2-DOF Pan-Tilt HAT For Raspberry Pi Wiki - DFRobot

FIT0731 Pan-Tilt HAT for Raspberry Pi and Jetson Nano

Introduction

This is a 2-DOF Pan-Tilt HAT based on Raspberry Pi 40PIN GPIO. It can directly be used once plugged into Pi board’s GPIO. Meanwhile, the board makes the Raspberry Pi GPIO pins led out for your convenient use. With I2C interface adopted, the module allows your Pi to control the camera movement and light intensity detection only using 2 signal lines. The on-board PCA9685 chip can output 12-bit resolution PWM to control the rotation of the pan-tilt. In addition, it comes with onboard TSL25911FN and built-in ADC that could make responses to light similar to human eyes, which is helpful for enhancing module performance; The on-board voltage level translators make it compatible with 3.3V/5V operating level. There is a specific acrylic board for fixing the Raspberry Pi camera on this product.

Through actual tests, it is proven that this module is compatible with all series types of Raspberry Pi and works well with Jetson Nano board. It can do 2-DOF motion in the horizontal and vertical directions, which is suitable for image monitoring/recognition and tracking when installed a camera. Besides, when combining an IR sensor or Ultrasonic distance sensor, it can be turned into a detection device to make a robot sense and avoid the surrounding obstacles. Of course, you can use it with other various sensors to complete your interactive works by Raspberry Pi or Jetson Nano.

Note:

  1. The Pan-Tilt HAT is unassembled, you need to get it done by yourself.
  2. Raspberry Pi board and Camera are not included.

Specification

Controller

The controller of the product is PCA9685, an I2C-bus controlled 12-bit 16 channel PWM output chip. The Pan-Tilt HAT integrates an onboard TSL2581 light sensor that can be used to detect light and work with camera, and it uses I2C interface as well, saving pin resources.

Communication Protocol

This product employs I2C communication. I2C-bus has one data line(SDA) and one clock line(SCL). There are three kinds of signals in the process of communicating: Start signal, Stop signal and Answer signal.

Communication Protocol

Write Data

When working, Raspberry Pi (hereafter named as Master) will first send a Start signal, then send a byte to TSL2581 sensor(hereafter named as Slaver), whose first 7bits are I2C address of Master and 1 bit write bit. Slave response with Answer signal every time it receives any data. Master send command register address to Slaver, then data of command register. Stop signals is sent to slave to stop communicating.

Read Data

When working, Master will first send a Start signal, then send a byte to Slaver, whose first 7bits are I2C address of Master and 1 bit write bit. Slave response with Answer signal every time it receives any data. Master send command register address to Slave. After that, Master will send a Start signal again, and then send a byte (7bits address and 1bit read bit) to Slaver. Slaver response and send data of the register to Master, master answer as well. Communication stops when Master sends Stop signal.

I2C Address

- PCA9685

PCA9685 and TSL2581 both use I2C communication. So we need to set different I2C addresses for them to ensure proper communication.

I2C Communication

- TSL2581

TSL2581

Note: The default I2C address pins of PCA9685 are set as A5=A4=A3=A2=A1=0, address is 0x40. I2C address pins of TSL2581 are set as Float and its I2C address is 0x39 by default. If you use the module with other development boards, for example, STM32, please add R/W bit to Low bit.

Tutorial Based on Raspberry Pi

Note: Before you assemble servos to Pan-Tilt HAT, please test the servo independently with test codes, otherwise, the servo may get stuck or even damaged since it is not at its original position. Refer to the instructions below:

HardWare Configuration

Enable I2C Interface:

Operate at the end:

1.  sudo raspi-config 
2.  Select Interfacing Options -> I2C ->yes Enable i2C Kernel Module

Enable I2C

Restart Raspberry Pi:

 1. sudo reboot

Install Library

1. wget http://www.airspayce.com/mikem/bcm2835/bcm2835-1.68.tar.gz
2. tar zxvf bcm2835-1.68.tar.gz 
3. cd bcm2835-1.68/
4. sudo ./configure && sudo make && sudo make check && sudo make install
5. # For more information, please refer to：http://www.airspayce.com/mikem/bcm2835/
1. sudo apt-get install wiringpi
2. # For Raspberry Pi system after may 2019, an upgrade is required.
3. wget https://project-downloads.drogon.net/wiringpi-latest.deb
4. sudo dpkg -i wiringpi-latest.deb
5. gpio -v
6. # run gpio -v version 2.52 will appear, if not, installtion error happens

Download Program

1. sudo apt-get install p7zip-full
2. wget http://www.waveshare.net/w/upload/9/96/Pan-Tilt_HAT_code.7z
3. 7z x Pan-Tilt_HAT_code.7z -r -o./Pan-Tilt_HAT_code
4. sudo chmod 777 -R  Pan-Tilt_HAT_code
5. cd Pan-Tilt_HAT_code/RaspberryPi/

Test

Connect:

Connection

Please Connect them correctly.

Brown GND
Red 5V
Orange S1/S0

Pay attention to the mount position of the tilt servo. Wrong place may cause servo damage. Do not assemble servos when testing. Make sure the servo can rotate 360 degrees freely and then connect hardware: pan servo to S1, tilt servo to S0. Open the test file in program folder.

1. cd test
2. make clean
3. make
4. sudo ./main

Assembly

Assembly

Assembled:

Assembled

A: pan servo
B: tilt servo

The module comes with a M2 screw package independently. It has 5 screws and 10 nuts. Screw the camera on with M2 screws, and then attach the Acrylic plate onto it, and fasten them together with nuts.

Assembly Camera

Demo

1. #Execute the following actions under the directory Pan-Tilt-HAT/RaspberryPi/:
2. #Execute this if using ambient light
3. cd Light_Sensor/
4. #Execute this if using pan servo
5. cd Servo_Driver
1. cd bcm2835
2. make clean
3. make
4. sudo ./main
1. cd wiringpi
2. make clean
3. make
4. sudo ./main
1. cd python
2. sudo python main.py

MJPG-STREAMER Software Monitoring in Real-time

MJPG-streamer can be used to collect image from a camera and transfer image to a browser by IP-based network in the form of stream.

1. sudo raspi-config
2. Select Enable Camera, select YES

11

For CSI interface camera, the system cannot find the device node of /dev/video0. It is required to add bcm2835-v4l2 in /etc/modules file.

1. sudo nano /etc/modules

Add:

1. bcm2835-v4l2

12

This module will be loaded when the system starts. Restart the system:

1. sudo reboot
2. ls /dev/video*

The device node of video0 can be found now:

13

Note: after the above steps finished, check whether the operation and command are correct.

1. sudo apt-get update
2. sudo apt-get install libjpeg8-dev cmake
3. cd Pan-Tilt-HAT/RaspberryPi/web_Python
4. git clone https://github.com/jacksonliam/mjpg-streamer
5. cd mjpg-streamer/mjpg-streamer-experimental/
6. sudo make clean all
7. sudo ./start.sh

Input the Raspberry Pi address into Google address bar(other browsers may not support this). Port COM 8080. Click stream, the content captured by the camera will be displayed in real-time. For more information, please refer to: https://github.com/jacksonliam/mjpg-streamer.

14

Remote Control via BOTTLE

Bottle is a simple and efficient micro Python web framework that follows WSGI, through which web control can be quickly realized. It only supports Python2.

Install Library:

1. sudo apt-get install python-bottle
2. cd Pan-Tilt-HAT/RaspberryPi/web_Python/
3. sudo python main.py

Input the Raspberry Pi address into Google address bar(other browsers may not support this). Port COM 8001. The following interface will appear, click the button to control remotely.

15

Support controlling by mobile phone and PC without using APP.

Controlled by mobile phone:

16

Contolled by PC:

17

Tutorial Based on Jetson Nano

Install Library

Install the function library

1. sudo apt-get update
2. sudo apt-get install python3-pip
3. sudo pip3 install Jetson.GPIO
4. sudo groupadd -f -r gpio
5. sudo usermod -a -G gpio your_user_name
6. sudo cp /opt/nvidia/jetson-gpio/etc/99-gpio.rules /etc/udev/rules.d/
7. sudo udevadm control --reload-rules && sudo udevadm trigger

Note: your_user_name is your user name, like Tom.

1. sudo apt-get install python-smbus
1. sudo apt-get install python3-pil
2. sudo apt-get install python3-numpy

Download Program

1. sudo apt-get install p7zip
2. wget http://www.waveshare.net/w/upload/9/96/Pan-Tilt_HAT_code.7z
3. 7zr x Pan-Tilt_HAT_code.7z -r -o./Pan-Tilt_HAT_code
4. sudo chmod 777 -R  Pan-Tilt_HAT_code
5. cd Pan-Tilt\ HAT_Code/Pan-Tilt\ HAT/

Also you can download from the Github:

1. sudo git clone https://github.com/waveshare/Pan-Tilt-HAT
2. cd Pan-Tilt-HAT/JetsonNano/

Basic Demo

1. # Execute this if using ambient light 
2. cd 1_Light_Sensor/
3. #Execute this if using pan servo
4. cd 2_Servo_Driver/
1. #python2
2. cd python2
3. sudo python main.py
4. #python3
5. cd python3
6. sudo python3 main.py

Test the Camera

1. DISPLAY=:0.0 gst-launch-1.0 nvarguscamerasrc ! \'video/x-raw(memory:NVMM), width=1920, height=1080, format=(string)NV12, framerate=(fraction)30/1\' ! nvoverlaysink -e

If everything works fine, the data captured by the camera will be displayed on the screen. Press Ctrl+C to stop.

Control the pan module.

Run at the end:

1. cd 3_Pan-Tilt+CAM
2. sudo python main.py

18

Data capatured by the camera is displayed on the screen in real-time.

Result

FAQ

For any questions, advice or cool ideas to share, please visit the DFRobot Forum.

More Documents