) eye.

DIY : Working With Servos

Most of us have used DC motors at one point or another. They follow pretty simple rules like speed is directly proportional to voltage and reversing terminals mean reversing the direction of motion. As we move forward from a simple differential drive our actuator requirements also change. This post is to familiarize beginners with servos- a type of electromagnetic actuators that do not rotate continuously like DC motors rather are used to position some object .Servo Motors were extensively used in the events Ballista and The Fugitives at ROBOTIX 2011. Servos prove better then stepper motors because the employ a feedback mechanism.
A servo has three wires coming out of it one for positive supply(red), the other for ground(black) and the last one for control (usually yellow or white).The positive wire is attached to a power supply of 4.8V to 6V. The various Hi Tech servos and their specifications are given here.

A servo motor only rotates 180 degree and has to be lobotomized if it is required to rotate continuously or to a higher angle. Here we shall talk only about non lobotomized servo.
Controlling a servo is easy by using a microcontroller; no motor driver circuit is required. Just a control signal is needed to be feed to the servo to position it in any specified angle. The frequency of the control signal is 50 Hz and the width of positive pulse controls the angle (Pulse width modulation). 20ms time period corresponds to 50HZ.

The voltage is fed directly from the power supply (Rectifier circuit/adapter/battery).
The control signal has to be fed a PWM. We can use the AVR micro controllers PWM feature to control servo motors. In this way the PWM with automatically generate signals to lock servo and the CPU is free to do other tasks.

DIY : Building a robotic arm

“Jacob took that coke in his hand, the Herald919, what he named his other “arm”, made of tinker and junk mimicked the same. As he undid the wrapping, Herald broke the other one.”

Take a look at your hand and now replace (mentally) every moving part of it with metal segments and joints and there you go! you have your very own robotic arm. Usually a robotic arm has six degrees of freedom, is made with over seven segments of metal framework, having over six joints and are designed for repetitive heavy manufacturing work.
For those who pursue robotics as a hobby here’s an easy way to make a simplified robotic arm.

You’ll need thin wooden planks, wires, a solder iron and 4 DC Motors with varying RPMs. Note one of the motors must be a low RPM motor that can help in gripping objects.

The base of the arm is a disk mounted on a motor fit at its center. Now fit a motor to the base with its axis horizontal. The other end of the two planks now hold the third motor which in turn is fit within two more shorter planks. This segment holds your gripping arm at its end. It has the last motor fit with the four bar mechanism (and preferably the one with the least rpm). Four bar mechanism involves the use of four short rods connected as seen in the pic below.

Now that we have made the chassis you can now move on to making better and more accurate arms using servo motors and even simulate a human arm using Image Processing. We ,Team Robotix had our very own robotic arm built under the project “Michelangelo” where the arm drew graphs of functions .

Find out these methods and lots more on our website www.robotix.in.

R.A.F.T. : A Revisit

RAFT was one of the most successful mechanical events in Robotix 2011. Receiving participation from robot lovers all over the country, this event was one of a kind. Based on a practical and industry derived problem statement, the simplicity yet the unique nature of the event received wide acceptance.
First years from IIT Kharagpur also showed deep interest in the event right from the start when Team Robotix conducted the first fresher workshop, submitting hundreds of mechanical designs for the event.

Trivia : The name RAFT was decided and then its full form (Robot Aided Flood Transportation) was concieved.

In brief the event comprised of 3 rounds:
Round one: Navigation of a raft on water through a pre-defined course. The raft was provided to the participants making the entire more practical and flexible.
Round two: Retrieve people (in our case, cubes) from flood affected areas (platforms) and bring them to safety (Victory zone).
Round three was left to be disclosed at the time of the event.

To read more about the event click here

To view pictures from the event click here

NEGOTIATORS: Backwards, Codewise

Negotiators the online coding event of Robotix 2011 was an original design which went on to be largely successful. The participant’s program was pitched against that of two others in order to negotiate a pattern on a grid with blocks.

Trivia: Some participants submitted multiple entries hoping that their dummy entries would be pitted against their main ones. The attempt at hoodwinking was detected pretty early though and dummy entries ignored.

Maintaining a long standing record, students from IIT Kharagpur won all three prizes in this one.

To read more about the event click here

BALLISTA: Blast into the Past

Ballista, an autonomous events of Robotix 2011 was a night event with a rotating light source. This event enjoyed its status as belonging to the action-adventure genre in terms of robots. The problem statement was derived from robot snipers on enemy territory . Robots were to detect walls and shoot at the light source from behind the walls. Lightly put, a James Bond interpretation of Wall-E.
Trivia: Several rounds of prototyping were needed to develop the exact light source with a sliver of white light that would be intense enough for effective detection.
The winning team had a perfect run on the arena, thus proving that if you are like a determined Ballista you’ll hit your targets spot on.
Detecting a light source, a wall and then caliberating the angle and force to shoot is not an easy task. But they did it! Robotix challenges you to the limit indeed.

The event had 2 rounds:
Round 1: Detect a wall on a chartered path and shoot at the revolving light source.
Round 2: Detect any 3 walls and shoot.

To read more about the event click here
To view pictures from the event click here

ROBOCOP: Back to the Future

Robocop, the Image Processing (IP) category event of Robotix 2011 was the first of its kind ever. The robots were required to be the judges of colour coded character, marked to be good (green)and bad (red). It required intricate programming and some teams even used stereovision camera feeds.

Trivia: One of the participating ‘robocops’ was so accurate that it detected the red colour in the spectators’ clothing and shot marbles at them. Digital cameras with red lights had to be removed from the arena because the robot also took shots at them.

Mechanisms ranging from scissor extensions to air pressurised marble guns were used to topple the red cylinders (criminals).
Rounds: Traverse 4 rooms while toppling red cylinders in a time constraint.

To read more about the event click here
To view pictures from the event click here

FUGITIVES: Recapturing Moments

Fugitives the autonomous event that looked simple but turned out to be the hardest nut to crack. Robotix 2011 saw some mind-boggling alorithms to corner the fixed fugitives which emitted infrared radiations. The problem statements were based on artificial intelligence used for guarding prisoners.

     Trivia: Fugitives was initially titled ‘Chicken Run’ as the idea was to have an event, analogous to catching chickens autonomously, meaning the fugitives were to be moving instead of static. But that would be taking the difficulty level overboard so the solution to the initial problem statement is being thought of as a project by our technical team.

     Teams used different kinds of sensors and mechanisms to tackle this problem statement from Bluetooth to simply bump sensor circuitry.

  The event consisted of 3 rounds:

Round 1:   Corner 2 fugitives kept anywhere on a 10×10 grid

Round 2 & 3:  Corner multiple fugitives which are kept randomly on the grid.

To read more about the event click here
To view pictures from the event click here