Sie sind auf Seite 1von 27

MAE 4733 MECHATRONICS DESIGN TERM PROJECT-FALL 2011 REPORT

Design and Implementation of an Autonomous Photovore Robot with Obstacle Avoidance


SHUVRA BANIK BHARATH PAKALA 12/12/2011

SCHOOL OF MECHANICAL AND AEROSPACE ENGINEERING OKLAHOMA STATE UNIVERSITY

Abstract
A Photovore Robot is a robot which moves in the direction of light i.e. it will always move to a direction where the intensity of light is higher. Obstacle Avoidance in a robot means an intelligence which will guide the robot in such a way that it will always be able to avoid collisions with other objects. In this project, mechanical and electronic designs have been accomplished for an Autonomous Photovore Robot which also has the capability of Obstacle Avoidance.

Table Of Contents
Page Number Introduction Components Hardware IC Working Principles Differential Drive Photovore Algorithm Obstacle Avoidance Operation Circuit Diagram Project Pictures Applications Problems And Recommendations Problems Recommendations Future Works Conclusion References Appendix Project Code 3 4 4 6 8 8 9 11 12 12 14 16 17 17 17 18 19 21 22 22

Introduction
Photo means Light in Greek and Vore means Swallow Up in Latin. So the literal meaning of a Photovore Robot is a Light Eating Robot! In practical sense it is a robot which has affinity for light i.e. it is a light-seeking robot. It will chase light and can be made using simplest of all sensor algorithms. Using two light detecting sensors i.e. typical Photoresistors, this algorithm can be implemented. On the other hand, Obstacle Avoidance is a familiar term in the field of robotics which stands for the capability of a robot to avoid unnecessary collisions. The simplest algorithm for this can be done using IR Sensors though with the use of image processing or artificial intelligence, very high precision algorithm for Obstacle Avoidance can also be made. In this project the simplest approach has been made to detect obstacles using IR Emitter and IR Detector.

Components
Hardware
Breadboard A 3.25X2.125 400PNT breadboard has been used to place the electronic circuits.

Figure 1: Breadboard Wires Wires have been used to connect electronic devices on the breadboard.

Figure 2: Wires Robot Chassis A Pololu Robot Chassis RRC01A with cardboard has been used for the chassis.

Figure 3: Robot Chassis Wheels

Two HS-5745MG wheels and one Tamiya Ball Caster has been used.

Figure 4: Wheels Resistors 3 1K Resistors have been used.

Figure 5: Resistor Batteries One 9V and one 1.5V battery have been used.

Figure 6: Batteries Servo Motors 2 Parallax Continuous Rotation Servo Motors have been used.

Figure 7: Servo Motor 5

Development Kit One CCS PIC18F4520 Development Kit has been used.

Figure 8: Development Kit Fasteners Velcro fabric hook-and-loop fasteners have been used for attaching breadboard, development kit and servo motors with the robot chassis.

Figure 9: Fastener

IC
Photoresistors 2 Photoresistors have been used to sense light.

Figure 10: Photoresistors IR Sensors One IR Emitter and One IR Detector have been used to detect obstacles.

Figure 11: IR Sensors

Working Principles
Differential Drive
The movement of the robot has been done using Differential Drive algorithm. To move forward, both front wheels are moved in the same direction. To move to right, right wheel is stopped and left wheel is moved forward. To move to left, left wheel is stopped and right wheel is moved forward. To stop the robot, both front wheels are stopped. A schematic diagram for Differential Drive algorithm is given below (Drawn in Microsoft Word 2010).

Robot Rotates Clockwise

Left Wheel Moves Forward

Right Wheel Stopped

Figure 12: Differential Drive For the servo motors used in this project, the Base Positive Pulse Time is 1.5 ms with an interval of 20 ms (From motor datasheet).

Figure 13: Base Positive Pulse Time Decrease in Base Positive Pulse Time increases speed in clockwise direction. In this project a positive pulse of 1.33 ms has been given to move the motor in clockwise direction.

Figure 14: Movement in clockwise direction Increase in Base Positive Pulse Time increases speed in counterclockwise direction. In this project a positive pulse of 1.67 ms has been given to move the motor in counterclockwise direction.

Figure 15: Movement in counterclockwise direction Time interval between two positive pulses are always kept 20 ms for smooth operation.

Photovore Algorithm
Photovore Algorithm is implemented by using 2 photoresistors placed at the front side of the robot. A photoresistor is a Light Dependent Resistor (LDR). So its resistance changes with the change in intensity of the light falling upon it. In this project, a Voltage Divider Circuit has been made for each photoresistor and voltage across each one has been measured using ADC. It has been found that the voltage across each photoresistor decreases with the increase in the intensity of light. So the difference between voltage of left and right photoresistors are taken to sense the intensity of light. The following figure shows the Voltage Divider Circuit made across each photoresistor (Drawn in Microsoft Word 2010).

ADC
Increase in Light Decrease in Vout

Figure 16: Voltage Divider Circuit (Photoresistor)

From calibration, it has been found that if the voltage difference is more than 0.08V, then the robot should change the direction to right of left based on the intensity of light. So a C code has been written using this criterion. The following figure shows a schematic diagram of Photovore Algorithm (Drawn in Microsoft Word 2010).

Robot Rotates Clockwise

Photoresistor

More Light On Right

Figure 17: Photovore Algorithm 10

Obstacle Avoidance
It has been implemented using one IR Emitter and one IR Detector. The IR Emitter emits IR light continuously in the forward direction. If there is an obstacle in the front, then IR light bounces back. The IR Detector receives the reflected IR light. An IR Detector is actually an IR Phototransistor which converts IR light into voltage. In this project, it is found that the voltage across the IR Detector decreases with the increase in IR light upon it. So a Voltage Divider has been made across the IR Detector and voltage has been measured using ADC. The following figure shows the Voltage Divider Circuit made across the IR Detector (Drawn in Microsoft Word 2010).

Obstacle

IR Emitter

(1 K) IR Detector

ADC Increase in IR

Decrease in Vout

Figure 18: Voltage Divider Circuit (IR Detector)

For detecting a distance of 6 inches in room light, if the voltage is less than 4.94V, then robot has been turned to right. This voltage has been found after calibration. The following figure shows a schematic diagram of Obstacle Avoidance (Drawn in Microsoft Word 2010).

11

Obstacle
IR Light Reflects

Around 6 inches Turns to Right

IR Emitter

IR Detector

Figure 19: Obstacle Avoidance

Operation
When power is given to microcontroller, by default, it works as a Photovore Robot. If anyone presses button (PIN_A4) on the Development Kit, then an external interrupt occurs at PIN_B0 and the robot works as an Obstacle Avoiding Robot. Another interrupt will make the robot to work as a Photovore Robot and these operations continue alternatively with each button press until power is taken off.

12

Circuit Diagram
The circuit diagram of project is given below (Drawn in Proteus ISIS Professional 7.7 SP2).

Figure 20: Circuit Diagram

13

Project Pictures
Several pictures of the final setup are given below:

Figure 21: Sensors

Figure 22: Circuits

14

Figure 23: Final Look

15

Applications
Photovore Algorithm and Obstacle Avoidance can be utilized in many ways such as: A Photovore Robot can be used with a solar cell to get direct sunlight all the time of the day. Obstacle Avoidance can be used to avoid undesired collisions such as in Mars Rovers.

16

Problems And Recommendations


Problems
Problems faced during this project were: Same speed in both servo motors is not found using the same pulse rate. The wheels slide on the floor due to lack of enough friction. In order to keep the module compact, the chassis has been made smaller which in turn made the placement of circuit boards and batteries clumsy. Calibration of photoresistors is difficult because their resistance changes with the change in light intensity and source. Calibration of IR detector is difficult because voltage across it changes with respect to light source (i.e. in sunlight or in room light), color and also surface structure of obstacle.

Recommendations
Some recommendations for the above stated problems are: Using servos of same speed. Using wheels having better friction. Allocating enough space for the circuit boards and batteries. Using better photoresistors with better calibration. Using Sharp IR Range Finder sensors instead of IR Detectors.

17

Future Works
As an extension of this project these future works can be done: A Photophobe Robot can be made by changing the code of Photovore Robot which will move away from light or will seek darkness. A Fire Fighting Robot can be made using IR Detector of this robot. In fact it was almost done but enough time to write an efficient code was not available. A Line Following Robot can also be made using the Photoresistors of this robot.

18

Conclusion
Our project can be considered as a success. We have applied our knowledge attained from MAE 4733 course and could successfully run the project. A video demonstration of this project has been given during presentation and questions from the audience have been answered. We hope everyone has enjoyed our work.

19

References
We have got a lot of information regarding this project from the following site: 1) Society Of Robots (http://www.societyofrobots.com/) Figures for the Components section have been taken from these sites: 1) Figure 1: http://simonpstevens.com/Content/Projects/Microcontrollers/Posts/MicrocontrollerIntroEquipmen t_Files/Breadboard.jpg 2) Figure 2: http://itp.nyu.edu/~bms415/blog/wp-content/uploads/2011/09/wire-jumpers.jpg 3) Figure 3: http://doc-14-8g3dwarehouse.googleusercontent.com/3dwarehouse/secure/hhulr73hmmak89paul31eote4ben7ngk /20i0m4vfbhvn47ueuk69mb4m7ob8tbq8/1323604800000/lt/*/275b87677bfaf417c048bd5374389c 59?ts=1233007659000&ctyp=other 4) Figure 4: http://www.servocity.com/assets/images/316SH_on_servo.jpg http://www.pololu.com/picture/0J263.300.jpg 5) Figure 5 http://www.basicmicro.com/assets/images/1k_resistor.jpg 6) Figure 6 http://www.a1gifts.co.uk/images/prodimages/Duracell_9v_imageL.jpg http://alkalinebatteries.us/images/AA_Coppertop2.jpg 7) Figure 7 http://www.parallax.com/Portals/0/Images/Prod/9/900/900-00008-M.jpg 8) Figure 8 http://www.ccsinfo.com/product_info.php?products_id=18F452kit 20

9) Figure 9 http://1.bp.blogspot.com/_4CTz0Ff0Msc/TNyeUn-4E-I/AAAAAAAAAOE/KunJ2Vd15Y/s1600/Velcro.jpg 10) Figure 10 http://imgusr.tradekey.com/p-5717779-20111012020505/cds-photoresistor.jpg 11) Figure 11 http://www.t2retail.co.uk/lg_images/Dayga-T2-infrared_Emitter_And_Detector.jpg

21

Appendix
Project Code
//photovore + obstacle_avoidance #include <18f4520.h> #device ADC=10 #include <stdlib.h> #fuses HS,NOLVP,NOWDT,PUT #use delay(clock=20000000) //for LEDs #define GREEN_LED PIN_A5 #define RED_LED PIN_B5 //for Motors #define Motor_L PIN_C0 #define Motor_R PIN_C1 #define RES 5.0/1024.0 #define CHANNEL_1 1 #define CHANNEL_2 2 #define CHANNEL_3 3 // ADC resolution is 5 volts / 2^10 bits // Select Channel A1 // Select Channel A2 // Select Channel A3

int direction=1; float voltage; // Variable to hold float result #int_EXT // Interrupt sub routine void EXT_isr() { delay_us(10); direction^=1; } void adc_settings(void); void interrupt_settings(void); void photovore(void); void obstacle_avoidance(void); void move_forward(void); void turn_left(void); void turn_right(void); void turn_right_obs(void); void preform_adc(void);

22

void main(void) { interrupt_settings(); adc_settings(); while(TRUE) { photovore(); obstacle_avoidance(); } }

void preform_adc() { int16 AD_value; // Variable to hold 10 bit result delay_us(100); // Time given to complete AD conversion AD_value=read_adc(); // Gets 10 bit AD conversion result voltage=(float)AD_value*RES; }

void photovore(void) { float voltage_left,voltage_right,vore_diff; while(direction==1) { move_forward(); set_adc_channel(CHANNEL_1); // Select channel AN1 for AD conversion preform_adc(); voltage_left=voltage; set_adc_channel(CHANNEL_2); // Select channel AN1 for AD conversion preform_adc(); voltage_right=voltage; vore_diff=voltage_left-voltage_right; if (vore_diff>.08) { turn_right(); }

23

else { if(vore_diff<0.0) {vore_diff=abs(vore_diff); if (vore_diff>.08) { turn_left(); } } } } }

void move_forward() { output_high(Motor_L); delay_us(1670); output_low(Motor_L); delay_us(20000); output_high(Motor_R); delay_us(1330); output_low(Motor_R); delay_us(20000); }

void turn_left(void) { int i; output_low(RED_LED); output_low(Motor_L); for(i=0;i<20;++i) { output_high(Motor_R); delay_us(1330); output_low(Motor_R); delay_us(20000); } output_high(RED_LED); }

24

void turn_right(void) { int i; output_low(GREEN_LED); output_low(Motor_R); for(i=0;i<20;++i) { output_high(Motor_L); delay_us(1670); output_low(Motor_L); delay_us(20000); } output_high(GREEN_LED); }

void obstacle_avoidance(void) { while(direction==0) { move_forward(); set_adc_channel(CHANNEL_3); // Select channel AN1 for AD conversion preform_adc(); if (voltage<4.94) { turn_right_obs(); } } }

void turn_right_obs(void) {

25

int i; output_low(GREEN_LED); output_low(Motor_R); for(i=0;i<88;++i) { output_high(Motor_L); delay_us(1670); output_low(Motor_L); delay_us(20000); } output_high(GREEN_LED); }

void interrupt_settings(void) { set_tris_b(0x1); enable_interrupts(INT_EXT); enable_interrupts(GLOBAL); }

void adc_settings(void) { setup_adc_ports(AN0_TO_AN5); // Specify ADC channel (here only AN1 is needed) setup_adc(ADC_CLOCK_INTERNAL); // Use internal clock to initiate ADC }

26