Arduino self balancing robot – Update #2 & #3

After much consideration, I think I will try to make a self balancing robot first before a ball balancing robot as to get a better understanding of “balancing” algorithms first. Hence, I’m working on a self balancing robot now instead. Actually I had finished a few stuff a few weeks before I wrote this post, so I will be writing Update #2 and Update #3 together.

#Update 2 – 17/10/15:

Don’t really have any pictures on it, majority of the time was spend testing parts only. I decided to build a self balancing robot first to get more understanding on how balancing robot work. This is because 2 wheel self balancing robot only have to handle 1 axis (the x-axis), whereas the ball balancing robot have to handle 3 axis, which is more complicated.

Micro-controller

• Using a Arduino Uno for testing purposes, intending to get a smaller micro controller soon
• Thought about using ATtiny85, but it is too small in memory size & slow in terms of speed.
  • Verdict: Stick to Arduino Uno for now.

Locomotion (Settled)

• Thought of using the 5V stepper motors for locomotion, but after testing them out, I realised it was way to slow to respond to changes……
  • Verdict: Nah, too slow. Will be choosing another type of motor to use.
• Tried out 5V mini rotational servo motors with an Arduino Uno, seems to be a better candidate for the robot
  • Verdict: Should be good (enough?). Most likely using 2 FITEC Continuous Servo Motor (FS90R)
  • Things to keep in mind while using rotational motors:
    • 0° – No movement
    • 90° – Clockwise movement (max. speed)
    • 180° – Anti-clockwise movement (max. speed)

Sensor

• Interfaced the MPU6050 (Accelerometer & Gyroscope Sensor) with the Arduino Uno, used it to control the speed & direction (aka velocity) of the servos
  • When the sensor falls in one direction, the servo has to react in the opposite direction. (~Newton’s 3rd law)
  • E.g. If the robot is falling to the left, the servo has to move to the right as to mitigate the fall

Coding

• Implemented a simple program to do crude correction of the falling sensor.
• Will be using only the gyroscope x-axis for now, with the values mapped.
• The velocity of the servo will be determined by adding 90 with how much change there is in the Gyro-sensor (X-axis). (Since at 90, the servo will not be moving.)

#include <Servo.h>
#include <Wire.h>

//MPU 
const int MPU=0x68;  // I2C address of the MPU-6050
int16_t AcX,AcY,AcZ,Tmp,GyX,GyY,GyZ;

//Servos
Servo myservoL;  //Left Servo (D10)
Servo myservoR;  //Right Servo (D9)

//Angles
int valL=90; 
int valR=90; 

void setup()
{
  myservoL.attach(4);  // Left
  myservoR.attach(5);  // Right
  
  //Initialise Serial at 115200 baud rate
  Serial.begin(115200);
  
  // Initialize MPU
  Wire.beginTransmission(MPU);
  Wire.write(0x6B);  // PWR_MGMT_1 register
  Wire.write(0);     // set to zero (wakes up the MPU-6050)
  Wire.endTransmission(true);
}

void loop() {

  //val = Serial.parseInt();            // reads the value of the potentiometer (value between 0 and 1023) 
  readVal();
  valL=90-map(GyX,-380,30000,0,180);
  valR=90+map(GyX,-380,30000,0,180);
  // sets the servo position according to sensor
  myservoL.write(valL);                  
   myservoR.write(valR);               

  Serial.print(GyX);
  Serial.print("t");
  Serial.print(GyY);
  Serial.print("t");
  Serial.print(GyZ);
  Serial.print("n");
  delay(15);                           // waits for the servo to get there 

}

void readVal(){
  Wire.beginTransmission(MPU);
  Wire.write(0x3B);  // starting with register 0x3B (ACCEL_XOUT_H)
  Wire.endTransmission(true);
  Wire.requestFrom(MPU,14,true);  // request a total of 14 registers
  AcX=Wire.read()<<8|Wire.read();  // 0x3B (ACCEL_XOUT_H) & 0x3C (ACCEL_XOUT_L)     
  AcY=Wire.read()<<8|Wire.read();  // 0x3D (ACCEL_YOUT_H) & 0x3E (ACCEL_YOUT_L)
  AcZ=Wire.read()<<8|Wire.read();  // 0x3F (ACCEL_ZOUT_H) & 0x40 (ACCEL_ZOUT_L)
  Tmp=Wire.read()<<8|Wire.read();  // 0x41 (TEMP_OUT_H) & 0x42 (TEMP_OUT_L)
  GyX=Wire.read()<<8|Wire.read();  // 0x43 (GYRO_XOUT_H) & 0x44 (GYRO_XOUT_L)
  GyY=Wire.read()<<8|Wire.read();  // 0x45 (GYRO_YOUT_H) & 0x46 (GYRO_YOUT_L)
  GyZ=Wire.read()<<8|Wire.read();  // 0x47 (GYRO_ZOUT_H) & 0x48 (GYRO_ZOUT_L) 
  
}

That’s it for Update #2. Now, moving on to Update #3…

#Update 3 – 24/10/15:

Micro-controller (Settled)

• I think I will be using the Bluno Bettle instead of Arduino Uno
• Features:
  • ATmega328@16MHz: Similar to the Arduino Uno
  • Bluetooth Low Energy (BT 4.0): Can be used in the future for control
  • Micro USB port
  • Super Compact Size (28.8mm X 33.1mm, 10g): That’s what I need!
  • Working voltage: 5V DC
  • Digital Pin x6
  • Analog Pin x6
  • PWM Output x2
  • UART interface x1
  • I2C interface x1
• Verdict: Most likely using this micro-controller.

Sensor (Settled)

• Interfaced the MPU6050 (Accelerometer & Gyroscope Sensor) with the Bluno Bettle
  • Encountered various problems while doing so
  • Got -1 for all the output (Gyroscope values), took some steps to resolve problem (in ~2h):
    • Changed jumper wires.
    • Used I2C Scanner to make sure the Bluno Bettle can detect the MPU6050.
    • Did not detect any address at first, so I swapped the wires around. And it detected the sensor. Opps!

Body

• Designed the body of the robot, it will consist of 3 body plates with stand-offs supporting it.
• I will 3D print the parts, the sketchup file will be uploaded onto Github soon.
• The servos will be attached below, sanwichiched between 2 of the plates.
• The top section of the plate will be where the sensor (MPU6050) & the 5V voltage regulator is. I will be using a 3.7V Lipoly Battery to power the Bluno Bettle.

(I think the distance between the top plate & middle plate is a little to small. Maybe I will extend the stand-off by one more the next time.)

Coding:

• Using the same code as previously mentioned. Will be adding the PID filter once I settled everything.

This are a few more other photos that I have taken: (Rather messy setup…)

And that’s about it. This is what I intend to do next time (hopefully):

• Organise the jumper wires nicely (it’s rather messy now).
• Find a good spot to place the Bettle & the battery.
• Modify the body a little bit, maybe 3D print wheels?
• Implement the PID algorithm into the robot.